Research on Performance Evaluation System of Electric Vehicles under Different Operating Conditions based on Entropy Weight Method and Grey Correlation

Abstract

With the rapid iteration of the new energy vehicle industry, the performance evaluation of electric vehicles has become the core basis for the iteration of vehicle enterprises, the improvement of industry standards and the decision-making of users' purchase of vehicles. The current evaluation system mostly adopts fixed weight and single standard working condition, which is difficult to adapt to the actual scenes such as congestion, high-speed, ramp and low temperature. There are problems such as single dimension, fixed weight, insufficient accuracy and "excellent laboratory and deviation from actual experience". In this paper, an adaptive weight evaluation system based on entropy weight method and grey correlation analysis is constructed. Screening quantitative indicators around the five dimensions of power, energy consumption, endurance, safety and comfort; The characteristic parameters and sensitivity model of working conditions are introduced to realize the dynamic adjustment of weight with working conditions; Objective weighting of entropy weight and grey relational fuzzy evaluation are integrated to improve the quantization accuracy; Based on Python software, a variety of mainstream pure electric vehicles are selected for demonstration under four typical working conditions: urban congestion, high-speed cruise, mountain ramp and low-temperature endurance. The results show that the consistency between the evaluation results of the system and the actual performance of the vehicle is more than 10% higher than that of the traditional model, which can accurately identify the advantages and disadvantages of the vehicle mode adaptation, and provide theoretical support and engineering basis for product optimization, calibration and the establishment of industry dynamic standards.