76 Electric and Hybrid Vehicles • Component sizing – Estimating the power and energy requirements of powertrain compo- nents such as electric machine power rating, inverter kVA requirement and battery capacity. • Component testing – Powertrain and other auxiliary components can be tested with regards to their capacity and constraints including loss estimations under different driv- ing scenarios. The loss estimations of the powertrain components are particularly useful in design the thermal systems for those such as electric machines, power electronics and battery systems.
For vehicle simulation, a forward-looking model is desired where a vehicle model responds to driver input commands to develop and deliver torque to the wheels similar to what it is in a real-world scenario. A modular structure of the simulation model is highly desirable so that components and configurations can be easily added or removed without having to start from scratch. This facili- tates comparisons among different component technologies or components of different ratings. The performance comparison of alternative hybrid architectures is also easily accomplished with the modular structure. The models are developed using both physical principles and empirical data.
The vehicle dynam- ics and roadway fundamentals presented in Chapter 2 are the physical principles based on which the vehicle model can be developed. The driver behavior is modeled using various drive cycles, which are standardized speed and road grade profiles used to evaluate a vehicle’s performance. For the components, either simplified models or look-up tables based on empirical data are used in the simulation. Detailed dynamic behavior of the components is often not required, since the time responses of many of the subsystems (such as the electric machine) are much faster than the vehicle dynamics.
Simpler models of subsystems minimize the simulation time, since otherwise it would be significant for such a highly complex vehicle system. With this approach, it is possible to run a large number of simulations in order to tune the control strategy to perform well under a wide range of driving conditions. Because there are both mechanical and electrical dynamics, a variable-step stiff system solver is required. The simulation block diagram of a vehicle system showing the interactions among the driver, the supervisory controller, the subsystems and the vehicle model is shown in Figure 3.19.
The driver is modeled with a drive cycle and a PI regulator designed to minimize the error between the drive cycle velocity profile and the actual vehicle velocity. The speed profile of the drive cycle is assumed to be the vehicle speed desired by the driver. The PI regulator output creates a pedal position which is interpreted as either the accelerator pedal position or the brake pedal position. It is assumed that both pedals cannot be depressed simultaneously. The pedal position is read into the control strategy (located in the SCM or supervisory control module) to generate the subsystem commands.
The powertrain, modeled in the subsystem block, produces the tractive torque T TR , which is sent to the vehicle model. The brake pedal position input is passed through directly into the brake subsystem model to address safety concerns; however, the brake pedal position is also required by the supervi- sor controller to determine the magnitude of the powertrain regeneration command. Driver Model Vehicle Roadway input (grade) Velocity profile (Drive cycle) Vehicle velocity FIGURE 3.19 Vehicle simulation block diagram. >0; Acce PI rator SCM Subsystem Feedback commands signals Vehicle Subsystem TTR Model Models le PI e