e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e good afternoon we're now on the record it is uh 12 pm noon on Thursday March 13th 2025 this is a joint technical conference uh before the Colorado Public Utilities Commission and proceeding number 24 A- 0442 Public Service companies's just transition solicit ation and proceeding on number 24 A- 0547 the company's distribution system plan uh thanks uh to the company and everyone else uh for participating uh I'm Eric Bank the PC chair uh can my colleagues introduce themselves for the record starting with commissioner Gilman good afternoon commissioner Megan Gilman hey there commissioner Tom Plant uh thanks um we scheduled this joint technical conference to allow the commission and parties to have more transparency regarding certain critical forecast models uh and other issues uh associated with a company's filing uh in both pings our hope uh it is our hope that while we're focused on commissioner questions today interveners in both proceedings can use the discussion in this technical conference to inform and approve their ongoing advocacy in both these cases through Discovery testimony and at the uh evidentiary hearing through a set of uh decisions in both cases issued on March 7th 2025 we clarified the procedures for today's conference and adopted an initial agenda as a reminder of the parties the agenda comprises the following six categories first EVs and manag charging uh second beneficial electrification third large new LS such as data centers uh fourth the uh uh 8760 process for Peak load forecasts and load Seer fifth the long-term residential customer rate analysis and six a final miscellaneous category if we take breaks or the need for a confidential session arises we may need to move certain agenda items or uh specific confidential sessions till later in the day if we do conduct a confidential s session the company will need to help us confirm that the parties in attendance are proper but we appreciate that the company may have coordinated with a preliminary list to help the legal assistance on this note we ask that the company let us know and confirm with a legal assistance of there are individuals who might need screen sharing capabilities at the outset of uh uh each categories uh discussion uh since this is a technical con uh conference for presentation and clarifications from the company we do not need entries of uh appearance but for clarity of uh the record uh uh I would like to ask the uh company uh to provide a list of the people who uh will be presenting but before I do that why don't I turn it over to my colleagues and see if you guys have anything to say uh commissioner Gan um thank you yeah very much looking forward to the discussion today to learning more about the basis and impacts of the forecasting um and especially important given the the significant uptake sorry the long day already um the significant up that we're seeing uh in the forecast of values and really understanding um the bases of those so looking forward to it and appreciate um all those joining for for the preparation and uh in advance for your answers today uh thank you commissioner plant thanks uh and uh yeah also looking forward to the discussion today I think it'll be really beneficial for uh certainly all of us to get a little bit of a a a preview about uh how some of the modeling is uh works and what the differences are between some of the different terminology that's used but also really valuable for the other parties that are participating in the proceeding to um to hear the clarifications on how um you know certain situations are modeled and how they're hand L and I think that'll really help to inform our our Hearing in uh I guess may may is that one where we have our hearing may I don't think so I think it's later June June June yeah wild far mitigation different one may yeah so gonna be a tough uh it's gonna be an interesting summer um so with that uh I don't know who's uh leading the discussion for uh the company whether it's Mr iy or U Mr Lon Mr iy if you would uh uh first thanks for doing this uh and second if you would uh introduce uh uh the uh people from the company that'd be great happy to do so and first thank you for the opportunity uh also want to thank the commission for uh willingness to flex the dates uh as people were heading into pre-planned spring breaks and whatnot is this was better we certainly recognize that this provides you a bit fewer hours we to be uh efficient and we aim to be informative I'm happy to provide the list of of Representatives but we have a we have a a set of slides that we're planning to draw on as appropriate throughout the day and I don't know if we can share that slide with the names um calling my team if they're able to do that um and while they're doing that I wanted to mention and if they can I will um I wanted to mention we if if it's at at the commission's pleasure we were um suggesting that Mr Steve Marts who's vice president of integrated planning he had one slide to kind of frame Our Generation and distribution planning um we'd suggest that he could spend about five minutes on that up front but that's that's at your at your pleasure uh and then when we get the fourth topic the the 8760 realm at the commission's requests we do have a load Seer demo um spooled up for you we also were planning or hoping I should say to present about four slides again about five minutes um for Mr Dave Mino in advance of that to kind of help frame out our distribution system planning uh process so that's that sounds great so it sounds like uh hopefully we figured out the slides sharing or if we can make Mr iy uh the hosts uh I'll I'll share it if we need to um it's be great it's a bit of a lengthy list I I will say that and we've Endeavor to simplify it and Mr iy I I don't know if we have the same um I tried to make everybody I could find from public service a co-host but if I've missed anyone please let me know if i' if I've missed them and I'm happy to make anyone needed a co-host yeah okay one one thing Mr Riley apologies to step in on you um I think Mr Cowen Neil Cowen um would be helpful to get him sharing access I don't believe he has that right now in any event we can move forward be extremely helpful because he was he was planning to kind of bring slides up as the various Representatives called on them but to the oh sorry is that with a C or a k yes ma'am c o w n all done thank you thank you uh so ju just starting here and then I think I'll I'll leave it to Mr Cen after this um this is our list of individuals and do you need them read into the record is that what we're looking for here yeah you might just go down the list uh quickly sure uh just just as a brief introduction we've tried to lay this out across your your stated agenda um identify the leads by area because there there's in each area there's a lead in one area we have two leads and then there are two or three reporting folks as well so if you go across the columns that should hopefully give you a good guide throughout the day but um with that uh I am Jack iy j a c k i hle e do you need me to spell all of these I can do it okay n if you could just read through them okay next we will have Mr Steve Marts uh next Mr John good enough Mr John J N Landrum Mr Tom Bailey Mr David Mino M Brenda mcdermit Mr Zach Pollock M Deb Irwin Mr Chris Allen Mr Mark Shen Hider uh Mr Shawn White Mr Michael Bliss Mr Travis whan Mr Jeff knon Mr Andrew Wilson Mr Brian with a Y Davis and that's all well thank you uh thank you for doing that uh do you uh want to turn to uh Mr Marts uh for the uh planning overview yes I think Mr Howen is sharing Mr Mars's initial slide in a moment good afternoon Commissioners nice to see you and like Mr Ry said thank you for the opportunity just want to do a quick audio check am I coming through you are we can hear you loud and clear wonderful thank you chairman um thank you chairman blank and commissioner Gilman and plant so we we constructed this slide just to provide a little bit of an anchoring for the sa conversation we certainly hear the commission's uh questions loud and clear and you know certainly empathize with the role that you guys have to play in terms of helping um or needing to understand the cases there's obviously a lot of content in the cases there's a variety of extremely esoteric topics embedded within both and so I empathize with you know the need to bring these these technical Concepts together and kind of understand the interplay um between the two um so that's what the slide Endeavors to do at a high level like Mr iy said we prepared a series of witnesses um to steer the conversation where you all would like to go that the goal as I see it for today is to provide transparency how this is working again I know this commission's concern has been on this question of how are we integrating I'm here to provide high level um discussion on how that works and again to kind of rorate my role Excel Energy is I'm the vice president of integrated planning which means that I have oversight and direct oversight of our resource transmission Generation Um distribution and gas planing area so a lot of these things uh Concepts that we're putting forward in the just transition plan as well as the DSP other proceedings are directly coming from from my team um so to get to the topics at hand here with the with jts or jtp as well as the distribution system plan uh there's you know basically High Lev idea of a VIN diagram here and when I think about how you construct system plans there's kind of like three broad buckets um of work that that fit those buckets the first is on forecast inputs and assumptions right so that's kind of like what you put into the models um and there's a variety of categories and you know there's there's an ingredients list if you will to the recipe for creating system PL um and there's certainly commonality to that um however there are nuances in terms of what these system plans do and we'll talk in detail about that um but there's kind of the inputs and assumptions section the next section um the the the second third of of the grouping of work that goes into system plan is really around like what are the tools that we use to solve for the system plans these this architecture and our model outputs and I think it warrants recognition that not all of our models and our system uh system tool set that we use necessarily solve for the same things so I think this commission and our interveners are very aware of like you know Encompass is a is a branded name that we use for resource planning um that solves to a very different output than say the power flow models that we use to construct our distribution system plan so the way I kind of think about that this again this is just an illustra of an example to talk about you know there's this broad set of tools in the middle and we use them for different reasons Encompass solves to um to meet capacity needs for the system but also has embedded in it a algorithm which solves to economic production costing as well as an economic selection of resources based on various inputs assumptions what we provided in terms of load profiles weather profiles and things like that as a as a object oppose that to a power flow model which is like what we use in a distribution system plan that is looking at more technical analysis looking at Power flow across the system based on conductor capacity um capacity of components within our system and essentially what that is doing is it's kind of balancing in versus out meaning how much Supply are you feeding a grid what's the customer Demand on it and then it's looking for basically system violations and then what our Engineers would do is they would look look at that and look for for Solutions which I'll talk about more in a second so the point of that is to say you know there's not really an economic solutioning that occurs like a powerful model so again happy you provide more detail but I just want to offer that up as a highle kind of you know axiomatic way to think about you know the differences in the tool sets that we use and kind of that second bucket and then the third bucket is kind of the output as well as the the solutioning of it and that's where I think we've made a lot of progress in terms of evolving as an integrated planning organization um and a little bit what's on the slide here so you know I've been in this role for three years and I think there's been a great evolution in terms of how we're thinking across all of our Commodities come up the best Solutions and I've seen us really modify our planning processes to do more as part of our as part of our system plans to better embed transmission with generation planning I see as kind of an evolution if you go back several vintages in our resource plans the level of transmission analysis was very high level we had some some high level understanding of how generation impacted transmission needs and the the history as I see is we've we've slowly but surely increased the level of analysis where you look at what we did with the jts you know we we conducted an entirely new and holistic evaluation looking at multiple scenarios stuff that we've never done in the past multiple scenarios looking at different potential outcomes on the system looking for specific stresses just to answer the question what would have to be true on the transmission system to get to certain penetration rates of things like deers um or distributed batteries or certain types of load profiles and so I think that we've demonstrated um a pretty consistent Evolution there doing more with our proceedings and then on the distribution system planning side you know I think we've also evolved our solution set there to Encompass demand side management demand response a lot of these things are in our forecast as well and so we'll talk about that in more detail throughout the course of the afternoon but again these are things that we're offering into the solutions sets of our planning engineers and so that way when we're solving for things like system violations due to power flow you know we're no longer just looking at this with the tool set of do I make the conductor bigger or not we're looking at you know multiple um uh multiple solution sets and constructing a portfolio around how we can solve some of these system challenges and again we brought that forward to this Commission in the form of demand side management demand response proceedings we recently filed our VP ter we've done a lot with with npas we've put them out in energy filing out there um and and and more things like that so getting back to the visual here if you think about the just trans transition plan what I think about are the core elements of what that case is about is we've put forward a new uh planning Reserve margin and resource accreditation construct um we deployed what we feel is one of the best um consultants in Industry to help us valuate that objectively especially in light of a high renewable system we've done a lot to move ourselves to scenario based forecasting I think we've proven that both in the distribution system plan as well as the just transition plan like I said before there's a lot of commonality for those and I think we're actually fortunate we filed those those two plans in very close proximity to one another and so again we're able to um show you the detail in terms of how these things are linked and you know how EV an EV forecast in one proceeding is linked to the EV forecast and other so I think we can we can demonstrate that commonality um and I like I say before right we've done a lot to overhaul how we look proactively at our transmission plan um to really understand different stress cases and then again I'm kind of reating here but you know we're opening up the solution sets we intentionally designed and pursued uh we called it an ader but I think you know the colloquial term that's kind of come to fruition here is called vpps just to use a broad term we intentionally created you know very heavy VPP um based scenarios so we as the company could really understand you know where our customers going with their energy use where's policy going and how do we best align with those so I feel that the scenarios with for in jtp as well as the DSP are very align to the scenarios and there's a lot of commonality between the two so I'll pause there um and open myself up for questions but those were the opening remarks that I wanted to provide to you all to kind of help frame up the afternoon and to um you know give you a sense of where we're coming from how we thought about conducting these filings and things that we've done and I I I'll concede maybe it's not always transparent to you guys but things that we've done to challenge ourselves to bring more commonality to the filings so just want to say uh appreciate it and uh seems like we're uh making progress on the the tools we're using uh uh and how we better integrate the planning process maybe uh I'll go first uh on the EVs and manage charging uh and um I think I want to focus on the distribution system planning and um more specifically on the company's uh supplemental direct testimony variance motion and some of the um you know sort of background um planning issues around it we're not going to decide that today we haven't heard from the parties uh um I certainly haven't been f briefed uh but I I just do have some questions uh that it raises and um I'll just start with the 75% of rated line capacity constraint it sounds like um from that motion um that's something the company uh has been using for years and not something new um is that is that correct and do you know if other utilities use it and what utilities uh that that 75% can rra on rated line capacity so I don't know Mr Morts if that's you or somebody else yeah and just to clarify too I think you'll see a mixture of of Mr iy and myself you know our goal here again is to be transparent ril so we'll help kind of play Air trffic controller make sure we've got the right people answering the right questions so just would ask for a little bit of Grace to ensure that we're getting you guys right folks um absolutely and we appreciate you turning out the the whole team so thank you yeah so than thank you so with that I'll ask uh Mr David Mino to come on camera and David David's our um distribution system planning manager for the state of Colorado and Dave's an expert in this area so thanks Dave of course yeah good afternoon can you hear me okay we can great yeah I appreciate the question yes the 75% design criteria has been in place by the company for um for some time now this is not something that's new right now um I think what is new in this case is that the more proactive approach we are taking for our distribution investment to keep the feeders near at that threshold to increase our operational flexibility um we have some slides pulled up here to dive more into what the 75% design criteria really is um but first I think it's really important to note that you know the 75% plan load limit is not a planning Reserve margin we still allow assets to be loaded up to 100% of their continuous rating um this is really in place to help us trigger and identify when new capital projects are needed to ensure that we have sufficient lead time to design procure and construct uh a given project to make sure that it's it it's in place by the time that the system needs it to be um ultimately this is an engineering and integrated design parameter that is based on the ratings of our substations and distribution equipment and it's also how we plan our distribution system to handle emergencies and contingencies uh as you can see on this slide here this is an illustrative example of our standard distribution substation configuration um our Standard Bank sizes have a name plate rating of 50 MVA um those 50 MVA substation Transformers serve five distribution feeders um so you can see below how these different ratings aligned with the rating of the asset that 75% criteria allows us on on a 13.2 KV system which is the vast majority of our distribution system in here in Colorado uh that 75% gets us around 10.2 MVA per Feer which if you look at all five feeders aligned with the name plate rating um as you get um as you look at higher utilization rates and ultimately like I mentioned before we do allow these feeders to be loaded up to 100% but as we increase the loading on each of these distribution feeders we start to chew into different um different ratings for these substation Transformers that are really meant to be operated at at shorter durations before we start damaging the equipment so this this right here really just illustrates that as we plan system is 95% we're really trying to align our overall substation Transformer equipment how many feeders are served by those distribution feeders um which aligns with that 75% based on how we rate our distribution system um if we go to the next slide it helps us illustrate a little bit more about how we use this 75% criteria to uh plan for contingencies on the distribution system so on on a 13.2 KV system that 75% criteria equates to 450 amps since we plan our assistant to a 600 amp continuous rating and our our standard distribution design is really to have three distribution feeder ties with adjacent feeders to support in terms uh to support our distribution system in terms of uh outages whether that's plan maintenance or if there's equipment failure or public damage to our our system by utilizing that 75% criteria it allows us if there's a fault at the close to the the head of the circuit it allows us to transfer a third of that distribution feeder to each of those adjacent feeders um and at most those would be loaded to around 600 amps which is that 100% continuous rating so this allows us to ride out that fall in indefinitely it and to allow our system to be operating at or below the 100% continuous rating so we don't have to worry about damaging equipment reducing asset uh health or or and lifespan so these these two criteria as well pretty simplified here in this diagram or are foundational to where that 75% planning criteria comes into play um this this planning design is is integrated across all of our operating companies here at Excel so all eight states um and um yeah just I believe that was answering your question happy to kind of dig in any nuances that you may have associated with that all right I appreciate that and I I get that uh for purposes of determining uh the capital spending relaxing uh that constraint could be difficult but putting aside the 75% uh regarding our request to look at a lower sales growth and flatten load shape scenario is uh the company saying in your motion that you can't uh do that for technical reasons Beyond looking at seven specific uh Investments that I understand that motion correctly that's correct so the the challeng really lies in our load Seer platform which is a living model that we continually update throughout the year uh adding new capacity checks that are approved on the distribution system as well as updating our assumptions that are inherent to how we plan the distribution system uh the the way that Lo here currently works is we're not able to just roll back to the assumptions that were used for a given scenario that we developed um so unfortunately we're not able to rerun the entire load Seer forecast with those new assumptions baked into it so what are we were proposing in the variance was to evaluate a subset of projects that we feel are representative of the overall distribution system based on breakdown of customer types served by these assets um location of these assets um it allows us to take a smaller sample size and and manually recompute what the forecast would be for those given asset based on the changing assumptions of those load curves um Miss Federico can you pull up uh the graph uh I sent you and let me represent to you that this is just something our team pulled together uh from uh the peak demand day in 2035 based on the uh 8,760 Matrix uh matrices uh you sent us uh did you get that Miss oh there it is yeah can you can you see that Mr Mina again yes so I guess I'm just struggling why is it uh appropriate to assume that all these water heaters heat pumps and EVS charge at uh the peak demand period you know really really between 8:30 and 10:30 in the evening you know it just seems like over the next 10 years aren't we going to spend hundreds of millions of dollars and set up time use and other rates to to better manage and flatten this load shape so if you can help me understand why this is I guess as I understand it now the only basis for uh the capital spending uh that's possible to look at in this case in a reasonable time frame yeah so the the low curve that we're looking at for the the water heater and electric vehicles these are based on actual use especially the electric water here is actual usage of a representative few that we have installed on the distribution system today using the actual customer usage data um I will say that from the distribution plan perspective we are using the forecast through 2034 and the influence of these Technologies especially the electric water heaters really becomes more impactful later on outside of that 10year Horizon where we're actually using to plan the distribution system and and what that invest is based on so you know I do think that and that's something we're going to look at with those seven representative projects is is looking at a more flattened U electric water heater usage for residential water heaters to look at what the impact will be on the forecast for those dist distribution assets that support those those Technologies how that impacts the the risks associated with those assets over time and ultimately how that that drives a need for those individual projects that are are representing those those subset of of assets so I do think that and again our distribution process is an annual process that we continually make improvements upon and and do an annual Cadence to make sure that the information that we're we're using is reflective of what we truly anticipate the adoption and performance of these Technologies to be and ultimately how that drives the needs of the distribution system from an investment standpoint wait yeah go ahead CH in yeah just want I want to butress Dave's response I'm actually going to ask two additional folks to come on camera because I think it will get more into the heart of that question that you're seeking so Zack poock and John good enough you guys can come on please and um add on to Dave's response to the Chairman's question I think there's two pieces here which is one in depth on how we really see customer demand um materializing as a result of economic signals I think that's that in the spirit of what you're getting at chairman and then Mr poock can provide a little bit additional detail on the planning process as it pertains the tool aspect of it yeah appr good I I'll start talking about kind of the load shape that we see here um and then I might reach out to uh you know somebody from our DSM team to um provide a little bit more detail if if it's of interest um so we will talk quite a bit today about the commonalities and the assumptions that went into the jts and the DSP this is one of the differences there is a different water heater shape in the jts than what we have in the DSP uh and it's really um a matter of what was available when the forecast was developed um so this water heater shape that you see here is based on as Dave said an individual unit uh that was the shape that was used in the 2023 DSM filing um the jts shape is based on the 24 to 26 DSM file which does have a flatter shape I'm sorry um just this this is a commissioner plant I had a quick question when you say water heater shape are you talking about what's labeled here as be uh yes um good good clarification so the the spikes in the be are being driven by by water heating I I guess I'm just struggling well I'm just struggling with litigating a decision about spending substantial Capital with this as the sole or primary forecast which just intuitively it seems like we can collectively do better um uh and was uh unhappy to see that we're not going to get a revised Capital forecast looking at a a flatter load shape but so Ju Just can you help me address why this is uh intuitively likely to happen this way and why we can't do better yeah if I may chair um I think all the evidence we have today of the deployment usage on our system reflects the load shape that you see here certainly there's there's opportunities in the future to reduce and flatten that load but I want to reiterate Mr mino's point that we're not really seeing impacts until the mid 2030s as far as this peak potentially driving the need for additional projects I think the other caution I would provide here is that in order to understand the impact of that load on our distribution system Mr mino's team looks at every individual asset we have from a substation Transformer bank and feeder perspective and he compares the aggregate load shape from 25 different types of load shapes depending on the makeup of the feeder whether it's res IAL commercial Etc and we compare that overall aggregate load shape as well as existing load and known load that we gather through a capacity check process against the existing Headroom that we have in our feeders so it's a little bit cautionable to to infer that this peak here may actually be driving investment because the reality is that we see a pretty good amount of headro in our residential feeders in these hours I don't know if Mr Min would like to add anything there but just wanted to point out that it's it's difficult to draw that conclusion from solely just looking at the load shape you really need to compare it against the overall load shape of the feeder inclusive of the existing load and the head room on the feeder yeah I would definitely agree with that um this is just showing the additional load but as as Zach mentioned