Experimental Model Identification of a Pure Electric Vehicle Using Standard Dynamometer Testing

The prevalence of electrified vehicles for clean and efficient transportation systems has grasped an increasing attention of many researchers and academics in the last few years. In this context, accurate, yet attainable, modeling of electric drivelines plays a significant role to ensure optimal sizing, power management, and online control of such propulsion systems. This paper proposes a systematic approach for online modeling parameters identification of electric drivelines during standard test-drives on chassis dynamometers. The proposed approach depicts real test measurements into a self-adaptive algorithm, that tunes arbitrary modeling parameters using minimal least- square scheme to converge to reference outputs congregatedly. The achieved results reveal the ability of the proposed approach to yield accurate modeling parameters of the electric motor and battery and enable an extended post-test analysis of the investigated drivelines.