Study On Combined Online Estimation Of SOC And SOP Of Lithium Battery At Different Temperatures

In the current social environment,lithium-ion energy batteries have been widely used in various industries,especially in the field of new energy electric vehicles,because of their advantages of no pollution,light weight and large storage capacity.As an important part of the battery management system,the battery state of charge and peak power have an important impact on the acceleration braking and climbing performance of electric vehicles.The estimation of traditional battery peak power is tested by the pulse method based on a rint resistance model.The model is simple and the estimation error is large.There are many problems,which can not improve the performance of battery management system.In this paper,the on-line estimation of SOC and SOP of 18650 ternary lithium battery is taken as the research goal and an experimental platform is built.According to the corresponding relationship between the on-line estimation of SOC and SOP,the two states of battery are jointly estimated on-line.Firstly,this paper introduces and analyzes the structure,working principle and main parameters of lithium battery,and tests and studies the terminal voltage,internal resistance,capacity and other characteristics of lithium battery.Considering the internal parameters and state of lithium battery,this paper selects the second-order RC equivalent circuit model as the model of this subject,and uses the improved identification method to identify the parameters of lithium battery online at different ambient temperatures,The variation of each parameter under different ambient temperature is studied to improve the identification accuracy.Secondly,based on the analysis of common lithium battery SOC estimation methods,according to the model parameters identified online at different temperatures,this paper optimizes the fixed process noise variance in the original extended Kalman filter algorithm through the improved extended Kalman filter method,that is,the differential selection algorithm,so that the noise variance can be automatically adjusted according to the changes of the environment,This changing variance is substituted into the original extended Kalman filter algorithm to complete the on-line estimation of SOC under cyclic conditions.By simulating UDDS working conditions at different temperatures,the estimated value of the algorithm is compared with the true value,which proves that the method has small error fluctuation and applicability.Finally,based on the test and test of battery parameters,considering the influence of SOC of lithium battery on peak power SOP and the limitation of single parameter estimation,this paper adopts the peak power estimation method based on second-order RC model and multi parameter constraints,introduces the estimation formula,carries out the joint online estimation of battery SOC and SOP according to the improved EKF algorithm,and carries out the experimental verification at different temperatures,The online estimation accuracy and applicability of the two methods are analyzed.