SOC Estimation Algorithm And Verification Of Power Lithium-Ion Battery Considering Temperature Influence

Power lithium-ion battery used in electric vehicles has become a hot research topic in domestic and international electric vehicle industry because of its high energy density,low self-discharge,good safety performance,long life and no pollution and other advantages.In order to be able to provide users with accurate current battery operating status,to ensure the stability and security of the battery operating system,battery management system(BMS)assumes its main function,and together with the battery pack forms a reliable battery system.And SOC estimation of power lithium-ion battery as the core algorithm of the battery management system,plays an important role in electric vehicles,including preventing battery overcharging and over discharging,providing accurate driving range and extend the battery life and so on.However,at present,most of the researches on the SOC estimation of power lithium-ion battery are for single-type battery in normal temperature environment,and the verification conditions are not sufficient.It is difficult to prove the applicability of the algorithm to other temperatures or other types of lithium-ion batteries.Therefore,it is of great theoretical and practical significance to study the SOC estimation strategy applicable to different ambient temperature and various types of lithium-ion batteries.The paper is based on how to establish a accurate and reasonable lithium-ion battery model,design a new SOC estimation strategy and realize offline and online simulation verification and other issues for analysis and research.The main work contents of the paper involves the following three aspects:1.According to the accuracy requirements of the model and the engineering practicality of the algorithm,it is determined that the control object model is the second order equivalent circuit model.The battery test experiments are completed at different temperatures using the machine for charging and discharging and the temperature control box,and then experiment data are obtained by statistic and analysis.The battery capacity,ohmic resistance and open circuit voltage are calibrated,and polarization effect parameters are fitted by using the least squares method.Dynamic battery model based on three parameters of temperature,current and SOC is established.And validation and analysis for Lithium-ion battery is finished by a variety of operating conditions and temperatures.The results show that the calibrated battery parameters are able to better adapt to the type of lithium-ion battery.2.According to the estimation requirement of battery SOC and the non-linear characteristic of dynamic battery system,it is determined that the estimation strategy used in the paper is extended Kalman filter method.In order to make the SOC calculation formula more accurate,it is proposed to use the relative available capacity to replace the original rated capacity,and temperature factor is also considered.On the basis of this,a variable gain coefficient is introduced to modify the extended Kalman filter algorithm.It can be seen from the simulation results that within a certain range the larger the gain coefficient is,the faster the convergence rate is and the larger the overall error is.On the contrary,the smaller the gain coefficient is,the slower the convergence rate is and the smaller the overall error is.So the modified gain coefficient can make the estimated value to meet the requirements of fast convergence and small error.3.The calibrated battery model and the modified extended Kalman filter algorithm model are established by Simulink software platform.Then the off-line simulation of the control strategy is completed for different temperatures and different conditions.The results can meet the expected requirements.On the basis of this,we further build a rapid control prototype platform based on Micro AutoBox and realize the on-line simulation verification of two types of lithium-ion batteries at fixed temperature,fluctuating temperature,current noise input and different initial SOC.The results can show that the SOC estimation strategy of the power lithium-ion battery at different temperatures can achieve the overall error of less than 5% in a variety of working conditions,with good robustness and accuracy,and to a certain extent can achieve its engineering application value.