we we look at over 800 distribution feeders each with their own unique existing load profile as well as 275 distribution substation Transformers again each with their own unique load profile so most of these are summer peaking assets where we have margin for margin of capacity to absorb winter electrification demands such as this so I don't know that this is it's fair to say that this is driving the investment I think this is driving some of the growth but it's not necessarily triggering overloads based on on just this information here so I guess historically uh distrib spending has been you know roughly 500 million and now it seems to be quickly doubling to a billion can you help us understand what's driving that if not this load shape happy to chair I I'll take the first shot at that and defer to Mr Mino if he wants to add any additional details so I I think you see the the change in spending being a reflection of of two practices one is the recognition to move to more proactive planning philosophy in the company's first distribution system plan we used a much more incremental uh forecasting process part of that was where we were as a company with the tools that we had available to us so partly what you're seeing here is the incorporation of the new load forecaster tool what loads your forecasting tool rather that tool gives us the ability not just to understand existing loads and historical patterns of adoption of load but it allows us to marry socioeconomic data Geographic and geospatial data and actually develop a forecast at the individual premise level of forecasted adoption in the future of specific load modifying DRS that process which we've seen a lot of other utilities and commissions explore um inherently adds more load to the forecast than you'd see under a more incremental forecast that's based upon a reflection of historical Trends the second piece of that is the company's effort to move more towards that 75% planning threshold as Mr Miner said before that doesn't necessarily operate like a planning Reserve margin where every project above 75% gets loed into the capital budget there is a prioritization process where we look at the most critical projects that have the highest amount of risk that we solve for so it's not sort of a blanket but I would argue that the the combination of those two factors is driving a lot of the increase that you're seeing in terms of the company's capacity budget in the broader distribution budget and other than looking at only seven specific project is there any way this commission can get a handle on what happens if you know load is load growth is slower than what you're projecting or the load shape is different than what you're projecting I'm not sure I can fully prepare to answer that right now I would say that you know before the commissioner right now is really the the next five years of investment that this DSP covers and I think specifically the two years that are consideration for the grid monitor ization adjustment Clause rer so perhaps isolating the request to focus on the near term as we think about longer term Solutions around managing load could could be a place to start but I know that's not not a fome answer yeah I would also add in our distribution system plan the study that we did with cavala looked at four different growth scenarios one was a pretty low adoption rate of Technologies all the way up to a pretty substantial adoption and the short-term growth Outlook all four of those different scenarios were in alignment uh which is really what this this Five-Year Plan is is based on so I I think that I think that gives us a good view of of those wha ifs in terms of how how this could play out in the short term um I have a couple more questions on EV that move to a bit of a different subject but why don't I turn it over uh to you commissioner Gilman do do you have questions on this or um or you want me to just finish out my questions and turn it to you um yeah I mean if if your questions are on EVS I think go ahead I I was going to save my um building electrification questions until that section but um I'll have some questions if you want to go to yours okay uh just uh quickly how did the company choose the amount of EVS that have unmanaged manage and manage future charging in the analysis uh and do you consider how customer incentive program GRS might influence neb uh owner uh charging profile yeah so I can I can take that one on on the assumptions that were made um so to level set and of where we get the existing number of EVS in our territory is from vehicle registration data for the zip codes that the utility serves and right now we currently have it's more than 100,000 vehicles in our service territory and about 5,000 participants in the managed charges pilot program um so that's kind of our starting point we've got with the shapes that we're assuming for manage charging which is informed by that manage charging pilot we're starting at roughly 5% saturation right 5,000 out of the 100,000 vehicles uh and then we just generally make an assumption that that will increase over time so we increase that from a current you know 5% to 70% on managed charging by the end of the forecast period um at the same time we're Shifting the charging window um it's maybe we can get into that a little bit a a little bit later um but yeah we're just we're assuming that that the percentage of uh customers on man charging gradually increases through time up to 70% and I see Deb coming on camera m yep thank you um Deborah one here um so to the part of the question about um incentives and their ability to influence customer Behavior so Mr goodu was talking about our active managed charging program um the charging Perks program we also have um a passive manage charging program that has um a number of enrollments as well around I think 10,000 and both of those have um customer incentives to participate um so over time here we are certainly expecting to see a lot more customer enrollments um and those incentives do play a a big role right so there needs to be kind of value to the system for these programs and value to the customer for them to want to participate um and so uh you know I think over time we'll be evolving those programs um how they work as well as the incentives associated with them but you know focusing on the incentives being a costeffective alternative um to the infrastructure um build that would otherwise occur if we were not managing charging uh just two last questions did you consider the possibility that some homes with EV will choose not to install level two uh charger and instead uh plug the car into a standard wall outlet I I think that's embedded in our just our general load sheet okay um in the uh DSP completeness filing it seems that the company is assuming that the customer the future has one EV and one plugin uh hybrid EV but that each car has a running demand of 7 kilowatts uh this seems to suggest that uh the company assumes that customers will have two level two Chargers one for the E EV and one for the plugin hybrid uh is that an accurate characterization of what's going on and is that part of underlying forecast I think we're looking for Mr Pollock or Mr Shen heider on on that aspect I think you're you're quoting from the grid of the future analysis Mr chairman I believe yes yeah Mr shider I can speak to the portion of that from uh the figure you were referencing so and and I don't know if we did a great job of explaining this in the past but that uh figure was really a illustration of what a potential customer of the future that is highly electrified could look like it is not intended to represent the average customer or even the majority of customers um but it's trying to illustrate a customer that electrifies the vast majority of their end uses uh to your direct Point uh the 7 kilowatt values that you mentioned those are are a typical Peak or name plate draw for an electrical vehicle charger for a level two depends a little bit on the you know there there's some variability there um and yes the the concept is that you know for that customer potentially both EVS would be plugged in at the end of the day and be both be plugged into their own charger overnight um the peak load if you uh and I I believe we provided an analysis and explanation of this in the completeness filing of how that total load on the home was derived it did not have both of those Chargers uh running at the same time I believe there was a you know each each of the Chargers contributed something like one kW to the peak which is sort of an assumption around you know there there would be some active managing of those Chargers uh I think I would turn it over to is it Mr good enough for the second half of the question and and maybe chairman could you repeat the second half of the question or did I answer um we currently do not have winter or summer rate we just have a single rating for our distribution feeders which is based on the summer rating um down the road we anticipate evaluating whether or not Winer Winer ratings are appropriate especially for underground systems which the distribution system is coming more and more comprised of so at this time we have a single distribution rating for our system okay um now there are parts of your system that already peak in the winter correct there are some yes um and then the projection is I I think it kind of depends where you look um sometime in the 2030s if I look across all the proceedings as early as 2032 I think and as late as 2039 across all the proceedings we have in front of us when you expect to be winter peing and of course that might change like locational um so are we approaching the planning window where you need to actually be considering when these things are peaking I mean to what degree does that impact the allowable capacity on the assets yeah so we evaluate the loading on the distribution system on an 8760 basis for each asset specific asset so we're already evaluating when these assets are peaking for how long consecutively and throughout the entire year to understand and quantify the the risk that we're seeing on the distribution system um there's a lot of factors that go into those Su winter ratings especially when you look at elevation and and ambient temperature and different things like that so um you know it is something that you know I think we are considering evaluating from a more long-term perspective but at this point uh we have not looked into that okay so what you're doing right now is taking that 8760 look at what's actually happening and if say it's a part of the system in the mountains that's already winter peaking you're identifying that Peak whenever that occurs and if it's winter peaking I'm presuming that's in cold weather and winter uh but comparing that to a capacity uh limit that is based on summer rating for is it's based on the maybe summer rating isn't quite the appropriate way it's based on the ambient temperature the elevation the the construction and capabilities of each asset um the mountain system is rated is a 25 KB system which is inherently different than our our 15 KB class system that uh we primarily been speaking of here but um yeah we we do look at it in terms of how how these risks are materializing what that loading looks like throughout the entire year and and addressing them as as we to fit okay but when you identify the allowable capacity of equipment that's based on temperatures in the summer or in the winter I'm confused uh it I believe I have to double check I believe it is um based on summer I think it is based on the the local ambient which is typically a warmer weather as well as wind speeds elevation and different different inputs such as that yes um Mr marz thank you commissioner Gman so you know just I think you're super faint is that any better a little closer to the mic yes yeah just speak up no no problem um hopefully that's better please of course correct me if if uh if I'm in audible again so certainly appreciate where your line questions going and agree with it um you know I would just add on to to Mr mino's answer that you know I think we also need to look at this more broadly than just the conductor itself when you look when you think about the DSP um and I think it's a good question we've had a lot of internal debate about that I would just note that the reason you're seeing kind of a difference there is there's a distribution system Peak that's that early 2030 number there's a bulk system Peak and they're different they're not the same and so that's where you're seeing that later 2030s number so just to provide a little insight into that that differential that you're picking up on there and that's something you know we're we're challenging ourselves to evaluate more more wholly to fully understand with Clarity when does that a curve uh it's a very good question on on ambient ratings uh there's more to it though than just the conductor we need to look throughout the substation and basically as I kind of put it like we have to chase down the limiting factor because it's not always the conductor like you can be in a very good spot from a conductor ambient rating perspective but it might be that the breaker switch um in the yard is is the limiting factor which needs some sort of cooling period or you know another area that we're concerned about obviously is Transformers given their high cost they have a thermal stress cycle that we have to contemplate and the thermal stress cycle of these are not always analogous or homologous to one another right so there's a different design basis that we have to synthesize across hundreds of different components in electrical system and so think it's a field that's right for examination however I also think it's very nent and so we're filing Don Dynamic line ratings uh on our transmission system of course I actually personally just came from a series of executive level meetings with Epi just this last week and they're endeavoring to create a test bed to really understand a State of the State on Dynamic line ratings and so we've done a lot of of evaluation of it we have our own demonstrations of it and specifically we're engaging on EP or EXC we're engaging with ERI on these kinds of conversations to get a more holistic understanding of how we can start to apply these kinds of constructs but one thing I'm very certain of at this point in time is they're not yet ready for broad system application so we're trying to find our footing of where we think this is best applied and like you I have the same open question as to as we contemplate a winter peaking utility what does that look like from a design basis perspective and then the last thing I'll say is we're also looking for and identify other utilities that already have a winter uh Peak so we can start benchmarking with them to see what operational practices and Engineering practices might need to change an area of rif to go explore that to the southeast given the high heat pump penetration um so we're in conversations with lots of company companies out there for example tvas um distribution companies or or lpcs is what they call them um Souther and Duke and others so with that I'll pause just want to add a little color and thank you for the opportunity to do that yeah appreciate it um I have a question now more specifically to EVS which I gu the topic but I just wanted to follow up on that capacity question um so in the jts technical appendix um which is jwi to um you show uh forecasted growth Broken Out by major component um and so we'll see EVS listed on uh two tables there there's a figure 2.25 and a um hold on let me see I think there's a table give me one second there's tables that list them out um and then there's also figures that show kind of graphically what the components of the load look like um in 2025 E's look to make up 2% of sales and 1% of the peak impact in 2028 they're 5% of sales and 3% of peak impact and then in 2030 there's 7% of sales and 7% of peak impact and sorry those that's from tables uh 2.21 and 2.22 so I'm curious um why we're seeing and and I get you know that you've started at this 5% assumption on managed charging for EVS but it actually seems as though and and then you grow that um over time assuming more people get in but what I'm seeing in these numbers is that the peak impact is at least at PACE if not outpacing the growth of sales which I wouldn't necessarily expect if we're getting more and more people into manage charging over time so I'm curious if you can enlighten me as to why we're seeing a quicker growth in Peak impact to bvs than we're seeing in energy sales impact to bvs yeah and if if uh Mr Cowen I don't know if we're able to it might help to pull up the graphic in the table um just for references um kind of speaking through that yeah it's in jwi 2 so this is from the jts um jwi 2 cables 2.21 and 2.22 I don't know if someone on your side has that or if we should try and get it pulled up commissioner Gilman we're working on pulling that up right now okay maybe as they pull it up I I'll ask you another question just to make good use of our time um does the company anticipate that manage charging will be primarily used to avoid um bulk system peaking conditions um or do you anticipate using it more regularly to avoid local distribution system uh overloading condition yeah the the way it's modeled in jts um is more more on the bulk side 're we're certainly not looking at individual kind of distribution constraints or or limitations as we're deriving the peak load shape or the EV load shape that's used in the jts um it's really it's really intended to be a proxy of the need to move to more manage charging over time as opposed to kind of prescriptive um what a new manage charging shape will look like um I know Mr C's working on pulling up the um the technical appendix but maybe after we go through that table there are a couple slides that I've prepared as well that might help guide the discussion around our load shapes um for EVS that we're assuming in the forecast yeah and if I could jump in on your question about um sort of the future of managed charging um so as Mr Good Enough noted we today's program is managing to bulk system right to bulk system signals um to maximizing the use of bulk system renewable energy resources um along with you know R uh which is generally based on time of use rates as well as um first and foremost the customers need for their vehicle when they need their vehicle charged um we are looking over time to start to focus more as well on managing to distribution system needs and so that's that's not what the program does today but that's certainly our intent um and we are starting to um you know look into that and I think you'll see that from us in the fure future um but you know think from a planning perspective you know that's why the load shapes are not specifically representative of just today's managed charging program right they're intended to to show what the impacts could be in the future as we evolve those programs Mr Pa thank you commissioner yeah if I can just add a few points to Mr and Mr G's points there um despite the fact that we're not targeting the distribution system as an objective today there is benefit from spreading out that EV load in the way that we've both modeled and implemented it in that overnight period so we are seeing benefit from that the other thing I would add is just one of the nuances that we're working through today is that if you think about the bulk system we have a dedicated commercial operations desk that is dispatching power plants Alternatives demand s side management Etc to be able to meet that single Peak that type of function doesn't really exist in most utilities today having some who's sitting around the clock dispatching manage charging or any other Dr for that matter to solve those very local and very nuanced constraints so we're looking at ways we can augment that capability through various Technologies you'll see in our DSP plan one of the things that we're moving forward on this year um is deploying what we call grid derms one of the use cases for that is specifically around finding better ways to integrate EV managed charging load as you're well aware a lot of that load can be pretty spiky especially as you go to the higher voltage charging like DC fast charging so all these things are very much in the works but it's still pretty much early days for that and I think that's pretty typical of the industry not just Excel okay so as I'm looking at these forecast I guess is the broad correct understanding that the jts uh forecasting for the EV component includes the manage charging starting at 5% and increasing uh out to 70% in I think 2015 um but that the DSP does not include any managed EV charging that's not quite accurate so I'll answer at a high level and then I'll allow the the appropriate to chime in here um we did apply manage charging and I think that's reflected in the technical appendex for the DSP which is DCM 9 the way we applied the shape was a little bit different just given some of the nuances with how load seere Works um in addition to that the scenario modeling that we conducted with cavala where we modeled four additional scenarios we also ran against our goal adoption or goal align scenario which had the highest adoption level of DS a 50% managed charging and 100% manag charging assumption which unlike how we model uh and operate the program today we assumed we would have the ability to effectively flatten the load at the customer premise which would inherently provide a much bigger distribution benefit um we have little little evidence today based upon our work with other utilities and vendors to suggest that 50% or 100% is is technically achievable but nevertheless we thought it would be interesting to see the impact from an infrastructure perspective of what that could look like so I'll probably turn it over to Mr good enough as far as the uh the impact and application of the charging shape or Mr Mino if he wants to talk about the specifics of Bloods here yeah just real quick in this 50 or 100% assumption on being able to manage to the distribution system like at the customer premise um is that included somewhere in the record of the DSP yes that would be a believe in the broad DSP filing which is inclusive the of the separate caval report I think we also allude to that in the plan believe that's attachment zdp 1 in the DSP record do we see the technical outcomes of such a run or it's just mentioned I guess uh I believe you should be able to see the impact on on load there um obviously one of the challenges is when you're doing it for 800 feeders and 275 substation Banks putting that into a written form is challenging but I think in aggregate those results should be reflected in the cavala report both from a an infrastructure needs perspective and a cost perspective and off the top of my head I believe the 100% managed charging scenario showed about A5 billion dollar infrastructure distribution infrastructure savings uh from now to 2025 now through I thought you said 2025 excuse me yeah 2025 through 2050 that's what I meant to say yeah okay uh yeah do you go want to take us through the chart to look at the difference in Peak growth and energy sales for Mees uh yeah if we can scroll down to the table here for others reference but I've kind of pulled it up to the side just while some of that other discussion was happening to familiarize myself with the numbers um frankly I'll probably need to dig a little bit as to why the that pace is showing up the way it is I see why you're asking the question it looks like the the share of peak is increasing faster than the share of of energy um I think this could be a good opportunity to to discuss the load shapes that we're using and how we're evolving those over time um I suspect a little bit of the interplay there is what's driving that I could confirm that you know later um but if if we think it's a good time to share that information I think it would be a good time to to talk through that okay yeah um so this would be slide for Mr Cowan this would be slide it'll be seven and seven eight and nine but we'll start with slide seven in our prepared deck so what I'm uh showing here is an illustration of the load shapes that we're using um at least kind of the the the base definition of those load shapes so for light duty Vehicles we have managed residential which is blue on the left chart unmanaged residential which which is orange and then we do assume some public charging for light duty Vehicles the way these shapes are constructed it's percent of daily load so it's normalized to the Daily load so you can really kind of pull out the shape um so it's not like a scale of of charging just the the percentage of charging for that type and then medium and heavy duty on the on the right side so for the for the manage shape as I mentioned uh it's um from our enrolled from the customers enrolled in our manage charging program we actually have metered data for the charging for those customers the public charging is based on uh information from I believe guide housee as well as the medium the uh L Duty manage charging profile and the medium and heavy duty profiles are based on uh data from external vendors please go to the next slide so so one of the assumptions that we make as we develop the forecast is what we call Ev tranching um if we leave that managed shape the way it is and we slowly roll customers to that manage shape over time eventually that's the peak hour like when we're when we're incenting customers to charge their charge their vehicles with that very you know sharp increase during you know during the time that we're incenting them to do that uh each each in each EV that somebody buys sets a new Peak um unless we start to move that that window over time which is what we've incorporated into the forecast here and that's what I show on the top left um just taking that same shape but recognizing that we'll need to have some sort of mechanism to move this load over time uh we do we have a total of five Tres for the residential uh light duty vehicles and then we have a total of two for each of the mdv and the the uh hdv and then the result is is really what I wanted to kind of lead up to which is on the next slide which is the flat of that EV load curve over time um so you can see in 20 2025 uh that residential EV low curve looks you know charging kind of in the evenings and then uh in the late evenings and then through that truning and through that move to manage charging we're flattening it over time so I think the answer to your question commissioner Gilman but I'll I'll confirm this is that it's what you're seeing kind of in that period that you referenced is that the we're still a lot of the customers are still on that kind of solid blue shape where the the new EVS that are being adopted are you know closer to a a a shape like that solid Blue Line than the flatter shape that we're assuming as we go through time yeah I guess what what confuses me is that the peak increases faster than the sales I would expect if we're moving more people to manage charging that the inverse would be true that we would be able to outpace the peak expansion by the expansion in sales but that's not really what the numbers are showing yeah I think there's also a component of the shifting hour as we get beneficial electrification as well and and other factors that are driving load differently so it's might not be as simple as um you know we're moving customers to not charge at hour ending 18 if that's no longer the peak hour uh you know that could those those Dynamics could be at play as well okay and these um these charts you're showing here um are these already in the record or are these new these these are new um they're yeah these are new that I uh created for this okay um and from what I'm understand the treatment in the DSP of EVS is different than what we're seeing here as far as how they were treated in the jts but accurate uh slightly different I think the The Source data is same and there's a similar kind of you know um methodology or or assumptions baked into the DSP forecasting there's some Nuance in loads here that doesn't allow the modeling to do crunching in a similar way that we did in in in jts um which Dave can speak to more um if if it's of interest um but they do the DSP is incorporating kind of this flatter load shape is their future managed charging shape which is based on this information um it's just not exactly Incorporated the same way okay um and and in response to the chair's question about if if some subset of customers do not get a level two charger and just charge a say 8 to 12 amps um normal charge um you mentioned that that's integrated into the company's existing forecast or like base forecast and you're wondering how like what are the assumptions about what percent of customers do that or when they charge or if they can be moved off peak yeah I think those assumptions are just embedded in the load shapes that that we're using that again were produced by by a third party um I don't think you would see individual EVS peaking at 7kw for example in our certainly not on average in the jts um so it's just some of that diversity is already built into these existing load shap okay um do you think there's detail about those assumptions in the reports um potentially uh that this is a forecast that's developed by a different team in Excel so I'm not sure exactly what information is in there but we could follow up on that okay and are those reports you're talking about the the cavala forecast or something else uh it was a a study um done in 2019 I believe with uh another vendor guide housee okay um and then your um kind of assumptions around I guess the pace and depth of EV adoption on the distribution system in terms of like geographically where are we seeing more quicker um is based upon what yeah and I think that's a question for for Mr Mino um he's kind of taking our system level jts EV adoptions