Abstract
With global warming & Fuel crisis, the interest in Hybrid vehicle (HEV) & Electric vehicle (EV) is at its lifetime peak. Development of EV/HEV power-train needs to be extensively supported by simulations and testing. Key test/development parameters for EV/HEV application are: range of distance a vehicle can travel in single charge, torque/speed performance of the power-train, accuracy of “State of charge estimation” etc. All these tests require real life driving conditions to be simulated for a few hours. A conventional Chassis Dynamometer can be used to simulate all the driving conditions for a specified driving pattern. However it is impossible for a human driver to apply specified driving pattern for a few hours accurately. To overcome this difficulty this paper describes a driver-less Chassis Dynamometer test set-up which can emulate specified driving patterns to an EV/HEV powertrain for desired amount of time. The system uses National Instrument hardware and a PC as the controller. The desired drive cycle can be programmed/selected by the test engineer. The controller emulates driver behaviour by sensing the vehicle speed and adjusting the throttle input to the EV drive-train & using the regenerative breaking in a way to get desired speed profile without using mechanical breaking. The speed control algorithm essentially implements a “feed-forward plus Proportional Integral” strategy so that the torque-command generated instantaneously by the controller in turn attains the desired speed as dictated by the driving cycle. The paper also explores powerful potential of this concept which can be extended for simulating various load conditions for battery and EV drive system.
Keywords: Electric Vehicle, Dynamometer testing, Electric Power train, Labview, Real Time Control.