Design and optimization of driving motor cooling water pipeline structure based on a comprehensive evaluation method and CNN-PSO

Abstract

Thermal performance has a significant impact on the efficiency of electric vehicle driving motors. Currently, active cooling methods such as liquid cooling are mainly used for driving motors. To investigate the effective cooling methods for electric vehicle driving motor, the structure of water pipeline is calculated under different working conditions through a comprehensive evaluation method and machine learning. This study aims to find a cooling water pipeline that enables the driving motor to have the best working performance. In this study, the electromagnetic losses were calculated and the temperature distribution contours of the motor under different working conditions were analyzed by simulations and experiments. A comprehensive evaluation method is proposed to calculate the influence of structure on the performance of designed water pipeline. Besides, the optimal feature size was selected by convolutional neural network and particle swarm optimization integrated algorithm (CNN-PSO) to find the best overall performance of the water pipeline. According to the results, the installation of cooling system can effectively avoid the thermal runaway of the driving motor. Based on the comprehensive evaluation method of the energy consumption and cooling capacity of the system, four different types of cooling water pipeline were selected and analyzed. The results show that the driving motor with spiral water pipeline has the best working performance. After optimization, the cooling capacity remains the same and the pressure drop of spiral pipeline is reduced by 24.57%.

Keywords

• Driving motor
• Electromagnetic losses
• Spiral pipeline
• Comprehensive evaluation method
• CNN-PSO