and and then the load Zer tool is what's doing the disaggregation or the distribution system so he's best suited to answer that question okay yeah thanks John um so the way that our low seere tool works is like as John mentioned we take that high level operating level forecast and we use the functionality within low seere called spatial allocation which takes into account a multitude of different inputs to help determine how we allocate that down to the individual premise level on the distribution system so that's certain things like socioeconomic data intensity for adoption um demographic data weather data uh Transportation data there's a whole slew of services that load Seer integrates into its platform to help influence where that adoption will take place by premise by year essentially in our forecast so that's it's not just a peanut butter spread across the distribution system it's just a little more sophisticated than that okay and I suppose that works for residential on commercial or Fleet is there any kind of more detailed information that's being used to identify locational distribution uh yes there so I mean I think all these factors go into um into that allocation process I'm not sure which ones are utilized for each type but there's commercial industrial data there's um there's Appliance data there's um transportation so travel patterns for for um traffic and things like that so there there's a quite a bit of behind the scenes Integrations for different different data inputs that help inform that allocation process but that allocation process is what we use to allocate all for every single all the 25 technology types that are laid out in my in the uh technical appendix which I think is attachment nine in the DSP um all 25 of those technology types use that spatial allocation process and are informed by those inputs okay thanks uh those are questions right now chair uh thank you commissioner Gman commissioner plant thanks um I was oh okay I'll come back to that one um uh couple of quick questions sort of following on um uh following on my colleagues questions but first um I I think I heard Mr Mars you say that there is not economic modeling that is incorporated into a powerf flow model is that is that correct hear that yeah thanks commissioner plant that's correct um you know right now the tool set around distribution level modeling is very focused on power modeling um so it's it's that kind of in equals out solutioning and looking for violation so it doesn't work like incompass in the sense of you you feed it you know options a through D which are based on like you know economic Supply curves and economically optimizes we have to provide that manual programming in terms of how we think about the solution sets as well as what we feed into it in the form of our load forecasts so if we were looking at for example you know some some powerflow modeling that was driving uh costs on the on the transmission system that we might be able to avoid through an investment in the distribution system that avoided cost would not be reflected in that power flow analysis is that correct cor do you mind rephrasing commissioner I'm not sure I'm quite following how you're linking together the transmission and distribution modeling right so so we if we're looking at powerf flow modeling that's driving some needs for capacity investment on the on the transmission system just based on the way that the power flow is going to go across the system and some of that capacity we can avoid through an investment in uh storage generation management whatever it might be on the distribution system that cost savings or that um that economic comparison would not be reflected um in your power flow modeling that you're doing with the Encompass model is that correct yes think we got maybe take a little bit of a step back I think you might be conflating how the production cost modeling Works within incompass which is in the jts versus the powerful modeling like the DSP so like example like how that works and ready see that direct result this is exactly why we constructed the scenario is you know looking at what the avoided like bulk system impacts are from like what we call the ader or otherwise known as the VPP scenario which adds in 900 megawatts directly and to your point it's agnostic location which say what happens how does it economic how does it economically optimize um based on this 900 megawatt infusion all you know all within the bulk system and how does that then change bulk system Investments so that's kind of like step one of that we see this though in the various where what guess where it's common is and that's the point of the G of the future work as well as the multiple scenarios we laid out in the in the DSP is we also look at that in the form of load shapes that feed into the DSP and how we modeled that so just by nature of the fact the powerful modeling doesn't solve for it doesn't mean that it's not in there it shows up in multiple places and that's why we've intentionally looked for different places to see how battery optimization and large large scaling of vpps and distributed dispatchable generation can can help the system and give us insights in terms of how we think about the impact of Investments and and those those uh that management on the distribution system system would then be reflected in the power requirements the capacity requirements on the bulk system is that correct or correct okay um thank you um so um I'll jump to this since we got this up on the on the screen right now um I know that we're you're going from an assumption of a 5% to a 70% over the planning period um of managed charging and I'm assuming that's what's reflected here um do you know is it uh is it uh do you have different assumptions for new EV load that is um you know current is 5% um you know and then by 2030 it's 100% And that averages out to your or uh how is that how is that reflected or is it just are you just looking at your overall 100% of EV owners yeah it's it's the percentages are based on overall adoptions so you'll have you know a certain percentage of of new vehicles coming on um on manage charging but also customers that aren't currently on manage charging switching over so it really is a a percentage of that that grand total um I will also mention I will just mention that this flattening of the low curve is is both the increasing percentage of of customers on man charging but also that truning that we're doing where over time it's a different load shape that we're incenting customers too and it and it looks as though this also includes work workplace charging uh incentivizing basically charging in the middle of the day at the work workplace or is that maybe that's fleets yeah I mean there could be a number of ways to get here this is specifically residential light duty vehicles um you know really this is just in intended to demonstrate the need to flatten the load not necessarily the way by which we will do it um but just just a recognition that holding the shape static you know just create Peak demands that are driven entirely by EV adoption and that's not something that that we can manage to or that we should be planning to okay and and I'm I'm just assuming that that time period between nine am and and 5 in the afternoon that we're seeing an increase over the base that would be probably workplace charging yeah it would be uh you know a way to do that would be give residential customers incentive to charge at their workplace instead and that would help kind of move that load to the middle of the day okay and if I'm uh am I correct in the in understanding that when you're calculating the 75% line capacity limits those are calculated without any of the management that you know you're working on developing in terms of um EV charging or ader whatever it might be um that that limit is sort of subtracting those out and then applying it and seeing where you reach at that point and then you would apply those things after is that correct uh I will let Mr Mino take that one the S yeah this thank you uh the 75% plan load limit is really based on what's the net loading on that given asset that we've seen historically so you know we're looking at the actual skated data that is being that is loading that's being supplied by Any Given distribution feeder and then as we forecast out we're accounting for those low shapes in the forecast and comparing that against that 75% um so does that answer your question or did um if not could you please repeat it yeah it sounds like so what I'm um what I'm assuming is that when you're calculating that 75% you're not assuming any um demand management whether it's from EES or uh other aggregated loads or uh it's just a straight up if everybody just did things when they were going to do things and there was no management at all that then this is when we would hit 75% is that correct or is are these assumptions of increased um management uh of EVs and and all that sort and you know the imp imp implementation of the VPP program all that stuff is already Incorporated for when you calculate your 75% so in our forecast we have a certain degree of measure managed technologies that are incorporated into the the growth forecast for each individual asset and thus being compared against uh the 75% rating for each of those distribution feed are supporting that assets like um vpps the ddg program uh we're using our grid needs assessment to inform where we can uh where we want to Target deploying those assets so we can they can support the distribution system but those are not inherently baked into the the forecast as it stands now okay great if I can maybe try to clarify here yeah so for the the EV charging assumption that's in load seere that informs the capacity projects the forecasted load um which is compared against the rating of the equipment is reflected of a manage charging EV load shape so that is assumed in there there's a little bit of nuance there in that we're not specifically in that forecast of that managed charging shape assuming it is specifically targeted to the very unique load shape of a substation Transformer feeder nevertheless the general load shape that we're seeing in modeling and VB charging right now spreads that load out across the nighttime hours in a way that does provide some benefit to the distribution system even though that's not what we're specifically seeking as an objective as we talked about the bulk systems really the primary beneficiary today yeah and so if I understand that what you're assuming is that increase from five to 70% manage charging over time I I believe those um that assumption is consistent across um I don't know if we want to go to the technical appendix and the DSP at some point but that has the load shapes and the assumptions for the manage charges specifically in the distribution context okay great um if if you could pull up um the chart that chair blank uh introduced I just had one question or a couple of questions around that Miss federo did you have that chart that chair brought uper that you could pull up again yes sorry I closed it when I was pulling up something else my fault give me a second thank you great thanks a lot so just a couple questions around this um Mr Good Enough one of the one of the reasons I was asking about how that 5 to 70% is actually calculated is that this uh graph is representing new residential demand so I didn't know are you are you assuming here and what we're seeing here and this is obviously 2035 so we're talking about you know 10 years out in the future so is what we're seeing here in this this uh hump uh between 6:00 and and uh 10 oock at night that's assuming a 70% um managed charging or are you assuming a higher level for new uh EVS um I'll take a stab at that and then Mr Mino can can either add on or clarify I believe this is all incremental load through 2035 not just um EVS that are adopted in 2035 so so this would be a combination of you know the that the entirety of those shifting load shapes between now and 2035 the load that is accumulated for the the vehicles on the various load shapes not specifically um one years of adoption when okay this is new residential demands looking from 2025 to 2035 is that what you're saying that's yes okay and so this what we're seeing here with this uh this increase between 6:00 and and 10:00 that's that's with 70% management um we we aren't at 70% by 2035 um we we get to that point in in 2050 in the in our assumptions so it would be on our way to 70% it's roughly linear so okay okay great thank you I understand better now um and Mr Pollock if I could ask you a quick question as I'm trying to understand some of this on the for example this peak that we see here um you know overall that's getting almost up to 4,000 megawatts when we're looking at this uh peak in demand on our distribution system and we're looking at um you know what the Char what the cost would be to increase the capacity to manage that those kinds of Peaks on our on our distribution system if we're looking out at you know 2035 hitting that Peak and I think as you pointed out we're not there yet this is you know looking out into the future that is that what's establishing those uh distribution costs if we were to just invest in uh capacity increases on the distribution system it would cost this much and so you're taking those distribution costs and you're putting those into the aggregator payment and so if we were to um as you said you know we what we hope to do with the um with the VPP program and the ADR program is to flatten these curves um out what would happen there is basically a savings for the rate payers that would be the number between what this capacity investment would have cost and what was paid out minus what was paid out to aggregators in order to achieve those uh those lower load levels is that a correct way to think about this appreciate the the question commissioner plan I'll I'll try to unpack that because I think there's a couple of subp Parts there so I think you're referring back to our VPP tariff that we filed just a few months ago is that right correct yes so how we we designed that is we did for the avoided distribution component take a look at the capacity Investments That would be needed over the course of the DSP Mr Miner I said before that's about three gws of additional capacity that would be unlocked and then we just took the cost of that and divided that out to come up with a cost of capacity and looked at it from effectively an annual revenue requirement of that capacity which came out to be about $70 per kilowatt month um the way we've designed the program right now which is reflective of a couple things one where we from a technology perspective but two I think also from what we see with the maturity of aggregators in the market we're designing the program to primarily be dispatched around bulk system needs to the extent that the load shape of the a distribution feeder may be coincident with that bulk system we've proposed providing that avoided distribution value which would be to question a reflection of potential cost reduction from reducing load um whether or not an aggregator may choose to enroll customer cust on a feeder to the extent that it actually reduces or defers a project I think is yet to be seen but that is certainly the intent um and then as far as I think the last part of your question as to whether there' be a cost savings there I think in Mr iy's testimony we did propose setting the avoided cost right now as what would be um equal to the value of avoided generation transmission distribution however we did introduce the potential idea that if aggregators were able to bring us that capacity more cost-effectively that absolutely could be set through some sort of competitive process that would result in a cost savings for our customers yeah I guess that's that's one of the sort of the core question is the you you set the aggregator payment based on what the savings would be or what the capacity would reduction would be and the price for capacity if you were to invest in the capital upgrades but is that do you assume that you make the investment in the capital upgrades and then you apply the the that value to the aggregator payment or are you doing the aggregator payment first in hopes of avoiding the capital upgrades I I think it's a combination of both and it really depends where in the the budget and the forecast of those projects uh that would happen I think that the challenge we're trying to reconcile right now we want to cultivate a an active market for aggregators but so far our experience with non-ar Alternatives has been somewhat disappointing so we obviously we can't pause projects because we need to be able to turn our customers but we do want to be able to leverage these alternative Solutions as part of our core solution set so I I think it's a little bit of both right now to be totally Frank okay and so that would adjust over time would that payment adjust over time then I think correct I think the payment could adjust over time I think the amount of of targeting May adjust over time again the challenge we have is that we if if we're opening this program to third party aggregators we're not sure where an aggregator may choose to conduct their sales and marketing where they may be able to enroll customers so it's at a level of specificity that's not as specific as a non-wire alternative where we might be asking a third party to propose a very geographically targeted solution so I think ultimately we just need to see how how the program goes and then we can potentially refine the pay and the you know whether it's being effective in reducing some of that loading that triggers projects sort of associated with that you know if you have I know that there are some feeders for example that are included in the distribution payment um proposed and there some feeders that aren't you may have an aggregator that Aggregates uh people who some of them are on that feeder and some of them aren't on that feeder how do you calculate what your when they are delivering x amount of capacity through this aggregated load management how do you determine how much of that is coming from the customers that are on a a targeted feeder and how much is coming from a customer that's not on a targeted feeder yeah so I think we've talked about that a little bit our proposal um depending on the resource type we're going to use a combination of telemetry data from devices in the case of things like batteries uh and then I believe we're proposing to use Ami data to to be able to effectively settle which resources are providing that information I will say a lot of that information is expected to come directly from the aggregators so that's one of the things I think we'll probably work out through the the process of that proceeding um whether that's workable for the company whether it's workable for aggregators but it is fully intended that we'd be able to um understand which capacity is coming from where and that customers that are part of that aggregation would be compensated appropriately if they are providing the distribution value on those years great um and to follow up on commissioner Gilman's question Mr Pollock you were you were talking about um I think the question was around bulk system versus uh distribution system Peaks and the management and I I know that your sort of um uh what we call phase one what you call aggregator derms is focused on the bulk system and then the idea is that what you're calling grid derms would then start to focus on uh the distribution system Peaks and communication with adms and SK systems and all that stuff and that's you're not you're not there with that yet but do you have an anticipated year that that starts to get implemented and is that year within this modeling and and procurement uh time frame yeah on on the grid Derm side or phase two Derm side we have made significant progress I think since we last spoke to the commission about it we've prioritized a couple of use cases which we're working to stand up the requirements around we expect to deploy that on a limited basis for a limited number of feeders um in the 2026 time Horizon that is primarily going to be focus on what's called flexible interconnection and flexible load connection so that's that's the intent and the timeline um as to whether or not it may have an impact I think that's really going to depend on how the deployment goes it's not just simply the the company deploying a piece of technology we need to get customers that are interested in the solution have them join into an agreement and ultimately enable them from a technology perspective as well so I think it's hard to say definitively whether or not it'll have an impact in this period okay so not so while you're expecting that you'll be implementing that Grid derms or phase two derms within this procurement period you don't have a program right now that you could model off of so it's not reflected in it the implementation of that program is not reflected anywhere in this modeling for the purposes of this um Erp uh i that's correct for the D well the DSP and the jts but really where we're looking to leverage grid germs is more from the the distribution side right right okay thank you sure no further question chairman i a yeah go ahead um on managed charging I think um I think Mr Good Enough had said there's about 5,000 out of 100,000 currently enrolled I believe in the dynamic charging um or the actively managed charging and then I think Miss Irwin had said there's another 10,000 currently enrolled in the passive um managed charging are those numbers right I just wanted to confirm I mean roughly I'm not going to hold you to the nearest the nearest 100 what what do we have enroll uh that's my understanding for the active manage charging but Mr can certainly correct me if my understanding is wrong on that she has a better sense of the participation in the programs okay U Miss misserin is reflecting that her Zoom is Frozen um if she could team chat I'd be happy to relay the numbers here yeah sure we get them speaking to m wi partially uh well and uh maybe we'll just start the next section uh commissioner Gan with that answer so if you want to move on to your next question my Mr mwin is uh unfrozen i' fre hery turkey okay am I back with you all yeah you are sorry um yeah what I said earlier is approximately correct it's around 10,000 in the optimize your charge passive managed charging program okay and that's where essentially the customers get I can't recall how long it is it's like a 10 hour window or something where their car charges but it is not actually like flexed or independently controlled by the company on a day-to-day basis whereas the actively managed charging that has 5,000 customers is more dynamically controlled by the company yeah that's right the the passive manage charging program has a couple of Windows of time that are like preset and don't change that they would be charging within um and uh the other one we would be actively managing that charging okay so as I look at the description of how the the manage charging is modeled with this kind of um steady load shape and then the tranches that just kind of move that behavior that to me looks a lot like the passive charging I mean like as far as what's modeled is is that a fair P to say so I don't think so um and I'll explain why so in general the way that you know manage charging is operating pretty much everywhere I think in the country has to do with sort staggering the start times of charging so you're not necessarily starting and stopping starting and stopping actively like during a charging session right it's more like thinking about when the vehicle needs to be ready and when you're going to start charging it in order for it to be ready in time for the driver's needs and so you know for a typical residential vehicle recharging overnight it's only going to need a couple of hours on a on a typical day based on how much people are driving right so so that General load shape is what we would expect to see just starting at different times essentially it's also the load shape you'd see from the passive charge system just that can iterate and like Circle back and and optimize but like the load shape still kind of looks the same right so I'm trying to um I know we only have 5,000 now in the dynamic charging with aspirations to go higher but we do have 15,000 it sounds like in total in managed charging that somewhat reflects the load shapes that we're seeing here so I'm curious why we're not starting with the assumption that we have the 15,000 people we already have in some form of managed charging that produces load shapes as I understand it fairly similar to what's being modeled um the Assumption 5,000 yeah good question uh you know that's the information we had at the time the forecast was developed um so that's that's the information that we that we incorporated is there a way to understand the impact If instead we started at what looks like 15% of current penetration of some managed char um good I would just say it's more like we're about 10% uh in terms of current enrollments versus current vehicles um just to be clear sorry what's that mean I I was hearing toal out of 100,000 yeah I think our our rounding in uh on today's call is pretty imprecise but it's around 10% in total 10% of vehicles it's at least double what's modeled let's let's give it double double to Triple so what would be the impact how could the commission and the parties understand the impacts of a modeling of managed charging that at least starts at the value of how many are already enrolled in managed charging yeah that that's an exercise that we could that we could do change the assumptions on the manage charging trajectory and see how that impacts the load forecast and as I understand from your motion that would primarily update to the jts um but if I'm reading the motion right no change in the load Seer because it's not possible um so I wouldn't see any impact in the DSP is that am I getting that right that's correct uh like I mentioned before L series of of Living Model that we continually update we're not able to revert back to the the state and and assumptions that were baked into our our forecast that we provided in in the PP filing so fortunately it's not as simple as just rolling back and reproducing that forecast and then it's it's also a lot of Downstream work that's pretty manual to recalibrate the risks associated with the system and ultimately the projects that are resolving them okay and I know I know we have a loader section later so I'll leave some of my more particular questions about loader in the updating process to that but just wanted to understand how we can understand if we better reflect what's even going on now on the system as far as man charging uh Mr Ry yeah I had a couple just couple quick General comments uh I certainly kind of hear the exploration around approval of the plan that's especially the DSP that's in front of the commission and just you know procedurally we will have pretty frequent DSP plans I think the next one is technically due in the first half of 2026 so um so that that's point one point two is that our scenarios are showing like a pretty close convergence in the in the immediate near-term these are great issues that all tend to have a greater effect in the longer term I would say um and then was going to offer I think uh we we've had some dialogue here on just you know just quickly addressing level one versus level two um various sources but we're you know the various data showing something like 2third to almost 90% of EV drivers are adopting level two um just to give some comfort there we could probably supplement that and kind of tie down the sources a little bit uh in the supplemental direct if that's helpful to the commission I was also thinking we could maybe uh try to maybe explain the algebra of the of the U what commissioner Gillan was Raising concerning the percentage of EV Peak versus EV sales uh in those two tables in that level to or I'm sorry that U appendix 2 I think it was called um so we we could probably try to explain that a little bit more fully too we hadn't heard that issue before before we're looking at it live here so just wanted to offer those those ideas uh I don't think we're going to take a vote and request it now but I would just say on my behalf i' I'd find that helpful if you wanted to submit it with a supplemental direct filing I I some forbearance on dates knowing that supplemental directs doe in a week um so if if these go past a little bit that might be what happens if that's all right okay um uh I guess commissioner Gman do you want to jump into the uh beneficial well why don't why don't we take like a little break uh till uh what don't we say 1 155 and then we'll come back and uh commissioner Gman why don't you start us off on the beneficial electrification and uh Miss uh feder Rico I uh sent you uh another graph so if you could be prepared to pull that up when my my questions come but we'll go commissioner Gman commissioner plant and I'll uh go last on the be so miss Fredo um I I had alerted her already I want to look at here at 105 BCM 9 which is the load your technical appendix so I think she's already got that ready but just as a note when we come back I'll be yes I have both of those ready for you I want to apologize I want to apologize and I'm going to do this transcript I was I was working out sorry uh all right let's come back atfe 1 15 let's come back at 155 thanks 155 thank you e e e e e e e e e e e e e e e e e e e e e e all right commissioner Gman when Mr Ry is back uh why don't you uh take us forward hello um so um Mr Riley just kind of I I guess more of a broad question in development especially of the DSP uh it seems as though you go from kind of the end uses and their anticipated uh profile of usage aggregate those up um and look to the different components Etc is that fair as far as describing the the DSP