[Music] welcome everyone my name is Kevin Oshiro and I am a member of the application engineering group at math works I will be presenting an introduction to modeling and simulation of hybrid electric vehicles I have a background in electrical and mechanical engineering my previous industry experience includes working at Packers Kenworth research and development center where I did research on hybrid electric vehicle power trains for medium and heavy-duty trucks my areas of interest include enabling a model-based design process using physical modeling tools I enjoy modelling mechatronic systems especially electrified power trains and I like to develop system level control strategies for these type of systems and I also provide training and mentoring for the eco-car student competition the objective of this presentation is to provide a foundation for you to model and simulate hybrid electric vehicles or HPVs so that you can conduct your own analysis such as HIV architecture selection energy consumption and performance estimates over different drive cycles and component selection these simulation studies will aid you in your HEV design process through a series of examples I'm going to show the initial steps of a model-based design process which will entail the following first of all model HEV architectures and powertrain plant models including the engine electrical and drivetrain components then we'll develop and implement HEV control algorithms and finally conduct an HEV design optimization process but let's take a moment to discuss in more detail the motivation for modeling hybrid electric vehicles there are many challenges when designing a hybrid electric vehicle you have to select the architecture or the topology of the powertrain you have to select the types and sizing of components such as the type of motor its maximum torque or power rating or the type of battery in its energy capacity it's also very difficult to model the HEV plant and control algorithms if you've never done so before you also want to do your best to optimize the HEV system performance over a wide range of op conditions and if you're intending to build real prototypes or production systems the control algorithms you develop have to be real time implementable that means you can run it on an embedded processor efficiently here is an example of the challenges in selecting an HEV architecture this is a typical parallel slash series parallel HEV architecture or topology the p and the pound sign indicates where the location of the electric machines in the powertrain are for example the p0 location is the front of the engine typically the machine is connected to the engine crankshaft through the front accessory drive belt the p1 location is the flywheel side of the engine the p2 location denotes the input to the transmission and always has a clutch between the motor and the engine the p3 location can be located inside the transmission or on its output and the p4 location is on the axle which is the axle that is not connected to the internal combustion engine you can have multiple combinations of machine locations and there are intrinsic pros and cons for each machine location here are a couple of examples of a parallel and series parallel architecture the p2 parallel architecture has a single machine with a disconnect clutch allowing for full electric or evie mode as well as parallel hybrid mode meaning the engine and motor will provide power at the same time the p1 / P 3 series parallel architecture allows for easy mode series hybrid mode and parallel hybrid mode there are other architectures as well including input power split architectures but you can see how many combinations there are just for the parallel slide series parallel topologies and it would be too time-consuming to build physical prototypes of all the combinations possible assessing the intrinsic pros and cons of these different architectures by modeling and simulating them virtually is a much more efficient approach here is another design challenge example regarding selection and sizing of components in the HEV powertrain for an electric machine or motor generator you can select the type such as interior permanent-magnet or induction machines you can also select the max torque or max power of the machine for batteries you can select the type of battery or the energy capacity or cell or module configuration such as the number of cells in series in parallel like the discharge and charge power capability there are also many types of transmissions including multi speed transmissions such as automated manual transmissions dual clutch transmissions or conventional automatics as well as fixed gearing one of the most well known transmissions are based on planetary gear sets for input power split h-e-bs such as the toyota prius or GM volt you can also inter connect multiple planetary gears in different ways to make a large number of topologies finally you have to select the gear ratios and these components including the differential you can see how large the design space is for HEV architectures and components another design challenge example is how to optimize performance over a wide range of operating conditions conditions include different type of drive cycles drive cycles such as FTP 72 have slower transit speeds to represent city driving or drive cycles can have higher steady-state speeds such as the highway VT cycle or a drive cycle can represent aggressive driving as shown in this u.s. 0-6 cycle you have to account for different Road grades and temperature ranges as well the HEV controller has to reduce the total energy consumption reduce emissions and meet drivability requirements such as maintaining or improving acceleration times and maintaining speed at different Road grades the solution to these HEV design challenges is to utilize a model-based design process in upcoming examples 1 & 2 I will show how to build plant models and develop control algorithms you will then be able to evaluate different HEV architectures you can use the system level model to perform component sizing studies and assess the system performance and most importantly you be able to do closed-loop control development early in the design process in example three we will discuss optimizing the control implant simultaneously and it's not covered during these examples but please note that the models you build now will be reused throughout the design process for tasks such as code generation hardware and loop and verification and validation this model reuse is the basis of model-based design let's now take a look at a completed hybrid electric vehicle example to see what it looks like and what it does the model I'm showing is created in the math work Simulink environment specifically using the powertrain block set tool which I will discuss in more detail in a later example this is an example of a closed-loop system level vehicle model the model works as follows first the drive cycle source block generates a standard or a user-specified drive cycle speed versus time signal this is known as the reference or target speed next a virtual driver model generates accelerator and brake pedal signals based on the vehicle target and feedback speeds controllers for the powertrain system used the accelerator and brake pedal signals along with other system feedback signals and output torque commands to the actuators in the system for example the engine or motors the controller commands go to the vehicle system or physical models of the engine electrical system such as the motors or batteries and drivetrain dynamically output torques and forces to move the vehicle and there is a visualization system which this displays the dynamic signals during the simulation if we press the play button the model will simulate dynamically by stepping forward incrementally in time this type of simulation is also known as a forward-facing or forward-looking simulation let's go into the visualization subsystem and view the scope as the simulation is running we can see the vehicle speed tracking the drive cycle target speed here in the upper left graph we can see signals of the actuator torques and speeds such as the engine and motor speeds and the engine and motor torques there are also graphs of the instantaneous battery current and instantaneous battery state of charge and there's also a graph showing the calculated fuel economy in the units of miles per gallon equivalent you can use this type of model for assessing the performance or energy consumption of different HEV architectures sizing of the physical components or optimizing the controls or system as you can see being able to model and simulate a hybrid electric vehicle is a very powerful tool that will help you find solutions to various HEV design challenges HEV design is very complicated but using a model-based design process to iterate and optimize designs faster is the key to being successful please watch the next three examples so you can see how to model an HEV implement HEV control algorithms and conduct an HIV design optimization process thank you for your time