The Research On Lithium-ion SOC Estimation And Implementation Based On The SVD-KF Algorithm

Since the 21 century,the issue of energy and environment has become an important topic of concern for the well-being of all mankind,and governments have sought to establish their own clean and renewable energy systems.More and more attention has been paid to the R&D and application of environmentally friendly and almost zero emission new energy storage lithium batteries.Compared with other batteries,lithium batteries have the advantages of high energy density,long cycle life and reliable performance.However,the lithium-ion battery is more active in chemistry and sensitive to environmental changes.Therefore,the lithium battery management system is needed to effectively control and manage the state of the battery.Battery charge state estimation(SOC)is one of the core technologies of lithium battery management system.Accurate SOC estimation is a prerequisite for the normal operation of the whole battery management system.And it is of great significance to the battery life and the improvement of the performance of electric equipment.In this context,a series of research work on lithium battery SOC estimation has been carried out in this paper.In this paper,the background and significance of the research on SOC estimation and battery management system of lithium battery are introduced,and the electrochemical characteristics of lithium battery and the influence of relevant technical parameters of lithium battery on battery performance are expounded.On the basis of understanding the working characteristics of lithium battery,the advantages and disadvantages of various mathematical models of lithium battery are analyzed,and the tradeoff between the engineering implementation of SOC estimation algorithm and the performance of the model is made.Finally,the two-order RC equivalent circuit model is chosen as the mathematical model of SOC estimation for lithium battery.In order to improve the SOC estimation accuracy and efficiency of the two-order RC model,this paper presents a method to identify the parameters of the lithium battery model by using the Parameter Estimation toolbox of the MATLAB software.The parameters of the identified lithium battery model are simulated by using MATLAB software.The simulation results show that the identification method is accurate and can simulate the working characteristics of the lithium battery.Secondly,considering the nonlinear characteristics of the lithium battery model,it is necessary to have an algorithm with high accuracy and good robustness in order to estimate SOC accurately.By analyzing and verifying the performance of various SOC estimation algorithms for lithium batteries,the SVD-KF algorithm,which is improved by uns-cented Kalman Filter,is determined as the SOC estimation algorithm in this paper.The algorithm combines the second-order RC equivalent circuit model of lihium battery with the current SOC and two polarization voltages of the battery as the system state variables,and uses UT transform to carry out the nonlinear transfer of the system state.A priori estimate is modified by combining the observation equation and the Kalman Gain,and a posteriori optimal estimate of the system state is obtained.In this paper,using the SVD-KF algorithm framework,the corresponding mathematical model is built in MATLAB,and the simulation work of SVD-KF algorithm is completed by using the custom working condition.The results show that the filter algorithm can track the SOC of lithium battery well,and the accuracy of the stability error of SOC estimation is controlled within 3%.Finally,in order to verify the performance of SVF-KF algorithm,this paper designs a hardware platform of lithium battery management system based on IMX.6UL chip.The management system can simultaneously measure 12 channels of lithium battery voltage,and has simple passive equalization function,Charge-discharge protection function and communication function etc.Based on the hardware platform,the embedded Linux operating system is transplanted to the platform,and the sub-module of system data acquisition,the sub-module of SOC estimation and the sub-module of equalization are completed.The system overvoltage and under-voltage protection sub-module and communication sub-module and other sub-module software programming work.Finally,the compiled and designed software program of battery management system is downloaded to the hardware platform by cross compiling tool chain,and the verification test of the working condition of the SVD-KF algorithm is carried out.