Research On SOC Estimation And Two-Stage-Equalization Strategy Of Power Battery

Lithium-ion power battery has a wide application prospect in the field of electric vehicles because of its long cycle life,high specific energy and low self-discharge rate.In order to ensure the stable and efficient operation of multiple series-parallel batteries,it is necessary to effectively monitor and control the batteries through the battery management system.Therefore,it is of great significance to conduct research on State of Charge(SOC)estimation and equilibrium management in battery management systems in this article.In order to establish a stable and reliable SOC estimation model and equilibrium management model for power batteries,research was conducted on model based SOC estimation and equalization strategies based on two-stage reconfigurable equalization circuits,and an improved grey wolf optimization algorithm is proposed to identify offline parameters of battery equivalent circuit model,and a SOC joint estimation method combining on-line parameter identification with state estimation algorithm with high accuracy is selected,and the intra-group equalization threshold and inter-group equalization ratio are improved.The details are as follows:(1)Select the target battery with model NCR18650BD,obtain the test data of the battery under different working conditions by the battery test platform,and fit the SOCOCV relationship curve of the target battery.Then the commonly used battery models and identification methods of model parameters are introduced.Proposed the use of an improved grey wolf optimization algorithm for offline parameter identification of second-order RC equivalent circuit models,and the equivalent circuit model of the target battery is built in the MATLAB/Simulink simulation tool,and verified that the equivalent circuit model has high accuracy.(2)The SOC estimation method based on battery model and the online parameter identification method of battery model are introduced.The SOC joint estimation method combining recursive least square method with forgetting factor and adaptive extended Kalman filter algorithm is selected,and the SOC joint estimation model is built in the MATLAB/Simulink simulation tool,and the reliability of the SOC joint estimation model built in this paper were verified through comparison with operating condition test data.(3)The equalization principle of reconfigurable circuit,equalization circuit based on DC-DC converter,and equalization control algorithm based on two-stage reconfigurable equalization topology are analyzed,and an inter-group equalization strategy with extreme value insertion and fixed proportional coefficient is proposed.The selection of intra-group equalization strategy adopts double thresholds as equalization criteria,including the range and root mean square of SOC.The balanced management model is built in the MATLAB/Simulink simulation tool,and the simulation experiment proves that the balanced control algorithm and strategy adopted in this paper can realize the balanced control between batteries effectively.(4)According to the data acquisition card and peripheral circuits,a hardware-inthe-loop simulation platform is established,and the SOC joint estimation model and balanced management model built in this paper are verified by experiments under different working conditions set by electronic loads.Experiments show that the SOC estimation model can maintain good estimation accuracy for real-time input experimental data,and the balanced management model can effectively control the onoff of switches,and finally reduce the difference between battery voltage and SOC.