Research Of Ternary Lithium/Iron Phosphate Lithium Battery Management System

In recent years,with the increasing prominence of fossil energy and environmental problems,people’s environmental awareness and the improvement of sustainable development requirements,the application of clean energy has become an inevitable choice for today’s society.Battery as the main power storage tool,among which lithium iron phosphate battery and ternary lithium battery are currently widely used battery types.In this project,the 18650 model nickelcobalt-manganese ternary lithium battery and lithium iron phosphate battery are the research objects.The parameter identification method and state estimation algorithm required to establish the lithium battery management system are focused on research,so as to improve the accuracy of the state estimation function of the lithium battery management system.Specific research efforts include:Two lithium-ion batteries,lithium iron phosphate battery and ternary lithium battery,were selected to design experiments and test the capacity,resistance,open circuit voltage and other characteristics of the battery for research,and pulse discharge experimental data and dynamic working condition experimental data were obtained respectively.The fitting experiment of the open circuit voltage OCV-State of Charge SOC curve was completed through the open circuit voltage experiment.Through the study of the definition of state of charge,the calculation method of OCV-SOC is improved,and the calculation model of variable current SOC is built on the SIMULINK platform to complete the fitting of the improved OCV-SOC curve.By comparing various lithium-ion battery models,the second-order RC equivalent circuit model was selected.The three parameter identification methods of recursive least squares,second-order exponential fitting and Parameter Estimation Toolbox of forgetting factor SIMULINK model and MATLAB script code were constructed to identify the parameters of the equivalent circuit models of the two lithium-ion batteries,and the results showed that the error of the parameter estimation toolbox parameter identification method of the Parameter Estimation Toolbox was minimized by both lithium-ion batteries.The parameter identification method was improved and verified by introducing the idea of variable current,and finally the accuracy of the method was verified under DST working conditions,and the error was within 2%.When using the extended Kalman filter algorithm for battery SOC estimation,the idea of variable current is introduced,and the improved extended Kalman filter algorithm is proposed to be applied to battery SOC estimation.Two models of extended Kalman filter algorithms are built on the SIMULINK platform,and the results show that the improved extended Kalman filter algorithm improves the accuracy of the two lithium battery estimation SOC algorithms.Finally,a simple estimation of the state of power SOP is made based on SOC.Design each hardware module in the battery management system,write the corresponding software program,and build an experimental platform test platform.Test the accuracy of voltage,current,temperature and other data collected by the platform.The SOC written to the hardware is experimentally verified,and the results meet the requirements for battery management systems for pure electric vehicles in GB/T 38661-2020 "Technical Conditions for Battery Management Systems for Electric Vehicles".