process kind of as a whole uhuh yeah I I think that is fair um you know I'd certainly look to Mr Mino to kind of color in the specifics who who does that work there there was one slide that I know we wanted to bring back which had about five or six process steps but I I agree with your your general characterization of it okay and then I'm a little less clear like with regard to the jts how the introduction of what individual loads will look like and what assumptions are made on those like how that factors in to the jts is it as clear of a line or is it more fuzzy how we get from the individual uses to aggregating all the way up to the jts forecast on on that and this is probably and I apologize for there's a lot of inter team work on this but I think Mr Good Enough could probably better answer that but but I think at a high level we're assembling all the individual components to build the aggregate forecast of energy and demand which are solved together as as a holistic system across the system coincident Peak through it through the Encompass modeling um so it's it's looking at the same component parts But ultimately summing those to uh the systemwide coincident energy and well systemwide energy and coincident Peak demand and that that's what Encompass is is is solving for whereas DSP and Dave could I'm sorry Mr Mino can fill in chapter and verse on that as as to how it it's done by each feeder and sublocation a different process but following the same basic building blocks um anything to add Mr Mino before I go to some of those end uses uh I apologize I came on about halfway through your question but I think um I think Mr iy accounted for represented the distribution system plan pretty well you know we look look at how these 25 different load shape or these different technology types with their Associated load shapes layered on to each individual distribution feeders curve and substation Transformers curve from a forecast perspective to really identify what we project on each individual asset how that risk is going to materialize and then we look at the locational specific nature of the system as we look to develop projects and mitigate that risk okay um Miss fero can we pull up that attachment DCM 9 so this is a technical um appendix the load SE technical appendix from the DSP proceeding um but it does show some of those um load curves for the end uses as I understand them so um and we asked you supplemental direct question on this so I hope you've had time to look at that and and kind of prepare for today um and this gets to some of the chair's question before when he had these kind of stacked on on top of each other on one chart but I just wanted to look at them individually for a few moments to just understand um my main questions with regard to each of these figures um it's like each of the next I don't know seven or eight figures is what is the base data of these how did we come upon these load shapes and then how are these load shapes used to develop those aggregated forecasts whether like on a specific set of distribution equipment and then kind of at large for the jts so um might make sense to start with figure 15 which is the residential um space and water heating and here we see the water heating has a very distinctive Peak at about 1M in the winter but not in the summer and then um very large Peaks uh H in the evening so just curious what data was used to assemble this um graphic yeah I'll I I'll uh ask Mr Shan heider to talk about the The Source data for this but how we use this to plan the distribution system is we receive a total amount of demand per each technology type um at the operating level of uper company level um and we use that spatial allocation process which I'll I'll detail it more during the actual lowere demonstration um but we use that to allocate those those adoptions of these these different Technologies to individual premises along the distribution system along with the associated load curves that we're seeing here for each technology type um those individual premises then get aggregated on an 8760 basis and layered onto the existing 8760 curve for each individual feeder and subst Transformer to develop a forecast for each that's unique to each individual asset so that that's the how um and happy to answer any more questions on that but if you wanted to get to the source data I might um ask Mr Sean heider to come on camera and explain where that came from yeah one followup before we get to the source data um so when you maybe maybe it's two offs I'm not so sure so when we get this load curve here um is this anticip you know you you determine kind of the the quantity I guess of adoption you you think you'll see through those processes and then apply them obviously to the existing loads and whatever um but is this same load curve used for everyone you assume adopts that technology by customer class yes so every residential customer received this curve um this is this is based on an you know an efficient water heater that um has representative of one we've seen on the system I'll let Mr Shan heider explain a little that in little more detail but um we're applying this to to you know hundreds of thousands if not millions of different points along the distribution system so this is an assumption we made to to uniformly apply this curve to to all adoption of these Technologies okay so you assume essentially every home follows this load curve if they've adopted this technology um and there's some just multiplication that happens there it sounds like um is there any diversity applied either on the distribution system or in the jtf um or is it just load curve times adoption uh it's it's low turve this is on a per unit scale here you see like 100% is the the peak that it hits so there's a there's a per unit demand that is applied that is multiplied by this curve to actually develop the how much demand that particular technology is going to add to the distribution system over the whole 8760 period and then that gets would be like per unit demand times the proportion shown in this curve times number of customers that that adopt uh bu asset essentially okay yeah number of customers behind such and such feeder that's connected to a given feeder correct okay okay and no diversity from that like oh maybe some houses are a little different than the typical load curve you know that's that that's a limitation of the tool right now to where we can't adjust the curve over time throughout the forecast is something that we're working with the vendor on to improve this and be able to account for some of that stuff going forward um but yeah that's that's that what you said is accurate okay thanks for clarifying okay Mr showen Haider welc me to the bast data that comes up with these water heating shapes that vary by season I I would be happy to and if we look at figure 15 we have effectively two technology s represented on there one is water heating and the other is the fully electric space heating on this graph and then and both for the residential case and then figure 16 is the dual fuel residential electric space heating which um we actually are going to be filing a correction to that graph because the graph that shows up there in figure 16 uh is is incorrect that's actually a ground Source heat pump load shape but the data used in the modeling was correct we just grabbed the wrong graph to plug into this particular uh appendix so going back though to 15 do you have a preference on whether you would like to start with water heating or space heating let's start with water heating since I think there would be I don't know if you have available the the proper curve that is what is modeled for the dual fuel um if you have it available that would be great to tea up so let's do water heating first and then we'll go into both of the space heating types perfect I I think that's a good uh a good plan so oh if we could pull that water heating back up um at a a super high level as I think it was Mr good enough and Mr Mino have alluded to we have a a different load curve for residential water heating utilized in the jts modeling than we utilized in the DSP modeling and at a high level the source of those load curves is the same but they're a different vintage and the underlying technical assumptions for the majority of both the space Heating and water heating load curves are the company's at the time most recent dsmb filing and for water heating specifically in the the the modeling to start the DSP process start started before the modeling for the jts and so the water heater load curve that is in the DSP modeling was based on information in the 2023 dsmb plan and if you remember that was sort of the first one or the last one before we fully implemented beneficial electrification and the modeling underlying the jts water heater load curve is from the 24 to 26 dsmb plan which we were uh working put together and and we got that done after the DSP modeling started so uh we have I think we've alled to before the what you see on the screen those Peaks show Hider real quick I'm sorry um does That Vary the adoption rate or does it vary the actual load shape the load shape is different between the two and and and significantly different between the two and it really comes down to in the DSP modeling you have a much spikier for lack of a better term load shape for water heating than we do in the updated jts water heater profile um and this is I think the you know sort of the Crux of the issue is the version that is shown here and that was used in the DSP modeling um it is more representative of what an individual or a couple of water heaters together would look like it doesn't really uh have enough diversity in it to be used at like a total system view right so you you do see very high peaks um those Peaks absolutely happen when you have a residential water heater and it's based not only on the uh usage of hot water but also the way that a heat pump water heater works uh in the vast majority of heat pump water heaters there is a heat pump heating element and there is also a backup electric resistance heating element and once you once the temperature in the tank gets below a certain set point uh that second electric resistance element will kick on and and you'll have uh very high peak loads for very short durations okay um now understandably like there are times today where probably a ton of people are using a lot of hot water and you see like a reasonable Peak I'm not clear why 1pm is one of those uh so the actual shape the the underlying data for the load shape from a water consumption standpoint uh in the DS plan comes out of enr study data where they actually study water usage and this particular time you know it's not this particular day is not necessarily uh every single day if that makes sense but this is the this was a day that popped up as as high usage right and I don't know if that's a you know a a preparing lunch or who knows what um that happens I think you're referring to that one o00 uh water heater winter Peak the the later in the day one is is similar and then you also see those kind of early morning when everybody's getting up to shower if that makes sense kind of like a tiny blip in early morning in a yeah big spike in 1M okay so it sounds like then the unreal data that you use if if I'm understanding correctly this does not come from your actual monitoring of equipment through your rebate programs which was kind of my earlier understanding in this um session but it's from enil data is that correct for for the water heater yeah water usage load shape we don't have uh to my knowledge individual metering on customer installed heat pump water heaters uh we do have you know through various pilots and research projects metering on air source space heating heat pumps but not on on water heating hot water heaters okay so this is from some enal data and it sounds like there were different days which give different uh Peak load curves upon which the company could pick which it used uh not specifically I'm sorry if I if I was unclear there okay you had said this was like from a higher use day than some of the other load curves I believe that you saw in that unreal data this probably what I could have said uh better is that you will have some days within a 8760 year that have higher usage so you will have some days that have high peaks and some days that won't have high peaks uh this is not a representation of everyday average together right and and in in when you think about planning a distribution system um and specifically as you get closer to the customer end of it you you can't really just average out every day and assume it's correct you're you're definitely going to see uh day night cycles especially in things like hot water usage um similar in space heating but this is definitely a peier shape than uh what we had in the updated version that was in the jts modeling okay but then the way Mr Moo uses this is and I get is this is supposed to be more of an outlier circumstance not like the everyday circumstance but this particular load shape is then multiplied by however many customers on a particular asset we're assuming get uh an electric water heater right a heat pump water heater yes that is yep um do do you do you think that that typical extreme condition involves that 1 pm Peak I'm just curious no and I I think as you'll see in the jts load profile for a water heater which is sort of like I said more recent it is a much uh uh more evenly distributed Peak okay um okay walk me through the electric the all electric space heat sheet okay so the all electric space heating curves again are they're founded in assumptions that are consistent with our demand side management plan and you really have three components that go into making up the curve one is the weather data you use and we use a an 8760 model so it has an hourly outdoor air temperature um and for the curves we use a tmy3 set of weather data typical model year uh three it's a it's an ml derived set of it's basically the they pick the most average month from the last 30 years for each month and then put that together into a year the of note the coldest temperature in that weather data is roughly minus 3 degrees fah um and it you know so it's not a an extreme year uh from sort of a historical perspective that's the first piece the second piece is we assume a sort of typical residential building and we assume that to we Define that by the the heat lost from that building as a function of outside air temperature right and when you uh think about it we have a a balance point of of 60 degrees which means at an outdoor air temperature 60 degrees there's no heat needed there's also no cooling needed and then the heating requirement on the space increases linearly as you go down in temperature and the uh the exact point if you will that or the the the equation that defines that is at uh 5° Fahrenheit we have a heating load on the space of roughly 38,000 b2s per hour so you could think about it as a a fourton I'm sorry a thre ton heat pump would almost uh provide enough heat for that on a five degree day right and that that linear extrapolation continues to to colder temperatures so those are the first two things that define the the hourly simulation then the third is the actual piece of heating equipment that uh that we employing and so for a cold climate for the for the all Electric System we use a cold climate air source heat pump with electric resistance backup um the specifics for that cold climate heat pump it's we design like that's where we deviate from the the DSM assumptions because we went with what I would consider an aggressively efficient heat pump um and so for a little bit of context if you're familiar with the the 25c tax credits they uh the the standard sort of for cold weather performance is the coefficient of performance the cop at five degrees uh there's a threshold of 1.75 which means at that temperature the heat pump moves 1.75 units of heat for every unit of heat it consumes or another way you could think of it is it's 175% efficient at 5 degrees so that's the the current 25 SE tax credit and which is B which is based on the basically the uh it's based on the center for consortion for Energy Efficiency ce's most efficient tier um there's also other standards out there like the doe cold climate heat pump challenge you may be aware of which is sort of trying to develop Next Generation heat pumps and the the compliance standards for that which are sort of a very futur looking super efficient heat pump is anywhere a cop of between uh 2.1 and 2.4 down at 5° and I sort of went through all that to give you a sense of the heat pump that we picked uh down at that five degrees has a a cop of 2.31 so it is a very efficient cold climate heat pump that's really not available on the market yet but it's sort of that forward-looking and and that's the one we assumed for every house um all of that said and done the that particular heat pump will meet the heating load for our all electric space heating um all the way down to about uh minus two degrees and so there's really only one hour in our modeling where you need the electric resistance backup and again that's an average weather year but that's what we're applying here um and on that coldest day that coldest hour the heat pump provides over 90% of the heat at a coop of about 1.9 which is really high still the electric resistance provides a little less than 10% of the Heat and that whole thing together is only using 7.15 KW and so the the overall this is a a very efficient very low demand all electric heat system for space heating a lot there I could detail um L luckily it's not my first rodeo with most big we' talked about this before got it um what I'm curious of is the shape um you had said one of the primary drivers is ambient temperature I'm not familiar with a day where the ambient temperature just goes down and down and down from 12:01 a.m to midnight um and then obviously you can connect this chart if you had the next day um there'll be a huge drop at at midnight um so I'm curious why are we not seeing any space heating bump in the morning why are we seeing some gradual increase of heating demand throughout the day which just in my personal experience is not how space heating yeah no and I I think what I I totally understand where you're coming from and I think part of it is what we tried to do in this technical appendix was to um help show the pieces that were that could help explain what was going on but we didn't really have a way to show all of the data in a very usable manner um but the other way we tried to show it here is if you scroll down to I think it's it's figure 18 so if you just scroll down it's probably on the next page of the PDF here um what we did is we showed there it is that uh residential be electric space heating 8760 load curve now I understand this is not um it's very hard to read in that the x-axis has 8,760 points on it right but what it does show is that day to night fluctuation in the electrical demand or the the load from this all electric space Heating and so what you saw up in in figure 15 above was sort of a very small section trying to uh to show that but if you were to look at this load curve like if you could sort of expand it out and see every hour you would absolutely see you know the the load on the heat pump or on the electric heating system very closely following but inversely the outside outdoor air temperature of that tmy3 model okay so is it that what we're looking at in figure 15 is not really a representative day um it's not necessarily a representative day I mean it's it's hard to the it's hard to sort of compress an 8760 load profile and the the actual what was put into load Seer was that 8760 profile right so we literally and and this is hopefully not stepping on Dave Mr mino's toes too much for when he talks about load Seer but there's a for every hour um you know first year all the way through 2050 or or till the end of the loader model uh there is a Bas load shape that is derived from actual system scada data right that's it's weather normalized and done whatnot and we we literally take this 8760 curve from the heat pump that's shown in in figure 18 and stick it on top now one interesting Quirk is that the load SE is based off a a a typical load year and so the weather used isn't exactly same and so you'll have like in both typical load year and typical meteorological year tmy3 like January is always cold right that makes sense but a particular day in January in the the tly like January 15th may not be a cold day but it might be a cold day in the tmy so you'll get some diversity there but the the easiest way to think about it is take the the figure all 8 8,760 data points in figure 18 and plop those on top of the same days the same hours in the the tly right okay um to that point I have a question on figure 17 which is the 8760 lcer for the water heating um so I I presume what you're modeling here is a heat pump water heater with some electric resistance back up and these spikes are when the electric resistance comes on am I right or wrong about that uh that that is probably very accurate yes okay so um if this just gets multiplied by adoption that we anticipate and then piled on to top you know at least in my experience with water heating like the um the circumstance in which you outrun what the heat pump can do and switch to Electric resistance it seems to me probably much more a function of what's going on in the house you're having a party a family reunion whatever um than weather necessar obviously there's some impact to like groundwater temp but not not that fast so what sort of coincidence like we're seeing some really big spikes here for the water heating that that are far different than the the typical load and I just wonder why would it be reasonable to consider that those would be coincident from house to house to house to house every house on the feeder has the same Peaks yeah so let me step back I think you're accurate in saying that those spikes are periods of very high water use demand um I would agree with you is and you can kind of see here there's Maybe seven or eight of them throughout the year um and that is typically some I say abnormal condition but those aren't necessarily coincident house to house um you know you do see periods around like the holidays as an example right where you have company staying with you or whatever and you might have more people in the house so that might be a little more consistent house to house than purely random um but no I I agree with you that the the the spikiness of this really should not be applied house to house to house and that's that's one of the things that as I mentioned earlier is different between the DSP and the jts uh the jts has a much flatter spikes right and a much more diverse load curve and that's definitely a a something that we would it is a later and newer version of our water heating load shape and it's something we will apply in future dsps and is there any way to apply that flatter Lo shape specifically on the water heating which I think we both agree is non-coincident largely from house to house in the DSP to see if that's influenced any any equipment sizing yeah and we I know we talked about this earlier about how to um what our supplemental direct testimony looks like but at a a high level and I'll I'm probably use slightly wrong words to summarize it we we intend to use the jts water heater profile and when we are going back and looking at those projects to see what the change will be and you know another like part of the part of it is is is and I can let Mr Mino talk to um the challenges around remodeling everything and going through all the grid needs assessment and mitigations and all that um but we do we do think it'll be in valuable to see those examples um of you know what would happen with actual projects and part of that is because there's they're currently is a significant amount of Headroom on the various feeders and so just one way to think about is just moving these Pikes down may not actually um change the load on uh a feeder significantly or change the mitigations and I don't know Mr you want to jump in yeah I agree you know as I mentioned before the the contribution from electric water heating to the overall individual system Peak like distribution theater Peaks really doesn't get expressed until mid 2030 which is beyond the is either the tail end or beyond the the the the 10year forecast that this five-year distribution plan is is accounting for so I I agree with Mark that I think that even if you were to flatten this out I don't think it would have a major impact on the overall loading of the system and and the need for the overall investment um and yeah to to what Mark was saying and I alluded to it earlier it's it's it's a lower is a living model we continually add information in there so we're not quite able to go back and just revert to all the assumptions and inputs that were used to generate the forecast that was produced for this DSP um that's why we were proposing the um the subset of projects and so this will have to be a manual effort to back out the existing curves and add new ones in there so and then not only that but then re-evaluate the risk and and look at all the need for all those different projects that are associated with those risks okay um so Mr Sean Haider just real quick on the space heating um the all electric your S pump are the bases I know you went through kind of the things that determined Its Behavior and performance um are those in in the record of the DSP or the jts in anyway now like the worksheets or modeling it sounds like this was company modeling this was not like from an outside Source uh yes this is company modeling I know we are in putting together our supplemental direct um we have a description of you know the the methodology the all of the assumptions that were that went into it as far as like you know the heat pump we talked about like that's not a real piece of equipment we we created that piece the specification so I know we detail all or planning to detail all of that in supplemental direct okay are you planning on providing the model uh I think we can yeah I me okay um uh let me see just one more question um can we go down to eight page 18 um so just a little further down the next one yeah um so here we see low curves for commercial um for electric space Heating and water heating similarly um and here we are seeing a little smoother water heating um but uh interesting shape obviously on the on the space Heating and understand a lot of commercial properties may not need Cooling in the middle of the day even in the winter but um the the ramp up is quite late in the day for heating I was just curious like what all modeling goes into this too we're we're only beginning to heat the building at 10 a.m. yeah so at a and I'm going to take a second to re acclimate myself with the figure okay um the so first and foremost and and this I think is is doesn't specifically answer your question but is relevant um the adoption rate for commercial space Heating and commercial water heating is significantly slower than it is for a residential right and so the overall impact of uh this work is is far lower um and and I think the later in the appendix figure 26 I think speaks to the the impact of spa of commercial space heating water heating and and there are figures and I'm just looking at my version on the side here uh versus like figure 20 talks about it for uh residential uh space and water heating so putting that aside for a second I mean the macro point there is this is a much much smaller load um the the methodology there's there's frankly a lot more variability in commercial space heating um and even in commercial just uses in general as well as there's far less uh practical applications meaning there's less actual electric heat pump space Heating in commercial buildings than there is in residential so with all of that said our methodology for building these um relied almost exclusively on external sources to tell us uh things like what are the various building types so we we said what's the energy use intensity for uh different building types in uh you know and that's from an external Source how many of each of those building types is present in our area and what is the uh you know what's the the performance of each of those building shipes and so what you see here is a much less refined um amalgamation of those sorts of data than what we did on the residential side given the relative amount of load from each of those Mr Jero yeah and and this is again this is representing the peak day um and and represents what the L would look like on on a peak day the 8760 curve I think it it balances it out a little bit more here um and as marle L to like yeah agreed that the adoption rate and if you look at the growth projector projections in this append to the commercial adoption is is significantly lower um and also you do lose some of the granularity on an 80 on an hourly basis realistically these these devices are operating on a much more granular basis so you do lose some of that that that Nuance when you when you produce things on an hourly basis like that okay um uh I'm going to move to another document if that's okay with you let Miss Frederico take this one down so she can fill up a different document for me um here you 105 which is um am I calling you right it's Mino or AMO I'm sorry uh we pronoun my family pronounce it Mino I'm sorry I'm probably gonna it's fine I'm used to Mino Mino no worries okay got it um has direct testimony here 105 uh just let me go to page 26 oh this is confidential I don't think it's got yeah this is not a confidential page so I think for and I pulled the public version anyway oh okay good thank you um so just looking at this uh graphic here which shows like the non-coincident kind of additions I think to the base forecast um you can see that that's second from the left it's a tan color is capacity