Research on Multi-objective Optimization Control of HEV Based on COBPNN

Abstract

In order to achieve the goal of energy saving and emission reduction of HEV (Hybrid Electric Vehicle), its control strategy should be able to carry out corresponding optimal control according to different running modes. In this paper, based on the analysis of the running state of the HEV, the overall model structure of HEV optimization control strategy is constructed, and the COBPNN (chaotic optimized BP neural network) is used to establish the running mode recognition and prediction model. On this basis, the multi-objective optimal control strategy of the switching process of the HEV based on the COBPNN is studied. The hardware-in-the-loop simulation experiment platform is built. The experimental results showed that application of the optimization control strategy, in the premise of the battery SOC balance, the total drive torque well followed the demand driving torque, met the power demands of the driver, and reduced the sum of engine and motor torque and the acceleration fluctuation amplitudes, maintained good driving comfort. At the same time, higher engine efficiency and better vehicle economy were obtained, emission characteristics had been improved. The experimental results proved the effectiveness of the multi-objective optimization control strategy.