Capacity estimation and energy allocation model of new energy vehicle battery management system based on optimization algorithm

With the charging and discharging cycle use of the battery, the capacity of the battery will decline, which affects its operational safety, and it is necessary to accurately estimate the capacity state and mixing degree optimization of new energy vehicle batteries. In this paper, we adopt the mothballing algorithm (MOF) to optimize the overfitting phenomenon of the random forest regression algorithm (RFR), and then optimize its characteristic parameters to obtain the optimal parameters suitable for capacity estimation, and establish the MOF-RFR battery capacity estimation model. Then, taking a pure electric vehicle as the research basis, the energy allocation control strategy of new energy vehicle battery based on multi-objective particle swarm optimization algorithm is proposed by comparing the existing power and economic recovery control strategies. The model of this paper is compared with different algorithms, and the model accuracy is verified for multiple sets of battery data of the same type. The results show that the established model can effectively estimate the battery capacity, and the average error is controlled to 0.33%, and the overall error is within 1.5%. The 100km hydrogen consumption (L) of the pure electric vehicle is reduced by 2.32 L. It can be seen that the model in this paper does not affect the dynamics of the car while at the same time improves the economy of the car, so that the performance of the whole vehicle is improved, which is of great practical significance.