checks I've been a little confused about this in the past like I call in what would I call the be barriers proceeding but we were looking um especially as as part of that at um constrictions to capacity for new projects in Denver metro area and I understood through that proceeding that while the company provides courtesy capacity checks to customers who might want to interconnect especially large loads developments bigger buildings um that those were not used in forecasting or used in any way but here I'm seeing you know a not insignificant addition in that tan from capacity checks so I was just wondering you know at least I understand that process that the company only forecasts for and builds for and counts on the load coming once there is an actual formal application from the customer and a capacity check is not a formal application from the customer so I was curious if you could help me understand does the company use capacity checks in their forecasting or not um just to help me kind of settle that in my mind absolutely so I think you may be conflating two things there's conceptual capacity check which is the courtesy that we do to customers that are not quite ready to apply and then once they do submit an application that goes through our capacity check process and that's the point where we distribute the all we allocate the capacity for those specific projects who have applied for distribution service so the values that you're seeing here are the customers who have submitted a service application and we've allocated the capacity on the distribution system to them for their projects the conceptual capacity check we do not in explicitly include into our forecasting process we do leverage some of that information to help inform propensity for adoption for something these Technologies but it's not implicit explicitly included in our forecast okay thank you yes I think it was the semantics there confusing the addition of the word conceptual for one and not for the other okay so these are like verified applications we this is our known growth as we refer to it because they've submitted an application they have plans they have panel schedules they're they're ready to start moving their projects forward okay understood thank you for that clarification I think we're done with this m freder um just wanted to follow up to to understand the the forecasting process for both the DSP and the GTS um you know there there's been a lot of discussion on what was known at the time those forecasts were started or finished or whatever it was can you just kind of enlighten me as to the broader schedule of developing those forecasts for each of the proceedings so I better understand the timeline um you know in the lead up to a proceeding in which those are done or were done in this case yeah I think it would be ful if we pulled up a slide in our presentation that shows our annual distribution planning process um there's a lot it's an annual process and it it does take the full year to get from point A to point B so I think that would be helpful to kind of illuminate how we plan the distribution system and when those inputs are really needed so that we we can account for them in our forecast so um Mr count if you're able to pull up slide I believe it's 29 in Our Deck we'll give it give it a I me to pull that up but um once this slides up you'll see you know what tools we use throughout the distribution process um and all the different steps that are involved from starting the the forecasting process all the way to getting our projects finalized and it looks like uh Mr Larson has his hand raised so maybe I'll pass to him in the meantime yeah Miss uh Miss Federico is it possible to undisable Mr Cen again so that he can put the slide that Mr mino's looking for it uh yes if if he left during the break that's probably why he lost the capabilities I apologies we won't do that one again he's all set thank you great yeah so um there's a lot this is a pretty busy slide here but I think it's really helpful to kind of contextualize how we go about the the playing of the distribution system um and and why this is part of the why it's it's not feasible to fully comply with the request in in the uh that was asked for in the supplemental direct here so we start our forecasting process for distribution in September after the summer Peak has concluded uh our planning Engineers look at historical loading data to identify our seasonal Peaks our minimums that we use for various portions of our distribution analyses that is that is performed using skated data that we have at our substations that measure feeder loading and individual substation Transformer loading and that typically takes about a month to do given that there's 813 uh distribution feeders 275 substation Transformers that serve our customers um once we have that ironed out we go through and and this is where lower comes into play it's it's not like Encompass that that does the entire um forecasting and and and mitiga a process and economic analysis the low seers is used for a single component of this process and that that core function is to develop those 8760 scenario forecasts by asset and asset being feeders in substation Banks so our our team scrubs historic skada data we remove anomalies where we are able to um come up with what we call typical loading year curves which are weather normalized curves using the the most local weather station um for each asset to weather normalize these curves uh and it produces three Tois um one at the the zeroth percentile one at 80th percentile and one at the 100th percentile uh ex most extreme weather within the last 30 years so we use that 80th percentile T to plan the distribution system the output of load Seer is that 30-year 8760 forecast by each node or asset um from there we we take that forecast and we have an algorithm within our a powerbi tool that we built to compute um 8760 risk analysis for every single individual node so we look at what the magnitude of these overloads are how many consecutive hours these are at these these um these overloads are at risk um how many hours throughout the year that's that's really helpful in quantifying and comparing these risks against each other as we look at how we're going to fund our our distribution system plan um helps us to identify the severity um helps us identify what's the most appropriate mitigation um shorter shorter overloads and shorter mitigations are maybe more conducive to an NWA whereas large magnitudes for you know multiple consecutive hours for a lot of hours throughout the year may not be well suited for something like that and may require more of a wi solution so we use a combination of our powerbi tool our modeling tools to really analyze those RIS RS um and then we start looking at how we're going to mitigate those projects um I skip some of the durations here but you know the the forecasting process in lows here takes about three to four months depending on um this all this all the different tasks that are really involved with that um I kind of highlighted a few of them but there's there's a lot of nuance that goes in there and we can touch on that when we get to the demo um and then the risk analysis that takes about a month to confirm certain system parameters get that information in there um make sure the output makes sense and is Computing properly um for the mitigations we're we're building these models and analyzing them in a powerf flow manner to really look at how these mitigations are going to be proposed we're it's it's very specific for each individual project that we're trying to do since each of these feeders are very locational specific so you know we're looking at how we can route this infrastructure what kind of assets are already deployed on on specific spefic roadways it's it's that granular so it does take a significant amount of time here um that typically takes about three months to kind of iron out what we think mitigations could be as well as some Alternatives that we could potentially evaluate um and then we go through a peer review process to dep pict the appropriate project and then we get to our Capital project development and and budget creation so this is where we're really honing in on the specific scope of the selected project um iring SMS getting that that into our five-year budget and and ultimately developing the specified scope for each individual project that we're pursuing so it's it's it's a pretty lengthy process um so that that's it from kind of start to finish here and then you see that you know there's a lot of tools there a lot of manual processes that go into it um so it does take quite a bit of time we are we are trying to improve this process and and create more automation around it but um there's a lot of engineering judgment that goes into it as well okay yeah I'm trying to and commissioner PL you you'll probably be up pretty soon after that's really like my last question I think um I'm trying to reconcile you know my understanding from the testimony that say like the clean heat plan outcome came too late to be integrated into this forecasting process and so we have in there an outcome that is not the actual outcome of the proceeding um you know it's not incredibly different but I think I talked earli the electrification is off by 20% maybe in terms of what's expected from the actual um commission decision so I'm trying to reconcile the idea that like this all starts so early and it's so baked that we can't get something like that that's more accurate in to your process and reconcile that with the company's recent motion that the load series is also so incredibly Dynamic that it's changing every day and it's constantly getting every day I added that word but that it's it is changing it is different than it was when you modeled it and so there's no way to even go back a few months to know what it was and so it's like on the one hand it's so rigid and maybe it's just that shot in time you know but that's what I'm really struggling with understanding here is so rigid we can't get re relative recent commission decisions in it but it's so Dynamic we can't still use it to calculate other things the dynamic portion of it so throughout this entire year as we get new capacity checks new new service applications we're adding them into load here as part of our process so um we can't just revert back to what this and and we're also updating assumptions as we prepare for the next forecast because this is an iterative process where we're constantly challenging ourselves and improving upon our our assumptions as you saw in in the filing and in my testimony in the plan like you know we've we've taken a pretty big uh advancement in terms of how we're forecasting and and we're going to continue to do so as we go forward but um we're constantly prepping for the next forecast batch um throughout the year so those assumptions do get updated um those new lows get added as we refresh the forecast and we use that for like especially for the cap capacity check forecast that we use to evaluate newp growth on the system so there is there's not like a snapshot in the tool that we can just revert back to everything that was in place at the time that we generated the the forecast for the the DSP filing here um that that's the dynamic nature of it from that that that we were referring to before and there's no like mat use silly terminology but there's no save as there's no way I we we're working with them on that yeah we're working on that with them right now that's not something offered in the tool um you know that was something that we've been actually talking to them for not just based on the outcome of this this hearing here but um or this this conversation here but it's something that we've been talking to them for several years now that was something that we would want to have so I think that's something that's in the pipeline to improve upon but it's not capable of doing that currently um and then also to your other point you know we need to make sure that we have all the inputs ready by the time we're starting to kick this process off within lower so all those different forecasts to make sure that we're ready to compute the forecast and and keep our our process on track here so um if if I think that's where the the challenges with the timing of all the different hearings and the outcomes if if we don't have all the the up most up-to-date information by that like in September early October we kind of have to roll with the the most recent version that we have that's ready that's complete and ready to use um so we can we can develop our forecast and it look like Mr Mars may have some additional comments to add there yeah thank you Mr Mino and and yeah all right I'd like to chim in I certainly hear the conern I think Mr Mino offered examples actually where we're helping the oems improve their products um as again I'll kind of reiterate there's not a One-Stop shop for modeling and I certainly empathize with the sense that this feels very rigid I think we've in a number of proceedings now laid out the planning process and we we've put information in front of this commission to try to figure out solutions to better work through these proceedings however I think there's another dimension here and I kind of think back to you know engineering school and math is you know as you're as we add planning variables and we increase the sophistication of the planning you know there's that many more equations to solve for if you will and there's a concept in planning which is you know around snapping the line so it kind of becomes this this very difficult to solve for construct if if we continue to add variables to the process of solutioning our systems it's going to increase that rigidity and that complexity to solve for so I'm very happy with the work that we've done with this commission the way that we're evolving and I certainly hear the concern which is why we're pushing our vendors but there there's a lot here that we're now asking our planning Engineers to solve for you know I'd say directionally it's an order of magnitude 10 times more complex to solve for a singular feeder now than it was 5 years ago so I think your comment on you know the the timing the delay and like clean Heat versus this DSP I agree with you this this DSP is probably conservative in comparison to what was came out of the CHP outcome that's something we need to think about that's where that issue of snapping the line where that comes to impact us on the planning construct and getting full um full transparency into how that's driving investment thank you um I think those are my only questions for now chair uh thank you commissioner plant thanks and I I don't I don't have uh nearly nearly as much so I'll go a little uh quicker appreciate it um so I think um I'm trying to remember who it was was answering a question from commissioner Gilman about the um the uh the the heat pump U cop I think it was a cop of two 2.31 at um minus it was Mr Sher that's right um uh at uh at five degrees and that's a highly efficient heat pump um and then if I understood you correctly at min-2 it's 1.9 and that's when you start to add in uh resistance heating backup heat is that correct yeah and and what might be helpful is if we can pull up slide 25 in Our Deck this kind of summarizes I and Neil are you running the deck if if you are if you want to slide down to slide 25 and as you're as you're pulling that up one of the questions I had around that is if you're if you're operating at a 1.9 Coop why would you then add in resistance heat which is you know one yep that's a great question and one of the uh pieces of the specification I did not go into detail on uh with commissioner Gilman is the heating capacity the output of the unit um which also changes with the outdoor air temperature and so uh not to dive too far into the thermodynamics but effectively if you have the same size heat exchanger and same refrigerant flow uh the larger the delta T between the interior space and the outside space from which you're getting the heat is the less not only the less capacity you have but also the less efficient it is those kind of make sense right if you have a certain amount of compressor power and you can move less heat it's going to therefore be less efficient to do it right and so the the table on the right side here is the uh operational characteristics for the piece of equipment that we modeled and you know as a footnote this is not there is no existing piece of equipment that exactly matches this um but this is the sort of hypothetical aggressively efficient future piece of equipment that we modeled and what you see is the heat pump basically maintains its capacity uh from 47 degrees Fahrenheit which is how heat pumps used to be rated um all the way down to five degrees I think we have a slight degradation there um and then from five on down it it loses uh about half of its capacity between 5 Dees andus 22 Fahrenheit and and we actually have uh the unit actually would shut off at at 23 below which I to my knowledge there aren't any heat pumps currently available that um have a a low temperature cutout um at or below minus 23 most well not most but the sort of the best in the in the business right now cut out at minus 22 and then they won't reenable until it gets up to about minus1 so um hopefully this chart kind of helps answer that so and the the other piece of it is that the heating load the amount of heat you need on a building increases linearly as the temperature gets colder right and so um you know if we have a a heating load on the building of I think it was roughly 38,000 BTUs at five that keeps going up as you get colder so you kind of get a double whammy maybe a triple whammy actually of the space needs more heat the colder it gets the heat pump provides less heat is able to provide less Heat and the efficiency of heat it provides is lower and so you'll see a uh more of a a you know sort of a quadratic function if you look at the actual KW that's used but the the other thing to note and I probably should have drawn a little line on this figure is the weather data that we used um and I think I I say it in the text here the coldest temperature that we experienced in the 8760 load profile that we built for space heating uh was that minus three degrees so we're still in the range of a very uh efficient heat pump you know versus if you think about like our gas design day um which is what you know like our gas system size for from a heating standpoint uh for most of metro Denver is in the range of 25 below so you could see that the the KW from an all electric heating system in a house would get significantly higher if we modeled it down to you know 22 25 below than what we actually put into Seer which was as I mentioned before it's down at the bottom there the the maximum KW from any from all and it's the exact KW from all is 7.15 KW great okay so it's really that that backup uh resistance heat is to meet the Delta of the BTUs delivered that because you're reduced in your number of BTUs that you can deliver per hour yes exactly okay understand well I think the other Salient point there is is that we didn't just switch to that electric resistance it's only shaving that that very small P your overall yeah your overall system cop is still uh 1.85 which is the combination of the heat pump plus the Electric resistance right gotcha okay thank you and um this is a more sort of basic question the uh the y axis on the the graphs that we were looking at and that um that attachment are Peak unit load per percentage can you describe to me what that's telling telling me is it are you talking about each individual unit are you talking about um what what's the peak unit load percentage actually measuring I think I'm gonna let Mr Mino take that one sure um Can Can you um clarify which curve you were which uh graph you were talking about yeah this was in your um attachment d i my glass D CM9 dm9 yeah so those 8760 curves are what we call a per on a per unit basis so the the scale is on percentage so the way that it works is each technology will have a a a a maximum demand that we're we're allocating to it and then it's Multiplied on that that per unit scale to scale that 8760 curve to that technology type okay so each technology type over the entire distribution system is going to have a certain number that's associated with 100% of what you're allocating for that technology on the distribution system and correct where you are within that number that you allocated on the distribution system correct so the peak would be that 100% so that name plate that we're referring to and then throughout the entire year it's a percentage of that Peak based on the operation of the curve essentially okay and that that Peak is essentially your total number of those units that you're modeling um uh and and if they were all running at once at maximum that would be your that would be your 100% uh that's the is on a per device standpoint but yes all the different technologies that were were allocated on the system would be coincident with that because they're using the same curve but essentially that 100% is the the maximum that we out that we're we're allocating to each device um and then it scaled down throughout the year um and maybe Mark you may want to yeah that a little bit if I may I I might just talk through the actual numbers that might help reinforce the concept so the the all electric space heating piece of equipment we were just talking about that has a the the the maximum that any single one of them will hit is 7.15 KW and looking at figure 18 that happens right at the end of of December and the reason it happens right at the end of December is that is when in that 8760 TM Y3 year the weather data that is when that minus 3 Fahrenheit degree temperature happens right and so what you would see and I I I this kind of relates back to a conversation that commissioner Gilman and I had earlier you know the the KW for the all electric heating will very closely follow the outdoor air temperature meaning when it's really cold out you'll have the highest level and so what this shows you is and I'm just going to kind of use that I'm going to round it to 7kw just for for discussion purposes we have a we would have a 7kw Peak from every one of the all electric space heating customers in that right at the end of the year at the end of December um if we move to the other side of the graph so it looks like early in January there's a point that is roughly just going to call it 08 for for to make the math simple on that day on that hour each of those all electric space heating heat pumps would be using roughly 5.6 KW the 7even times8 5.6 right so you can scale all the way through right and so it's if we were able to zoom in on this figure you would see that throughout the day you know it's going to be when it's colder outside you're going to have it higher and it's going to vary a little now one other caveat is we um in order to smooth the load out we did not use any setback thermostats in our modeling and so like furnaces if you like many homes that have furnaces have a setback thermostat and when you you know when you wake up the temperature may increase four five degrees whatever and you'll have a much uh this it'll be running as fast as can for you know period of an hour or two to catch up we have none of that in our model we very conservatively sort of layered that you know smooth that all out so you will see like I said a very uh inverse but following relationship between temperature and KW okay great thank you very much for that description I appreciate it um last question I had was um as it relates to the the aerosource heat pumps I didn't see much uh discussion uh related to Cooling and how that's being modeled within the within the model so I mean you're going to there's certain assumptions you'll make around existing air conditioning that's replaced by beneficial electrification with more efficient uh cooling but then there's also homes that didn't have any AC that now all of a sudden have some so I'm just wondering how did how do you handle those assumptions within your overall model yeah the the short answer is as we added the space heating heat pumps we did not make any adjustments to Cooling and it's exactly for the reasons you indicated in that within those adoptions you're going to have some customers that previously did not have Cooling and they will now and you'll have other customers who previously had Cooling and the new heat pump may be more efficient than Cooling but actually many of the the cold climate heat pumps especially may actually be less efficient at Peak than the existing Cooling and so as a simplifying assumption for the new heat pump adoptions we effectively neglected the the cooling so I just assume that it balances out basically yeah for the new adoptions and Mr Mino I know there's some underlying assumptions around like embedded Cooling and whatnot I if you want to speak to those yeah so we also we we allocated residential growth commercial growth industrial growth and then the cooling demand is is baked into the allocations there so it it's more inherent and and the load shapes also reflect that for those allocations so the the the cooling demand is baked into the the other tr the other technologies that we we allocate not not necessarily represented here if that makes sense okay thank you appreciate it I have no other questions Mr chairman uh thank you commissioner plant uh uh given that we're three hours uh into this and less than a third of the way through it I'm gonna just uh focus on uh a handful of critical my most interesting uh or most relevant questions um you guys mentioned that the water heater load shapes are based on uh enal data at some point can you just file with a supplemental direct for uh separately uh what enal data that's based on we can yeah thank you uh Mr Mino as I understand it your forecasting goes through 2034 and that new distribution infrastructure is proposed in the current five-year planning Horizon and in this case for failures of the N minus 0 or n minus one test through 2020 through 2034 am I wrong about that oh no it's accurate we're using a 10-year forecast and to influence our five-year budget that's correct so the seemingly somehow somewhat ad hoc water heating assumption you discuss with commissioner Gilman are impacting the new distribution investments in this filing consistent with the 2030s graph I had previously shown isn't that right I will caveat that the graph that showed new growth on the distribution system it didn't show the existing load shapes um those are residential hot water heaters and and typically residential feeders Peak at an earlier time but typically between like five and8 is what I've I've primarily seen so I I think that that graph fails to capture the overall existing Demand on the distribution system and how that new growth factors into it I I so I I I know caveat that there that that is being that is part of the growth that we are anticipating in that 10year portion but I think that graph doesn't fully capture the full picture but but that I mean that's just the that graph is just your 8760 I'm just show showing it graphically and that's what you use to drive your Capital spending right that was layered on to the existing system demand for each asset so that that graph that was shown was just the incremental growth on top of that it didn't capture the existing load profile of the distribution system um and also it's kind of hard to extrapolate things at at an aggregate level for the distribution system um it really the way we look at it is each individual asset and how that that loading pattern adds on to the existing system demand for that individual feeder or individual substation Transformer it's it's really hard to make generalities about the distribution system at and aggregate good uh Mr Pollock did you want to add something I was just hoping to clarify a little bit so while there is a 10year load forecast reflected in the DSP the The Five-Year budget capital budget is based upon the next five years of activity to serve those capacity needs so with the exception of some long lead time projects like substations um you wouldn't see any investment being triggered by let's say a 2031 load or 2032 load being reflected in that capital budget uh that's not what I OD Mr Miner to say what he said was that if there's a 2031 n minus 0 or n minus one test failure that is going to show up in the current five-year Capital plan so there we initiate projects within that Five-Year Plan a substation takes a long time to construct um you know I don't think we anticipate all 35 of the new distribution substations to be be in service by the end of 2029 which is that 5-year window but given the amount of time it takes to build a new substation we initiate them in that five-year budget to look for land to start the process to start some high level engineering things like that so some of that cost is reflected but it's not the entire um scope of that equipment that we be building out essentially I think sorry go ahead Shar no uh please I I think it'd be fair to say Mr Mino that most of the projects aside from those substations of which there are significantly fewer quantity those are projects that we'd expect to occur long before the 2030s they would be budgeted and and put into service well they'd be budgeted and reflected in the budget in the next five years and likely put into service within a few years of their in need date is that a fair characterization yeah new substation Transformers new distribution feeders that we have funded in the five-e budget we're anticipating to be in service within that Horizon so that's more some of those longer range assets that we need to initiate early on to make sure that we can have them in service by when the system needs it which may stretch beyond that fiveyear budget uh Miss fed Rico can you pull the second uh graph I sent and despite