Electric Vehicle Modeling and Simulation.

Abstract : The forecasting of the performance of electric vehicles has revealed many discrepancies when compared to actual test data. An electric vehicle computer simulation program was developed to ameliorate this deficiency. The approach was to establish a very comprehensive and flexible vehicle model and simulate its operation on a realistic driving cycle. The driving cycle selected was the Federal Urban Driving Sequence. A thorough vehicle model was established that incorporate aerodynamic drag, rolling asistance, both rotational and translational inertial effects, and component by component dynamic power train efficiencies. Battery discharge performance is tracked by a fractional-utilization algorithm with corrections for short-term discharge effects. The simulation compares required power obtained from the driving cycle speed schedule and vehicle model characteristics with the available power at the motor for each time increment of the driving cycle to compute battery fraction used and deviation from the speed schedule when available power is insufficient. These results of the simulation can be used to evaluate an existing vehicle's performance or, if desired, to develop vehicle parameters to obtain a specified performance level. An application of the program to develop a suburban passenger vehicle is included to demonstrate the simulation's utility. A test bed vehicle was constructed and tested to verify the simulation. Additional aspects such as microprocessor based controllers including implementation of an optimal control law were investigated to gain insite into the efficiency and performance trade-offs of such a system.