Linear Interpolation Based Parametric Model of Electric Vehicle for Tractive Force Analysis

Abstract

Electric vehicles (EVs) are designed with accurate and computationally efficient models for vehicle dynamics assessment, and for estimating the tractive power demand and real-time controller design. For the analysis of total tractive force and power requirement of an electric vehicle under different operating conditions, this study proposes a linear interpolation-based parametric model. The developed model contains the key resistive forces such as rolling resistance, aerodynamic drag, grading resistance, and acceleration resistance from the basic vehicle dynamic equations. The influence of the vehicle mass, road gradient, and aerodynamic drag coefficient on both passenger cars and heavy electric trucks was examined in a three-dimensional parametric study in MATLAB. The vehicles span a range of sizes from 800 kg to 1500 kg for a passenger car and 2000 kg to 3500 kg for a truck, and road gradients varied from 0° to 10° and drag coefficients from 0.19-0.4. The MATLAB interp1 function was used to predict intermediate values of tractive power and to minimize the repetitive computer calculations. Road grade and vehicle mass were shown to cause a marked change in the demand for tractive power at low and high speeds in the simulation tests. Passenger cars had a traction demand rising from about 0.4 kW at low speed and flat roads to almost 8 kW at high speed on slope roads, with heavy trucks having a demand of up to 16 kW. Aerodynamic drag had little effect at lower speeds, but at speeds in the vicinity of 100 km/h, it was dominant. Extensive real-time validation with ADVISOR vehicle data showed good agreement between estimated and model-based tractive forces; a mean prediction error of about 160 N and a maximum deviation of 347 N. The prediction errors of the proposed model were found to be less than 5%, which proves the model is suitable for fast EV performance estimation, sensitivity analysis, and real-time control applications.