the July date in this heading let me represent to you that this graph is showing the system Peak day in August of 2030 again from the 8760 that you sent us through manage charging or aggressive demand charges and time of use pricing couldn't we flatten out this load shape in ways that might uh greatly reduce the distribution Capital spending don't mind giving me a minute here just to to process this graph it's the same as the other one it's just uh on you know given that uh your comments that um you know a winter peak in 2035 was less relevant in determining investment in the uh next five years although it sounds like still somewhat relevant um I asked the team to pull together a 2030 summer coincident Peak day um and it's still pretty spiky um so I guess the question is um couldn't we flatten out this load shape in ways that might greatly reduce the distribution Capital spending yeah so I would just reate a comment I had I mentioned previously is that this doesn't capture the existing demand shape and and how the overall new load impacts each individual asset it it's it's it's really challenging to look at the distribution system at an aggregate level like this we really look at this on a locational specific basis where each of these assets are located how these shapes are are in influencing the future shape load shape based on the existing demand that's already on the system so I I struggle with with that question a little bit because this isn't fully capturing the entire picture yeah maybe simplify that a bit um it's hard to distill or understand whether there'd be any Investments triggered just by looking at this load shape at a coincident basis you have to take the load shapes of the the 25 load shapes that Mr Mino team models and load Seer and look at the aggregate load shape and then go apply that against the existing capacity on a feeder which there's not unlike the bulk system where we've got you know one turb limit we're planning to with our planning Reserve margin each of the available capacity in each feeder is going to be different so it's hard to say whether or not this is going to trigger investments just from solely looking at the peak even this the the second part of your question chair I think there's absolutely opportunities to explore reducing the Peaks think time of use can be challenging because those rates are generally designed to trigger and avoid the the bulk system needs I'm not aware of utilities that have implemented locationally specific distribution rates in the way that time of use rates functions today um on the managed charging side I think absolutely you heard from Miss Irwin a little bit earlier we're working with our managed charging vendor to start looking at ways we can optimize charging specifically for reducing distribution impacts we're also working on some initiatives with a vendor which we mentioned in our DSP to look at opportunities for load flexibility using our Ami data to help flatten and better integrate both load and de so we're absolutely committed to looking at those opportunities um but I also want to reiterate that this graph alone is not necessarily conclusive enough to say that it's driving investment you have really have to look at it in the context of available capacity in every individual feeder and asset well I'd just say we asked for the residential uh uh the total residential load and uh I don't think we were able to get it although maybe it comes uh in the March 21st supplemental direct uh filing but uh back to your supplemental direct uh testimony variance variance motion is it reasonable to ask us to approve a multi-billion dollar um set of distribution of Investments without understanding whether it could be significantly reduced or avoided by authoring the the load shape during maybe a a couple of hundred hours per year I'm going to jump in Mr chairman um I'm G to maybe take a guess that you're looking at especially that the shape of the orange the EVS you've got a lot of peak driving from that new load whe whether or not it's it's hitting a specific feeder or sub which you know Mr Pollock is is appropriately caveo against that assumption but just looking at this Orange uh you know Miss IR probably has some thoughts about the cost of uh managing EV charging that that's also not free and just eyeballing this graph which everyone at Excel is seeing kind of for the first time this is this reflects to some extent you know this is a residential graph this is to some extent when the vehicle is home um we're seeing you know hour 18 here that that's 6 PM the vehicle has arrived at home um it may be gone from 8:00 a.m. to around 6:00 p.m. or so um so that is some degree shaping that it may not be as shapable as as we might like um and I think that's part of the factor here so I don't know misone if you wanted to add more onto the um you know kind of our more recent thinking on managed EV charging yeah happy too um so I think um to expand on Mr iy's point you know we always want to be managing load in a way that is cost effective compared to Alternatives um that would be why we would be doing it and so you know when we think about moving from managing EV charging on the bulk system and responding to bulk system signals and as Mr Po talked about earlier um being managed through our commercial operations um you know sort of signal dispatch to an approach that manages charging at the distribution system that is more complex to administer and requires additional systems like the durm system that Mr Pollock was talking about there're also implementation costs associated with that primarily with regards to data and integration with auto manufacturers and vehicles directly um so you know the more complicated of an approach we're trying to take the likely more it will cost us to accomplish that and so while we very much intend to do that um we don't yet know the full scope of what that costs and so I think you know as we move into the future here you will see us moving towards managing charging for distribution system needs but we will also be learning more about what that costs and you know from our initial look at that you were expecting that to um to have a material cost right and so I think we don't want to sort of jump to a conclusion about what is a coste effective approach to managing that charging and and need to be thoughtful about that I guess just two questions it just looking at the graph and again this is just the 8760 you sent over it does seem like the distribution Peak is fairly coincident with the generation system Peak maybe it uh hangs around a little later um um but two I mean well um all right I I guess I'll stop there um uh if I could share one quick comment on you know our current approach to managing charging which is focused on the bulk system is also substantially oriented at maximizing the use of renewable energy and so I think that's just something to also think about the hours of the day in which that renewable energy is available um as that is you know what what we are aiming for at this point with our program understood um well uh that that's it for my questions on this uh I just would say um I'd like to uh on advice counsil I think we'd like to take up uh um the variance motion next Wednesday so if parties have uh comments uh they might try and uh get them in uh say Monday although I don't think we're going to issue an order in this case um but I just uh throw that out there um given that a number of the parties uh are on this call um uh should we move to the next uh subject item Mr iy you want to move to the uh Data Centers do you have a um an introductory uh set of remarks or should we just jump in the commissioner questions I don't believe we had any introductory remarks I think we're bringing up uh Mr Bailey to uh primarily represent that issue uh in conjunction to some degree with Mr goodno but I I think we can go right into it unless Mr Bailey tells me otherwise uh he looks uh ready he's poised uh why don't we uh just keep rotating uh commissioner uh plant uh questions on data centers for Mr Bailey yeah I was wondering if um in in your model you are um I mean you break out large loads among your sort of different kinds of large loads um data centers are all kind of uh as far as I could tell treated pretty much the same and I was wondering if you're looking at different types of data centers are they you know crypto mining are they AI are they you know some other sort of uh sort of data center and are you incorporating any assumptions around um flexibility um within that demand that might be um that might be feasible for example um my understanding is that about 50% of the compute um at a AI Center is is on learning and so that might be movable or adjustable to a certain degree and I'm I'm wondering if there were any sorts of uh assumptions around uh flexibility of the demand among those different types of data centers uh yes thank you uh thank you chairman blank commissioner plant commissioner Gilman appreciate the question um I may I'm gonna answer that maybe in a couple parts and I think the modeling uh have Mr good enough for Mr landram jump in but I think to your question on I want to be very clear the the data centers that we're modeling in I think tlb1 that you referenced um are are not crypto they are are truly a either a data center AI think about AI or machine learning in that regard um we we do not in this initial phase of what we've produced differentiate between you know an inference model or a learning model Etc we're utilizing the feedback we're getting from those customers in verbatim um regard to load flex and I think as your your follow-up question we have significant discussions with all of these customers we do this across other Industries as well but specifically with data centers we've talked about their capability um to shift load or reduce load uh currently a majority predominant on-site backup generation is our diesel gin sets um in that regard the technology for let's say battery storage or maybe converting to um natural gas as an example which could be more efficient the the the cost and the technology is just just not there and so so we will continue to have those discussions the uh the point about the model we don't take into account any type of shift in that regard to your question but I may have Mr Good Enough weigh in the last point I'll make on that is as we have spoken to these customers when you think about their capital investment in in the data center you know it's multiple billions of dollars and what they are challenging us with is we need the reliable generation and capacity there we want to utilize our technology um to run the syst to run their system internally that's that's their most profitable source and so they are still continuing to pursue load Flex I think what they will look at as they pursue that is can that bring generation on sooner can they move their timeline to the left of getting onto our system and so um we will we have had those discussions and we'll continue we're looking at other resources and challenging other utilities to maybe give us some feedback on what they are doing um it's just not advancing as quickly as we would hope so so my understanding is the assumption is that there's not really any load Flex capability and if there were any that would that would be a additional benefit to the system but in terms of modeling there's no there's no load Flex that's assumed that's actually uh yeah that's correct and and you captured it uh correctly earlier commissioner plant and then and then Mr barley said it correctly as well we are making the same assumptions for all of the data center load at this point uh I would note that they're all you know anticipated to be extremely high load Factor so you know there might be a differential between a 95% or a 90% but either way it's going to be a pretty flat load and as we start to see these customers connect and actually you know ramp up and and start to use the system we'll learn what their load shape looks like and kind of make that adjustment going forward as we do our modeling but now we're assuming a a very high load factor and do you have any sense um of I mean obviously as you increase load on the system and you increase sales that can have a downward uh impact on on on the rates however you know you're also having to add a bunch of generation and possibly transmission and possibly other infrastructure improvements that add cost cost of the system are you getting any sense of where that balance comes out uh does it come out to you know the positive uh for the for the rate payers in your modeling or you you coming out to the negative or are you coming out basically flat uh I I can I can jump in uh commissioner plan this is really the subject of the long-term rate forecasting part of the agenda but we can we can hit a little little bit of that now I mean we see a fairly um fairly robust Trend where we when we have the additional loads on there and subject to some pretty pretty simplified and somewhat fungible cost allocation principles but the strong uh trend is that when we when we're adding the new emerging large loads we see a positive benefit to rates meaning the other customers could could realize a benefit to that and and we can we we laid out the there have been several requests from the commission we actually laid out the results on a slide that we can bring up later but maybe um we can park that for now but I think the answer is yes there is benefit we're happy to discuss it in more detail and do you know if the modeling on that and again if you want to talk about this during the rate uh impact discussion does that include any potential transmission uh investment yes uh it's it it sort of elevates the capital across generation and transmission and sort of moves it up and down depending on the assumption that we have made for the large loads coming into that um and that's that's what we we modeled at some some level of complexity in our long-term rate forecasts that that we have I know that um you yeah thanks for that I know that you um you've modeled these these various impacts and obviously the impacts are different depending on where the data center or the large load is located on the on the system um and I know that you have you know some incentive uh proposals around uh locating load in some of the JT the just transition areas um have you noticed a a um decrease in cost to the system based on where you might Target those loads and have you identified those potential areas where you could decrease the cost of the system overall in terms of um trying to attract those loads uh Mr Marx I see your hand up and maybe maybe you want to come in on this yeah thank you Mr iy thank you commissioner plant I say actually had raised my hand because I wanted to offer a little commentary as well commission plant on the L Flex question but we'll go back to that that's okay um so so yes we we certainly do see some advantages when you think about modeling this um and you know a high outgrowth of data centers um on our system I think we made our data inter assumptions very clear in our jts filing and so you do see around kind of like what we've have referred to as The Fringe of the Denver metro area we do see you know like a lower system cost kind of occur out there in the sense of transmission um investment impact um so that's a scenario that we put out there um for consideration we've also Al had many conversations with data center companies because there's more to the equation than just balancing infrastructure costs they need access to Talent they need access to other resources such as water um you know certain raw materials right they have to run their business and so there's not a utility only construct to balance here um so we've sat some feedback on that um kind of emergence of this you know I think we've Loosely taught it data in a row um in that kind of Eastern Aurora area um is where we see you know one of the better opportunities uh for our system second to that I'd say it's it's something that we continue to to evaluate um bulk Power Systems are dynamic they change um with the level of resources on the system and the type of resources um bulk power flows based on both demand as well as where generation wants to push uh the electrons as it were and so that's something that we expect that will continue to evaluate on an ongoing basis but for now it's that Eastern metro area um which seems like it's the best Target opportunity okay and I think you wanted to address I appreciate that you wanted to address the the flex question too yeah so that's another topic area I've had a lot of conversations personally with data center companies and their executive leadership it's question I asked with frequency I agree with Mr Bailey's um description of it uh many of them are willing to entertain the idea but they've said you know they've said at the end of the day they're going to run their Data Center and they made you need to contemplate diesel or natural gas fuel resources to do so um I'm aware of a couple data center um companies and Technology outfits that are contemplating novel load FX excuse me load Flex technologies that in their words um can kind of work seamlessly across a suite of Energy Solutions it's not my place to divulge there um you know competitive advantage and how they're contemplating that however it's a couple that I'm aware of um that said I'm also aware of several industry research initiatives um one that would possibly occur here in Colorado to start testing some of these Flex Concepts um so I just wanted to State pretty transparently from what we've heard from the data center companies it's not yet fully commercially mature and it's not yet part of the business strategy of the majority of these data center companies okay thanks and are there any um assumptions around self-generation that are associated with any of the potential uh data center projects as in like how we model them is that what you're getting at yeah yeah yeah so we have had some data center companies that have requested both Electric and Gas Service um we have not done sensitivity modeling looking at like the play of you know trying to retrieve some sort of Arbitrage value from mixing between the two ultimately I mean that's that's going to be limited by local Air permitting um and that's something that you know we we would need more information from the data center companies and we' need we need to have a different kind of conversation with them and I'm like what that would look like I've not heard though much interest in that type of of energy arrangement I think Mr Bailey characterized that very very well in the sense that you know they don't view that as their Hort business so the so any any self-generation discussion has been around basically gas uh turbines and on-site gas generation is that what you're uh Mr Bailey if you want to chime in you're feel feel free I'm not aware of Technology specifics on reips versus turbin so go forward if you have the answer to the commissioner's question sure thank you um commissioner plan I think it's it's a multiple factors what what the customers are looking for are the most economical Solutions and in our opinion and Steve Mr Mart's um characterize it correctly it's not as advancing as quickly as they would like to see um they would love to see reduced cost and battery storage I think in some areas of the country where there's access to uh natural gas Solutions you've seen some P hopefully you've all seen some public announcements about utility and and customer Partnerships for Combined Cycles um CTS that are are you know in front of the meter but are also utilized as a system asset to help and support reliability and so those are the areas that I think could Advance more quickly depending on the jurisdiction and the decisions by um the inuse customer okay thanks um dang it I had one more question but I can't remember um I'll I'll leave it there sorry yeah we can go to commissioner Gman come back at the end if you'd like uh commissioner going yeah um just curious um uh based on the the confidence level I I think you all I what is it when it's about 80% confidence that they're going to join the system allocate 100% of the load just a portion thereof below that 80% is that accurate yeah that's that's accurate and those confidence levels are really more categories so anything that's kind of tabed with a greater than 80% is highly likely and then 50 to 79 is likely and those are the loads or portion of loads that we're including in the forecast okay um and has has there been I guess has anyone else contributed to that kind of Matrix in terms of the evaluation of the likelihood or confidence of them joining the system or has that purely been an internal process of Public Service so I'll let Mr Bailey speak to the assignment of the probabilities so thank you uh commissioner Gman for the question uh My Team Works across um jurisdictions but specifically for psco you know we work with the director with the customers receive the data go through the process that is outlined in my testimony and and really trying to be deliberate about the the loads that we forecast and we want to be um certainly direct about this is going to be real or we feel like we will get to a Finish Line with an Esa and so we have those customer interviews then we work with Mr Good enough's Team U to provide that data and then internalize uh measure that data and forecast it and so it is to to your question it is an internal work product but we also are doing multiple interviews with customers and ensuring that we're getting the data that is needed um to give Mr good the right information okay and I'm curious um uh I think this will U be kind of across each of the things the the large loads and potentially even be and EVS I'm curious to what degree you all factor in kind of the uncertainty around economic conditions I mean obviously I think things have may look different now than they did when you developed your forecast um way back when the slide showed you did but um you know there's clearly increasing uh concerns and discussion about a potential uh recession coming don't know if if that's what will play out or not but I'm curious to what degree do you understand like the sensitivity of economic conditions and all of these forecasts and maybe start specifically with large loads which might be easier to pin down you know what that impact might be and then on the development of BV and be and E which are driving um this forecast so just curious um how you consider that or how it could be considered um yeah Mr barley would you like to take the economics of the large loads yes I I think thank you um commissioner Gilman I I I won't be predictive here but I I certainly share your your in respect to your question you know the one thing I will point to is we are consistently um meeting with these customers I think since my um my filing my testimony I think it represents I think we had nine customers who are going through the the sis study um since that time we've we've taken on an additional 10 who have paid us and we're going through that process so um we believe and we watch closely that the advancement of AI and the need for machine learning machine learning is only going to increase now efficiencies will happen um I can't predict the econom omics of you know a potential recession or otherwise but from what we're receiving in from customers and as we speak to them they are not seeing a Slowdown in the needs for um advancement in their work and advancement for New Generation and transmission service so moving on to to EVS if if if we can um so the the way the EV forecast is modeled It's a combination of the bass diffusion kind of techn Technology adoption s-curve modeling and a payback model um so and then it's an average of the two that's taken as the forecast so I mean certainly in that economic payback model there are you know assumed decreases in battery prices there's assumptions about gas prices assumptions about incentives um to adopt EVS so the the adoption rate will be very sensitive to those assumptions so if those were to change dramatically from two years ago almost two years ago when the forecast was creaded that would that would result in a lower forecast um of course there could be you know countering impacts at the state level right that off offsets some of the federal federal changes potentially um and then the bast diffusion model you know we've seen accelerating adoption so that's what that model continues to give us if that does slow down we'll pick up that slow down in that modeling and that'll result in a lower forecast as well but certainly some risk um to the EV forecast given you know what we've been hearing in the news lately yeah thanks um for addressing that um curious with regard to you know commissioner plant had a lot of questions on backup um Power demand flexibility you know the extent to which backup power can be used for demand flexibility um I share a lot of those questions but I don't need to to rehash all of them but I'm curious in terms of looking at serving these large loads um from what would be needed you know locally on the ground to just making sure we have the generation in and time to serve them is it possible that there is a speed Advantage like as I understand in some of these projects um they're looking for quick so is there a speed Advantage potentially to having you know full onsite backup or self-generation capabilities in terms of being able to develop before say the Generation Um you know which we're facing some difficulties in Contracting right now um could come online you know could that be an incentive to ensure that customers do have those sorts of capabilities which also you know could in turn um lessen their impact on the peing of the grid yes thank you commissioner Gilman for the question um in full agreement and please know we're we're obviously Advocates of having the right load Flex Dr that's that's part of the system that that helps and protects all customers to your to your question you know directly as we work with these hyperscalers and these customers they do want to move as quickly as possible but they also want to make sure it's safe and reliable and it's it's not impactful to a region there are some test cases I'm not as familiar in the US but I think outside globally where these customers are doing on-site gener ation and effectively islanding in some areas now there are there are nuances to that in the US related to permitting an access to to potentially natural gas pipelines for base load generation but also it takes an immense amount of land if you think about if you want to do a micro grid as an example to put um you know battery storage if you wanted to do you know a couple hundred acres or a thousand acres of solar so they like the impact of working with the utility and being connected to to a native system um that gives them that security but they are looking and they will continue to look at opportunities for investment if it if it is right for them that's why we believe that as we work with them and work through this process as efficiently as we can be we can meet their needs because of our reliability because of our rates because of our access to um to our progress with carbon free res resources it's a great compliment and so um our goal is to do that but we also in a competitive World got to make sure that we are not providing restrictions that force them to make those decisions if that makes sense yeah um and then just kind of curious maybe from more of a practical standing a practical standpoint given that these forecasts and proceedings take so long to put together let alone to litigate um you know and and the forecasting even now we're looking at is a year old if if not more um H how do you make the actual decisions on the ground if you get a a large facility that wants to come online quicker but you're showing that would trigger some resource adequacy concerns or or something like that and you're forecasting you know how do you decide which loads um get connected just curious Mr iy I I want to make sure I'm not thinking I'm thinking about policy and process here but I don't want to overstep your thought yeah I was I was thinking of like to what degree is it a commercial type of decision I mean we we have to ensure the reliability of the system for the customers that we do have and then it's a question of when can you add new loads and to some extent that's sort of a question of the queue in which that they come to us I mean and that that speaks to sort of the load interconnection agreements as well which is its own process so we would I mean I think we all deeply respect the regulatory process but we also recognize the time that it does take and we're encountering this this era in which the pace of the large low is what they prefer to get and then you know would argue that they need to get is the slope sort of exceeds our ability to process this both to build it and then the Pro process it regulatorily as well so uh it's not an easy question to answer and I don't know Tom if you had more to say on the commercial front yeah and I yes thank you um Mr Ry I I think also it's the balance we have to really use our com our commercial efforts to understand when we commit to an Esa when we get to that final stage gate of our process outlined in my testimony that we are confident that we can get to that inuse asset for them and obviously it takes takes this entire group and the commission um to get that done and so we do have to be diligent about that and ensure that we're we're walking those projects appropriately and I think um as you've seen we've we've worked to forecast appropriately but also maybe on the conservative side of the amount of projects we have because we do have great demand and so we W to we want to be in the right time frame for those customers to meet their needs and we have to be upfront with them and communicate properly that this is the process