The Research Of Online SOH Estimation Of Power Battery

With oil prices rising,PM2.5 continuous burst table to bring no small pressure to the traditional automobile manufacturing enterprises.,which forcing these enterprises towards a more promising new energy vehicles.Most of these new energy vehicles with battery powered and combined with battery management system to meet people's daily driving needs.However,new energy vehicles are still at the fledgling stage,most of the research focuses on the precise study of the battery State of Charge(SOC)and obtained good results,while the health of the battery is less involved.At present,the health of the battery has become the focus of attention with the phenomenon of spontaneous combustion,explosion and other phenomena occur frequently,which prompting people to focus on the battery State of Health(SOH).The health of the battery has become an important part of battery management,how to predict the development of new energy vehicles SOH accurately is of great significance.This article is based on this as the background of the battery health status of the study,the specific work is as follows:First of all,the paper analyzes the life decay of lithium battery and referring to the existing literature and found that the value of ohmic resistance can be regarded as the criterion of SOH.Then the equivalent circuit of the two order RC cell is selected by comparing and analyzing the existing three kinds of models from the aspects of accuracy and calculation.Finally,the parameters of one battery has been measured by the recursive least square method,and the identified values are added into the solution to determine the corresponding models of the battery.It is concluded that the model is a nonlinear system by analyzing the equivalent circuit of the battery,so the particle filter algorithm is proposed to deal with nonlinear and non Gauss system.In this paper,by using the idea of Sequential Importance Sampling(SIS)in the sampling of high dimensional function,the algorithm of particle filter is replaced by the time function.However,the degradation of the particle disappears with the increase of the number of iterations when the particle filter is used to deal with the problem,which will seriously affect the accuracy of the prediction.In this paper,the concept of resampling is introduced on the basis of the original particle filter.The idea is that when the effective particle number is lower than the set threshold,all the particles will be reassigned.The assigned particles continue into theloop until the specified number of iterations is completed.The dynamic equation and observation equation are established through the equivalent circuit model.According to the steps listed in the programming and simulated by MATLAB.It is found that the resampling particle filter has a high accuracy in predicting the internal resistance of the battery by the observation of the simulation graphics.This paper introduces the idea of genetic algorithm to replace the resampling process in order to estimate the internal resistance of the battery more accurately.The algorithm not only can solve the problem of particle degradation,but also enriched the types of particles by gene recombination and gene mutation.The idea is that when the effective number of particles is lower than the set threshold,the genetic algorithm will be processed.The treated particles continue into the loop until the specified number of iterations is stopped.Finally,the specific operation steps of genetic particle filter are listed using the dynamic equation and observation equation.According to the steps listed in the programming and simulated by MATLAB.It is found that the curve predicted by genetic particle filter is gentler and the jitter is smaller by comparing the two algorithms.Therefore,it is indicated that the genetic particle filter is more accurate to estimate the internal resistance.