Research On SOC Estimation And Equalization Control Of Electric Vehicle Battery Management System

Recent years,the global environmental pollution has become increasingly serious,frequent fog weather at home and abroad,affecting people's quality of life and threatening people's lives and health safety seriously.Therefore,the scientific institutions and the government departments have begun to look for the environmental-friendly new means of transport-new energy electric vehicles.The power battery pack of the new energy electric vehicle is nonlinear during the operation.And there is inconsistency between each single battery in the power battery pack.Therefore,it is value to study SOC(State of Charge)estimation and its equalization control research of the power battery pack.Panasonic 18650 Tesla electric car with the battery is used to carry out research work in this paper.Firstly,this paper expounds the importance of the BMS(Battery Management System)for SOC estimate and the equalization control of the power battery pack.In order to meet the power demand of the new energy electric vehicles,should take many batteries in series or in parallel to form a power battery pack,because there is inconsistency between each single battery in the power battery,it will affect the overall performance and safety of the electric vehicle battery seriously during the operation.In order to solve this problem,this paper analyzes the inconsistency between lithium-battery and explains the significance of the research of equalization control,and contrasts the charge and the discharge equalization control methods on the simulation with MATLAB software.Results show that when the SOC is less than 10%,the discharge equalization is better than charge equalization.Secondly,in order to estimate SOC of battery,this paper uses the BP neural network algorithm(Back-Propagation),through the experiment to obtain the sample data and analyzes the relationship between them,uses the MATLAB software training experiment data,and the maximum error of the model is estimated by using the trained model to estimate the SOC of the battery,which is less than 4%.Finally,the overall design of the dual power management system is designed.