and certainly each jurisdiction has diversity in that but we will we will work through that it's a hard question I mean it's a great question we grapple with certainly every day Mr Mar I see you have your hand up yeah may I chime in Comm M Gman sure um yeah great question I think it's one of the hardest ones the utility industry is trying to answer over the next five to 10 years and more um you know I think about this in within the framework of the order of operations there's you know I'd say i' put it like this the last 18 months to two years of how you know the data center excuse me data center industry has really grown up and expanded Beyond kind of the footprint of Virginia you know has has impacted many utilities many commissions we've had to learn from one another very quickly and these data center companies which are normally used to dealing with you know maybe one to two five maybe at most regulatory constructs are now trying to figure out you know more of a national and even Global footprint trying to identify you know how do they how do they create the next business opportunity and I think speed to Market is certainly an aspect of that our our reaction to that is try to be transparent with them about how our commissions work um you know we work with multiple commissions obviously being a multi-state jurisdictional utility and there are differences and so we try our best to educate them on those uh differences what I think is unique about Colorado and you again you going back to this going back to this order of operations concept we do have to maintain a regulated Open Access transmission tariff as well most of these are Transmission Service customers and so there's some interplay there in terms of how you manage that oat as we call it open access transmission tariff Q um and their their request for transmission service and how that works with State regulated constructs um and so again our our what we've done is try to be transparent around that we've worked with data center companies directly as well as their industry trade groups to inform them um of what the construction timelines are for transmission as well as generation as well as our regulatory Pathways and right now unfortunately that's a somewhat linear sequence and so uh we're actively challenging ourselves to try to find ways to streamline that but it's it's something that we need to sort out I think the other fear that I have as well is like as we as we're thinking about you know the total Energy Service to a data center there's a generation piece and a transmission piece in general and if we start to scrub out data center L from one you know from one of those elements it's going to impact the other because if we start to contemplate lower load forecasts and we we kind of depress the jts forecast when we come back and we look at our transmission plan we're going to base it off the same forecast and so we're not going to holistically look at how data center um load can impact both G and T so that's what I mean when I say there's a significant order of operations here that we have to sort out thanks I appreciate that um back to you chairman uh thank you um it's uh understanding that Colorado may not be uh an attri uh particularly attractive State at least for the largest new data centers because we don't have a manufacturing exemption uh to sales and use tax that uh 30 other states have so just uh does the company share this concern how are you addressing it and does it give you pause uh about or should it give us pause about the forecast uh thank you chairman blank and and appreciate the question and um recognize that those incentives in other states and I've actually in my past life testified in favor of those in certain States related to large manufacturing um I don't my response would be I don't think you all and we should take a pause based upon the fact that that there's not that economic incentive currently uh understanding it's gone through the legislative process the last couple years and and and and it's there now um certainly we will support the general assembly's decision and the governor's decision to sign that um if it if it is voted on and approved where we really look at the advantages of Colorado um we are seeing great demand based upon um the urban Metro Center the access to Western markets and this is this is feedback from customers the access to Western markets an underserved area Denver as kind of a launching point when we go when we go west great fiber great access to land and Workforce and so those are complimentary um when we've spoken to these customers obviously we've got approximately 20 in my tlb1 um they're coming here and asking for this service without the guarantee of those incentives and so uh we will watch it carefully we will certainly you know support the the actions of uh any decision made at the legislature I will say as I've watched this in other jurisdictions in the US um and a passive that type of incentive will increase demand in my expectation and so we will have to manage that appropriately uh as we've done as as we've done now and so I think it's very important to kind of watch the macro side of the economic decisions but I do not take pause right now that Colorado is at that disadvantage um I just say appreciate the filing of the lower load forecast at least it gives us a a record to look at you know all the options uh uh and have the parties have a full dialogue so appreciate that uh you mentioned that you're seeing a number of data centers located in Aurora and elsewhere in the Denver metro area if we uh site these data centers in Denver which is as I understand it a transmission constrained region won't it greatly uh potentially greatly increase the new uh transmission required to serve that load um so just factually is it fair to say Aurora is in the transmission constrained region I think it is because it's on the 230 KV not the 345 uh KV system but um won't sighting all the data centers there greatly increase our need for new transmission investment yeah I'll chime in commissioner sorry chairman um so partially accurate but not fully accurate um that's not exactly off of that system it strides it straddles both systems the 230 and the 345 and in our last uh system cluster study that was looked at as a result of this jts looking at data center row um um we don't see it as impacting the Denver metro area as far as data CER load that's further within that metro area um that would further require transmission investment we have not seen that so I don't I don't believe we stated that um so it is on The Fringe part of it touches the 230 system but it mostly impacts the 345 system which is outside of that so in the jts you're showing something like me what is a very large number 38 billion dollar of uh new transmission over the next uh 20 years uh uh if the data centers aren't contributing uh to that U new Trans I guess I'll just ask it as a question are the data centers contributing to that new need or uh you know what's driving it just the same be and EV uh load shapes we were looking at in the the DSP well I can't speak that figure specifically I don't have it in front of me however I the work that we did on the transmission as part of jts was not an exercise in guessing demand and then trying to figure out how transmission meets it it's as much about servicing and providing a flow path for the new and incremental generation onto the system as well as alleviating excuse me alleviating system constraints and I'd say this and if Mr L would like to come on and comments I would welcome him to do that but not all generation is the same it also has its own Generation profile and so we what we did in jts was look for the specific stresses on the system to contemplate a number of different scenarios one of which was a 900 megawatt VPP scenario to see what kind of interplay happens on the transmission system so I I don't agree with the generality that data centers are the sole driver of transmission investment there depending on the scenario there's somewhere between 6,000 and 14,000 megawatts of renewable additions that are cing the jts and there's multiple drivers for transmission investment as part of that jts analysis you have uh estimates of the transmission Capital spend uh if we assume that say 80% of the new data center load was located uh you know in the Southeastern portion of the state where uh the renewable energy resources are you know perhaps uh if there was a a a a Sal and use tax exemption we could uh put the new load where most benefited the the system uh I you guys look I can't speak to that sensitivity sh we did not run that I think we'd have to go talk to the data center companies like I said a little bit ago mean they have to balance business model as well which includes having Staffing for their data centers having other access to resources and I've not yet seen a request from a data center company looking to locate there so we could contemplate a a sensitivity there uh I would caution we need to talk about that with with our customers because again that's not part of what they've expressed interest in I mean maybe but if it was more valuable to the system and it was linked to a sales and use tax exemption um um I don't disagree as a theoretical but again that's a that's a business model they need to contemplate they need to go talk to those municipalities talk about how they get Talent there to run run those data centers right now I can tell you I think it's it's possibly a theoretical exercise I can't tell you that I have faith that that would produce a tangible outcome fair enough um the next uh Mr iy I was just I was just gonna add onto your question concerning the 38 billion um there is some degree of transmission investment I I know exactly where that estimate came out of in our modeling there there's some degree of transmission investment that may be covered by the data centers and we frankly need to dig into our assumptions to see what that the 38 is covered perhaps not by the Danga centers but to some degree there's additional amounts that are um and I think we could probably clean that up and and kind of confirm that a little bit you know with followup information to the commission yeah that'd be uh that'd be uh very helpful as the the the dialogue uh continues in both these cases but but agree with Mr Marts that it's not the only driver the large loads are not the only driver in that differential in fact it may a somewhat muted driver given that I think there is some large load contribution that's not on on the books that you're seeing here okay Mr Mars yeah Mr chairman if I can so I tried to raise my hand with my zoom app but it says it's not working I apologize for that just another thing this works thank you um I agree agree with M Ry opportunity for us to clarify assumptions that just another thing I note though as well is with transmission level service we do assign those costs for up to data Center so when we do our first our system impact study and then our facility service we call FSA facility service agreement um you know we would look at our tariff and our tariff would generally assign those those direct transmission service costs to the customer so agree with Mr Riley we'll go back and clarify that yeah that'd be helpful it seems like uh at least in the first uh EDR rate case we saw uh maybe we didn't quite collectively get that uh cost all uh as right as we might have um uh commissioner plant you remembered your question my last question yeah um thank you um so one one uh question that I had is I think you know as you were talking about it's been a while since you developed these uh projections of loads particularly around data centers and since then you know nvidia's up the efficiency of their trip at least two times and then of course there was the deep sea uh model which is a much higher efficiency um I'm wondering if if you have incorporated or if you thought about how you would incorporate any of these technological advances and efficiency advances in some of your load forecasts uh thank you um commissioner plant I I'll speak maybe a little bit more to your question on the macro I I felt like you're kind of going into maybe how are the customers viewing the efficiencies that are gaining and then maybe ask Mr good enough to talk about the forecasting um we had once that that deep seek announcement came out probably a month ago um we went out and interviewed you know our major customers so think about the top five to 10 customers that are looking across globally to expand but in the US with the largest expansion um they're not pulling back because they know you know kind of efficiency will be will help this business and they so they expect those efficiencies to happen they just believe that they'll be able to utilize um that same amount of energy in a smaller space and so they still need that demand but they can do more with it they also are very cautious about uh efficiencies that um can't U be produced or confirmed and so while they respect the what deep seek has announced they want to make sure that it is accurate in in in what we're doing so um I don't believe and and Mr good enough that we have taken into account that type of of um of change but we've just been conservative in what we put in the forecast yeah agreed given the kind of the commentary that Mr Bailey Just provided on the discussions with the customers that they have not indicated that they now require less power so we have not kind of folded that into the forecast at this point so so even with the improvements in the efficiency of uh the Nvidia chips and things like that there's there you haven't seen any uh reduction you're what you're saying is they would just make their data center bigger as opposed to uh reducing the amount of load that would be required of the system they would you yet yes commissioner plan they would actually try to use the same per square footage and and factor in more chips and more Computing but expanding the energy consumption just in a more efficient manner okay thank you and that's I appreciate the question we're getting that throughout our jurisdictions about how to manage that and so we're we're also learning too I mean it is a major paradigm shift for all utilities and so um we're watching we're learning we're gaining access to our counterparts and so um it's it's an exciting time but certainly something that we have to to manage appropriately well given that we're now four hours in the five hours and we're about halfway through why don't we take a little bit of a break to 405 and then uh I'm going to come back to you commissioner Gan and why don't you just ask whatever question is uh most relevant most important to you and we'll just go around one by one and just uh try and get our questions uh answered uh as best we can or at least say asked uh does that make sense uh to you give more we're at commissioner G yeah that works what time are we coming back 405 just five minutes okay does that work for you mrly yep yeah we want to try to help you the time we've got here yeah thanks all right let's take a break till 405 and we'll come back starting with you commissioner G thank you e that was really St [Music] me [Music] God e e e e e e e e commissioner Gan yeah I'll just um maybe ask a few and then go to you and I'll see if uh I have any more when you're complete um Mr Riley I have a few questions kind of for Mr Mino Mino I think most particularly um some of the informations contained within confidential attachments this testimony but I'm fairly certain I can ask the questions broadly enough that we don't get into trouble here um Mr Moo um as I go to the um confidential attachment uh DCM 2C which is the grid needs assessment which we won't pull up because I I don't think we need to get into confidential session to just talk about my general question here um the um number of new Banks and feeders and subs in your testimony here I can't quite reconcile with the number of projects that I'm seeing identified in DCM 2C and I was just wondering if there might be a a reason for that um are you talking future substations or existing substations um well the the projects identified um here within the grid needs assessment uh tab of the grd CCM to see so um guess I'm not sure what discrepancy is um on page uh 88 of your testimony you go through the number of um new feeders Banks and substations you expect in each of the 13 planning uh districts and it wasn't exactly reconciling the same number of projects from the attachment um I'd have to double check that I'm not quite sure what the discrepancy is um without a little more specifics but I don't know if we can jump into the specifics um might need to take a look at that and and get back to you okay yeah if you if you could just take a look uh and maybe uh file it with a supplemental direct that'd be great we'll do um another question um within um DCM 1C which is another attachment to your testimony um I'm curious for um substations that have only one bank and that that's a thing right there are some substations that have only one Bank yep um also within that attachment I was seeing some discrepancy between the bank loads and substation loads which I would assume if a substation only has one Bank the bank load should equal the substation load but let me know if I'm mistaken on that that should be the case um okay so if you could take a look at that also in attachment um BCM 1C to see why um it's not on every single one but it is prevalent that the um well actually maybe so you're saying the bank load is different than the substation load or the feeder loads are different than the yeah the bank load is different than the substation load but they are single Bank substation sometimes we serve non-distribution customers or non Excel distribution customers from those substations so there there could be some of that being factored in but uh without knowing the specifics I can't quite speak through I'll take a look at that and and respond in my supplemental okay okay I appreciate that um and then I just wanted to get into without getting into like sub you know IDs of the substations or Feeders um within the Mountain Energy project filing the company made um it identified about $28 Million worth of um potential distribution system upgrades needed to be made uh I think primarily in Sumit County um in order to accommodate those and they're broken on I think 8 million of out was for um some different uh feeder upgrades and then about 20 million for substation upgrade in Breen Ridge does that sound familiar it does yep I'm familiar with the upgrades for that project with for that effort yep okay so as little un clear you know in the testimony in that proceeding it seems as though um that those costs are kind of being allocated to that project um it says that the feeder projects in left Ville and Dylan were are expected to be introduced in a future 5-year plan but the need was already identified and then it says that the bra substation upgrades were required solely because of the NPA in that uh portfolio um but when I look up the actual substation ID it seems as though it is given a project ID um and so that to me might indicate that it is a project that's already planned in this DSP so I don't know if you can clarify that for me as well um i' have to take a look and see which ones are actually in the the DSP here um I know some some projects we do identify that don't quite get funded because of the level of risk that is there um and the timing of that but um I'd have to take a look and and see which specific project projects we've called for in here relative to uh to the Mountain Energy project okay yeah it would be helpful if you could make that link because I was not immediately understanding I kind of understood from the Mountain Energy filing that that substation was not intended to be worked on or funded um but it is not listed as it is not listed as not funded it's listed as having some I will say some of the projects represented the gitty's assessment are ones that have already in place like in 2024 end 2024 early 2025 so there may be some of that being factored in here but I'd have to take a look at the specific ones and yes so fair fair fair question I'd say let's take that back and reconcile I get what you're after just not know would just note that Mr Mino is not a witness in the Mountain Energy case so he's kind of on the spot not totally familiar with that so just would respectfully press give us a chance we'll go back I get where you're at reconcile that okay that would be helpful thank you for that um let me just go back to my notes for this one real quick um let's see I I think those are my questions for now chairman if you all want to move on and I'll let you know if I have anything additional uh do you want us uh so uh we I could like I have two or three questions left do you want me to just S one and rotate back or does that uh uh are you not looking for us to rotate back I'm fine for you to go go for all of them and if I I'll I'll raise my hand if I have an emergency question uh commissioner yeah I I just had a couple uh quick followups on one on one of the questions that commissioner Gilman was asking so if I might just throw those in and then I'm and then I'm good but um uh not to bring up the actual numbers associated with any of the confidential uh substations but Mr Mino if can you tell me if for example the tarmigan substation is one that serves other distribution customers that would fall into that uh that situation where we might see a discrepancy between um between the numbers it's a single um a single Bank substation um where this the substation load and the bank load should be the same but they they were different I was just wondering is that fit into that um category that you were talking about that maybe it serves a different distribution utility or something if that's what I understood correctly not that I'm aware of okay um can you uh just describe what the difference if is when uh you're talking about a bank and when you're talking about a Transformer there those names are synonymous with each other they're they're the same thing substation Transformer substation Bank we use those names interchangeably for the same asset all right great thank you I no further questions Mr chairman thank you commissioner plant uh just two last questions uh I'm concerned that there may be Supply constraints for critical uh equipment including for new CTS and maybe large high voltage transmission Transformers such that it may be difficult to coste effectively obtain delivery before 2029 uh given the expected time frame when phase two of the jts proceeding is likely to conclude as you know also maybe issues surrounding the approved 2021 Erp resources can I just ask if the company has uh combustion turbines already reserved under contract are you guys in the queue or uh or not Mr Marks you want to take this or shall I yeah happy to chime in thank you for the question chairman uh the the sample answer is yes um for not to go into more detail just given you know there's competitive nature to that um on the supply chain given what we've seen with recent world events but sure our answer is yes and uh uh uh and is it a similar answer for the high voltage uh transmission Transformers associated with uh both the 2021 Erp and and I guess the the jts to given the long lead times uh uh are you uh reserving equipment for uh uh those potential needs that's that's correct chairman uh very s very similar um case there I I'd also go on to say as well you know obviously being a multi-state um utility you know we're looking at the total pipeline of system needs and when we're kind of looking on a holistic basis looking at the supply chain capacity is that's out there turbance and Transformers looking at the total need set and then you know deploying a supply chain strategy that that services that um uh Mr iy yeah I was just going to add maybe just a little bit more than than Steve offered on we have the turbin reserved for our clean energy plan assets that we're planning to bring um psco doesn't have turbin reserved for jts at this point um we do have tools that we have proposed in in the jts uh specifically the pre-construction development assets which which evolved out of the last Erp and you had you had a hand in in um you know creating that concept so we have that the the the pro the challenge with that is that we can't really use it until the phase two is approved um the way it's structured so I would say we're open to thinking about tools that would allow some degree of ability to for the company to use its Financial you know its business model in the way that it does and create some reservation slots not not just for gal CTS but maybe probably for other equipment as well and sometimes frankly like ENC firms are the are the supply chain constraint not just equipment it's it's skilled skilled labor as well so we're actively thinking about that let me put it that way without maybe offering a super specific thought well I appreciate that and just say uh not trying to prejudge or pre- litigate the uh uh proposals in the jts but given the concerns about timing and rigidity that you dis that you know sort of you discussed with commissioner Gman the Erp process the more optionality we can create um I think we'd like to find ways to work with you to to to do that uh just the timing is uh just uh not consistent with what we're seeing in the market so um Mr Mar and a comment thank you propose chairman I I had a followup for you on actually for chairman blank's question on uh tarmigan sub so maybe let's finish up this thread and if it's okay we just ask that we go back to that because I think we can provide an answer there all right uh did you have more I mean that's all I had I'm just pleased uh you're preserving equipment and I don't have more on that uh unless you you do uh yeah why don't you go back and uh answer uh commissioner plant's question on tarm again yeah Mr Moo thank you chairman yeah I thank you I I just looked at my um confidential attachment tcm1 and the bank forecast and the substation forecast are indeed the same I'm wondering perhaps did you sum up the feeders on that substation because if you sum up the individual feeders and compare that to the substation forecast those numbers will be different due to The Coincidence of the loading on those assets so not sure if uh maybe you can just PR me a little more context there so I can I'm sure I'm I'm getting you the right information yeah that that may have been it um commissioner Gilman did what are your thoughts on that I don't have any additional deta right now I did have more questions though uh let me ask my last one and then I'll turn it back to you unless it was on that specific uh equipment ordering no um uh more question for you guys uh we compared uh this is on the rate impact stuff we compared total sales from the 8760 matrices as against the long-term rate forecast model uh sales in each available year uh we found that the sales in the long-term rate forecast model are as much as 10% 10 to 15% higher in each year if you could just help us understand the differences between the two forecasts that'd be great it looks like uh behind the meter solar is only about 5% of sales by 2045 and doesn't account for it so if you could help us uh with that that'd be great uh commissioner Gan thank you um uh I have a question I I don't see probably a Mr Mino question I don't see um cost or really any discussion on um distribution Transformers and I was just curious like if we're seeing this much growth this much um expansion of the system if there are costs there that appear elsewhere if they're folded in with something else just help me understand if they extra costs there we're not seeing specifically identified um we do have the cost accounted for in our five-year budget also it calls it out in the the overall project cost and the grid needs assessment um so we we are reflecting the cost of the substation Transformers when we have a single project for like a substation Transformer there's cost associated with the actual Transformer and bus work there's cost for the the communications that we bring into the substation for the skus system and then there's cost for the distribution feeders so there's typically three line items for a substation project um so maybe this just not clear in terms of how the naming conventions are are broken out but uh they are reflected in the The Five-Year budget that be filed okay um and I just wanted to understand a little better to the updates to load here that have been made since the the original forecasting and filing what's the nature of those I mean it sounded like you said new customer applications will get entered and that's something that gets manually entered yep so our my planners on my team enter those new Kachi checks that they have reviewed and approved into loads here we've actually have already produced our 2025 forecast so we're on a whole new forecasting cycle compared to what was already produced in and load Seer for for the DSP so um those are some of the examples of of changes other there's other ones in terms of just other assumptions that we've baked into it but um yeah that that's those are some high level examples can you give me other examples I'm just trying to really get a full view of what all has really been updated we have discreet known new customer loads have gotten added manually what else would be different any new distribution feeders that we've placed in the service since the last year those will be reflected the loading pattern on the assets that were offloaded by the new infrastructure those teal those Cur 8760 curves will have been updated to reflect the new loading pattern on each of those assets um we've introduced additional feeders that have been placing the service since then so um the topology of the system has changed since the 2024 distribution forecast from that perspective as well how many are we talking about how many new distribution system veters they went into service in 2024 I don't know the exact number off the top of my head um it was pretty substantial we ins serviced a couple new substations um at least a couple of new substations at least three of that I can think of off the top of my head but um please don't hold me to that so and then all the distribution feeders serve from that so we're talking a decent amount um in terms of how the the system topology has changed since the 2024 forecast okay and there's what a handful of feeders per substation uh depends on the design of the substation um and how many we bring out day one um typically we'll at least bring out two feeders um some of the examples that I could think of I think we've extended three so um there's plus there feeders from existing substations as well so I don't I don't have a clear account on how many we placed in the service but it was um yeah I'm I'm not sure exactly how many okay so then those um aspects of the system would have been manually adjusted or entered to recognize the existence of the new substations and feeders which then impacts kind of adjacent or places where power flow change as a result of these new additions correct yep the new assets are taking on load that has served other parts of the distribution system so now their loading has been reduced um to reflect that just based on how they're they're now operating so um those are some major changes in terms of how that that's changed okay but the the assets that would see less load um probably wouldn't have been identified as upcoming projects anyway cuz you kind of knew there was some mitigation that was going to reduce their load right identified where in terms of in like for projects in this CSP um correct I mean if they were placed into service there may have been some lingering cost for those projects in 2025 if they were plac in our goal is to get projects in service before the summer so some of them may be implemented now um that we know about um also yeah for anything that went into service in 2024 they would not be reflected in this this DSP right but but you knew you were putting them in so whatever other equipment they took load off of has probably not been identified as a near-term project I would suspect right um it guess you may me ask that question again sorry yeah I mean it seems like the biggest impact of this new substation feeder situation would be like on the adjacent or related equipment that got load removed from it and I guess knowing that you were concurrently installing this other equipment that would reduce load on that equipment I don't suspect the things that have seen the biggest load changes due to the new substations of feeders would have appeared anyway as upcoming projects because you were literally doing a project to remove load from them during that process right I think that's fair I mean it depends on how the the forecast was looking on those new assets or those relieved assets if if New Growth pushed them Beyond as their their operating limits um it could have potentially triggered a downstream project as well but I I don't believe I don't I can't recall any top of my head that would have fallen under that kind of category okay um another question um in the the previous DSP the company submitted a forecast and the DSP rules require some of that planning at least to plan around a forecast and meet State policy objectives and other aspects um and um through a workshop in the be barriers proceeding the company had stated that they did not actually use the forecast from the DSP proceeding for their actual planning and I was a bit surprised by that so I was just curious on this DSP I thought I would ask is the forecast you've submitted here the forecast you actually plan to use to plan the system and the construction projects yeah I think I would just clarify the original portion uh the scenario forecast that we produced through ICF was the consultant that helped us in the 2022 DSP we did not leverage those scenario forecast to plan the distribution system but the the forecast that we filed in the DSP was what we used to plan the system um here yes the we we produced two scenario forecast this year uh for this DSP our capacity check forecast which just contains our known load growth on the distribution system see capacity that we've already allocated to customers through that service application process um we've provided that one and then we've also provided our base forecast which accounts for the future growth that does largely align with the state goals and and policy goals so that forecast That Base forecast that contains all those other growth factors is what we are plan the distribution system to okay U Mr PA I'll just add a little bit of additional color there to what Mr Mino said in addition to the state policy forecast that we were planning our system to that was modeled in load Seer we also ran four additional scenario um forecast with cavala one of those was a calibration using very similar input assumptions that we Ed for load here the other three of those was a another updated State policy given vintaging a goal scenario that exceeds the policy and then what we call the adoption form trajectory which was based more and more heavily influenced by historical Trends and I just want to reiterate that the sort of key takeaway from that process is that if you look at the the next five years across all those scenarios both modeled by the company and load here and cavala it requires similar level of infrastructure investment than um that we see across really all forecasts it's really the tales of those forecasts where you see pretty significant Divergence okay thank you um one another question about the use of load here so um you explain that the system risk analysis the n0o overload and N1 contingency test n minus one I guess you can't see minus zero so I don't nus z n minus one is y n minus Z is still just 10 um and the mitigation selection process are not conducted in loads here but by distribution uh planning Engineers who have to account for specific factors including the nature and scope of the overloads distribution system topology and potential options and locations for new infrastructure to alleviate the risk so I'm just curious um it sounds like the N minus1 overload and the N minus one contingency tests are conducted outside of load Z am I understanding that correctly that's correct yeah if I I briefly t on that on slide 29 of our presentation where low Seer is really just our forecasting tool it it it's not like in compass where it it resolves all the system needs through on the transmission and generation uh system is it incapable of doing that or do you all choose to not use that feature that it might have I don't believe it's capable of doing that um it's it's really geared towards producing forecasts um scenario forecasts especially on an 8760 basis um really the the quantification of those risks we perform outside of it because um this because the the tool is not really geared to to do it that way okay um and when we look at dcm1 in the DSP proceeding the attachment to your um testimony and we see the projected loads for each piece of equipment um do those represent the max 8760 that are the combination of the actual skated out and the foric acid um that that would represent the maximum at any point in the year for where those two things overlay correct the attachment dcm1 contains the the peak essentially the peak value of the 8760 for each year for each asset layering on the base load which is the existing SK or the existing loading on the system that we've obtained through our skada system it it factors in the capacity check growth as well as the allocation of those 25 different growth factors outlined in in my uh DCM 9 attachment the technical appendix um behind that Peak value though was an 8760 curve for each hour for each uh year for each asset okay thank you uh appreciate that uh I think that's all I got sh well we had talked about throwing a a showing of loads here uh I think uh you got three exhausted Commissioners and uh probably your team is pretty beak too uh can you do do something for us uh quick uh what's your thinking Mr iy Mr lson uh we're we're here if it if it works for you um I think I don't know David Mr Wheeler uh is it how long is that demo take I'm trying to remember how long that we can we can be very efficient with our time I think the length is really depend on the amount of questions we get during it but I think we can keep it between five and 10 minutes to highlight the the critical functionality of the Tool uh and just illustrate how we use it to generate our forecasts uh commissioner Gman commissioner plant would you be all right if we uh yep uh all right thanks uh thanks uh Mr Mino great so right now we have the asual ere tool up right now we we have it in our Dev environment but um nonetheless uh we wanted to start on this page here which is what we call our hierarchy this allows us to put together the the connectivity of the entire distribution system from the operating level um node all the way down to individual feeders and it's this connectivity that allows us to take operating level forecasts and Al at them down to individual premises along the distribution system to create forecasts for distribution feeders distribution uh substation Transformers and substations as well so I'm jwing down through the hierarchy here um here you can see all of our 13 different planning divisions if we drill down further you'll see that um within that planning division these are all the different substations within that um plane division joined down further will will further get us um nuances around um the assets contain within that substation and we've anonymized this particular substation to avoid having to enter in any kind of confidential uh information so within this generic substation there's two substation Transformers um off of this substation Transformer there's four distribution feeders so these dark blue nodes represent the feeder so we have a connectivity model from our operating level company uh operating company level all the way down to individual feeders that allows us to do this uh spatial allocation process which we you know is what we use to generate the forecast um I want to switch over to the uh typical loader curves um so this is ultimately the output of or yeah sorry this is the actual base load that we develop for the distribution system so distribution planners on my team gener or say scrub three years of historic loading data um using our skada system to remove anomalies to clean up the data set and produce what we call a t typical load year which is an 8760 load profile in the distribution system this is essentially the base load that we layer new new growth on top of to develop the forecast so this is using actual skate information uh there's three different low curves here low typical extreme low is the zero Z percentile typical is a 80th percentile and extreme is the 100th percentile most extreme weather over the last 30 years so this is a weather normalized load curve that we use to uh as our base starting point for loads on the distribution system we produced these T curves for all 800 plus distribution feeders and all 275 substation Transformers so there's quite a lot that goes into generating these curves um switch over to our node viewer page here so this is kind of a landing page for any any distribution feeder uh this is actually a real distribution feeder we just anonymized everything here you can see the capacities on the distribution system um as commissioner Gilman mentioned before you know evaluating summer and winter ratings right now they're the same uh something that we are going to evaluate down the road to see if that makes sense as we look at how you know the ambient temperature effects Underground systems as well as all all of our overhead equipment but nonetheless we've we keep these the same here so our normal capacity here is that 75% planed load limit the emergency rating is that 100% load limit um we're going to skip over to the map window here so if we click on this map and uh drill down um this these this is showing premise level information um these homes are not on that feeder we selected some random homes so we could further anonymize this but this is an actual home on our system uh if you drill down into the individual premise you'll see quite a bit of information we call these Services here so that Top Line is the existing demand for that individual premise um again this has been anonymized to uh protect the sensitivity of this particular home's uh usage um as well as all these other services that you see here so EV residential PV resid IAL behind battery system uh different kinds of space heating electric water heating these are adoption points essentially so as we drill as we allocate these operating level forecasts um down to individual premises they get based that we allocate these forecasts based on the individual technology types you see here and services the the individual premises that have these Services assigned to them are able to adopt those Technologies you'll see that there's a per unit value so this is the value that those those um or per per device uh value so this is what we scale those 8760 per unit load shapes to um we can drill in and look at the actual load curve that's assigned to each of these here so if this premise were to receive a dual fuel heat pump it would take this load shape this is the this is the 8760 that was in the tech DCM 9 techn techical appendix and it would scale it to that per unit value on the line um in the services P page on the previous um item previous page to show how much that load growth will increase at this specific premise um behind the scenes and we can't quite illustrate that but there's a propensity to adopt these Technologies behind the scene which is essentially a percentage likelihood of adoption that helps inform that allocation process premise that are more likely to adopt or have a higher chance of receiving a specific technology during that that allocation process of different technology types um see if there's anything else I want to hit on on this part here um so we have like I mentioned before there's 25 different technology curves that we allocate um these are a lot of the residential ones here there's different ones that are spec specific to commercial workplace charging um um different different uh customer types and different technology types um so as we allocate these these forecasts down to the individual premise we then aggregate these up to the distribution feeder level so um we layer the 8760 curves for these different technology types that these premises receive that are connected to a given feeder and then we layer that onto that typical load year curve that we we we showed earlier um and this is all this is all enabled through that hierarchy tab that we we first looked at um the output is um what we're seeing here this is what we call our capacity check forecast so this just contains the the growth that we've approved through our capacity check process um you can see there's two lines on the on the y- axis here this is the 100% rating the the bottom one is a 75% rating so you can see that this feeder is above the 7 5% rating below the 100% we do allow the load to grow up to that 100% rating that that 75% is not a ceiling that we stop at but it's really just meant to help us trigger a project and give us enough time to execute on it but you'll see that this forecast is remains flat beyond the the portion where we actually know growth is going to occur um but on on the flip side of that our base forecast which includes those 25 different growth vectors you can see that there's substantial amount of growth that we're forecasting on this on this feeder um based on the spatial allocation of those 25 technology types um and eventually we exceed that 100% rating um so you can see that there's a substantial amount of growth here between the different residential um technology types that this this feeder is the customers on this feeder are planning to adopt or forecast it to adopt so the capacity check forecast is used to evaluate new growth on the system is it's what we call our source of Truth when evaluating new capacity requests this base forecast is what we use to evaluate risk and ultimately drives the need for our the projects that we've outlined in our Five-Year Plan here um trying to be efficient for time here that's a high level nuance and and the critical functionality there's a lot of manual steps to kind of prep this and also to uh to kind of get an actual forecast out of this tool so I think I'll maybe just pause there and see if there's any questions or um anything that you maybe you're curious on based on what we we spoke about yeah can you just spend a minute and talk about how that uh leads into sort of the dollar investment decisions uh and maybe why it's so tough to look at an alternative load shape and get to a different capacity expansion forecast absolutely um if we go back to the actual map showing the entire feeder routing um the way that we use that forecast is we we analyze the risk on this particular feeder based on that base forecast growth um which will materialize anywhere along this blue line which is the actual routing of this feeder um to come up with a project is pretty unique to this area we we would look at are there any is there an ability to extend a new feeder from this substation or any substations we transfer load to an add to an adjacent feeder um I there's a table in my testimony that goes through all the different projects that we identify but you'll see that the topology of this feeder is is unique is specific to this area so when we look at developing new mitigations or new projects it has to solve the need of this particular asset um that's what I was alluding to previously when you look at things on the aggregate you kind of lose the locational specific nature of the distribution system itself um which makes it really challenging to discern those those nuances of it so when we're coming up with a project to Rel this feeder we have to Route either a new feeder or you know add a substation Transformer to an adjacent sub and extended feeder from there the routing is really dictated by where we can actually physically Implement that or route that feeder in rights of ways or obtain easements we have to contend with existing infrastructure that's already routed through these streets so it becomes very specific to the asset that we're trying to resolve here um could couldn't you though just uh uh alter the load shape and reduce the sales forecast and see if it uh delayed the Investments you've already scheduled so instead of uh having to come up with a new topology uh wouldn't an altered load for uh um couldn't you just see more quickly and more easily if you could uh delay the the spending you've already uh put forward um not quite sure I'm I'm following that question I mean we're we're building out the infrastructure based on the the customer and the system need here um really how we mitigate the risk is really dictated by how we can actually out the infrastructure to come to these different load Setters that are at risk to to fully mitigate the project or the the need for the system need um I know but would the system need potentially disappear if you had an altered underlying load shape um I mean if you go if we go back to the base forecast here um the the amount of growth that we're seeing over the 30 years is substantial it may maybe delayed a year or so but I don't think that it's going to totally eliminate the need and we also when we come up with a project this is just one example if you look at the grid needs assessment we can typically Al we can typically try to resolve five to 10 you know risks you know it depends on the the mitigation we're putting forth but it's not it's not just one single asset per one project that we're we're mitigating we we we try to fix as many issues on the distribution system as we can for a for a single project to make them more economical against you know the risk that we're resolving um Mr poock really briefly I think we're close to the end of our time I'll try to keep it brief yeah I think the challenge Mr Min was alluding to is that this forecasting process this is repeated for all 800 feeders and 275 substation Banks so to run that on a systemwide basis and then go through the the technical process of doing powerflow modeling to resolve the risks is really challenging um that I think is what you're seeing as far as our proposal to model representative projects where we could show the impact of those load shapes being different I did want to clarify as well um something where I think we created a little bit of unintentional confusion around before but the the five years of investment proposed in the company's DSP is reflective of needs that are forecasted in a short time Horizon it doesn't include the impact of projects or loads um in the 2030 and Beyond Time Horizon so we could absolutely do this but it's a little bit different than the the Erp modeling process where we're solving to a single objective not only is the number of forecasts that Mr minus team is conducting substantially ERS of magnitude greater but the nature of solving the risk that seems identified goes through a much more technical process um maybe to draw an analogy here that might be helpful there's no generic distribution capacity mitigation as Mr Mino alluded to each one of those mitigations solves a variety of unique and multifaceted risks it's not like we're solving to a single objective I don't want to overfly oversimplify the the generation modeling process um but there's less objectives that we're solving to there uh commissioner Gman commissioner plant uh any final last minute questions I don't have anything else thanks so just to um yeah thanks just to clarify Mr Pollock what you're saying is I mean if we were to look at um addressing capacity constraints on a on a specific component of you know one of the 13 Distribution Systems um that that would not we you wouldn't have a way of of running that with that um with that you know energy storage or whatever resource it is on that distribution system to see how that might impact capacity constraints that we see on the bulk system could you maybe rephrase the question I think I understand what you're getting at but if you don't mind rephrasing it so if I'm trying to figure out if um I mean I get at the the load seere models looking at 13 different Distribution Systems it's you're taking those projections and I think you're putting that into the Encompass model and then the Encompass model is telling you what kind of resources you need in various different places across the system uh in order to meet that load what I'm wondering is if you were to take focused distribution based um uh Investments so for example energy storage or things that might mitigate capacity uh requirements mitigate Peaks things like that on the distribution system that would then reflect in uh a change in uh generation requirements or transmission Investments on the bulk system you wouldn't have a way of running that is that what you're saying in a scenario analysis yeah I might I think jump in here first real quick so just a cave it's 13 planning divisions we have over 800 distribution feeders served by over 275 substation Transformers um but we in each of these assets Peak at different times they're not always peaking coincident with the bulk system so when we're looking at mitigating distribution risk we really have to address the local need some of that does coincide with the bulk system but not always so we just have to be cognizant of that as we're trying to mitigate the local system need here on the distribution system but may uh May point to Mr poock to maybe add on to that I think the challenge is that in order to be confident that it's going to resolve a distribution need it need the asset or Dr that you may be modeling or dispatching needs to be tailored to the specific shape of that very specific asset that could be a feeder it could be a substation bank there's not really a way to model a generic resource on the distribution system given the unique asset speciic specific nature of every load shape and every profile while there's some generality there it's important for resolving distribution needs that you start with the loading on that specific piece of equipment it is possible to provide bulk system benefits to Mr mino's point if the load shape is pretty similar between the bulk system and distribution system but one of the other things we have to be cognizant of as well is making sure that if we're going to use an asset to say provide capacity on the bulk system we're not unintentionally creating a Train by using that asset on the distribution system a good example of that would be inadvertently charging a battery so that it can be dispatched for the bulk system in a window that might actually create or exacerbate an existing thermal loading issue on the distribution system so it's it's possible um there's a lot of complexity there as Mr Mar alluded to earlier there's not a single modeling or planning tool that co- optimizes uh economically and from a powerful flow and technical analysis perspective across generation transmission distribution understood thank you thank you oh thanks uh so much uh one last issue on timing it sounds like uh on the DSP uh um the open questions can be addressed on uh supplemental direct uh Mr iy Mr lson uh perhaps if there's jts you can just uh you know modify the uh testimony that's already filed um you know may maybe on the similar time frame as the the DSP uh I don't think we want to order additional supplemental Direct in the uh uh jts proceeding uh would that work for you Mr lson um yeah just just to clarify so let's take the DSP first you know we would file supplemental Direct on the topics in the order but for the ones that are the subject of the motion for variance on Friday the 21st yeah and it sounds like your team was okay with uh addressing some additional issues that came up today as part of that filing yeah with with some whatever with some understanding that some of it may not be uh uh capable of being done on that time frame as I think Mr Mino uh uh suggested correct yeah and we'll do we'll do the best we can kind of with the the dialogue today to to at least put some things some items in there that are responsive to the conversation today I think almost them are in distribution are in DSP as I'm looking at my notes but we might have to look a little more closely yeah I'm I'm not aware of something U Mr chair that affects the jts but I I may have I may on that if if if there is uh if we could uh handle that through maybe a testimony modification uh yeah we'll do an inventory of the items we'll do our best to respond in the DSP on the 21st to the extent we identify something that changes something within the context of the jts we we would do a a modification to the extent it was necessary if that approach works for you yeah and it's uh the EV manag Charing charging assumptions uh from uh good enough uh I think she had uh something on 10% manag charging so that that that that would be the one if Mr good enough maybe could adjust his you modify his testimony or supplement his testimony that'd be great yeah getting the actual numbers on manage charging would be helpful yeah we'll do our very best to do that by the 21st it's potential it might bleed a little bit but we'll we'll do it as expediently as possible yeah and there's uh no there's not going to be an under underlying order we're not taking a vote today it's not properly noticed so interested well with that I just uh want to express my uh appreciation to the company for uh enduring this French tribunal uh approach and uh uh helping us uh I mean it's just uh there's so much uh in front of of us uh our processes uh can be uh um rigid and we're trying to just create optionality and uh understanding as quickly and as best we can so just a a great deal of appreciation for the company mobilizing and uh helping us uh move this uh move this forward and hopefully it helps the parties with the discovery with the cross answer testimony and makes for a richer and better hearing so you know greatly appreciate you guys we we appreciate the opportunity for the discussion we certainly recognize these are significant cases uh we we know that you and your staff are working very hard on these amidst a lot of other things so the dialogue is helpful in both directions thank you thank you uh commissioner Gman uh anything you'd uh like to add before we uh adjourn no I agree would echo my appreciation to the company as well as to the commission's team for putting this together um and really figuring out how to address these two very major uh filings uh together so that we can better understand the forecast and I don't think it would have been beneficial to wait until the evidentiary hearing to get more clarity on how some of these really foundational issues uh impact these proceedings so uh it it certainly was helpful and appreciate you all for participating uh commissioner plant yeah I'd Echo that I appreciate you spending the time here with us today and um you know these are two areas of the energy system that are both changing really fast and you guys are modifying your processes I know um and uh and and so uh helping us understand how all of this works together is really uh helpful for us so I appreciate it and I just say the capital spending is uh unprecedented and we're trying to do our job to make sure it's uh thoughtful productive and uh appropriate so appreciate your help uh help for us uh in doing that so Mr iy Mr Larsson uh final comments before we adjourn nope I think we're all about to turn n minus one so I have no more agre nothing further thank you uh on that uh we're Jour thank you thank you