Research On Modeling And SOC Algorithm Of Power Lithium Battery

Power battery is the core component in electric vehicle and energy storage system,and the mainstream power battery is usually lithium battery.In the process of providing energy and storing energy,the battery management system is needed to monitor the working status of the battery and manage the energy.Due to the high safety of the power lithium battery pack in the process of operation and the complex working environment of lithium battery,the equivalent model and the estimation of the charge state in practical applications have become a hot topic of research.This paper firstly introduces the basic concept and principle of power lithium battery,analyzes various characteristics of lithium battery,and lays the foundation for the subsequent equivalent modeling,parameter identification and charge state estimation analysis of lithium battery.Through the analysis of the equivalent circuit model of lithium battery,the improved PNGV model is selected to simulate the characteristics of the power battery,the hybrid pulse power test is performed on the battery,and the data are recorded,and the tested data are fitted with the state space equations of PNGV through Simulink to obtain an accurate mathematical model of the power lithium battery,and finally the curves of the test are compared with the curves obtained from the actual test Finally,the tested curves are compared with the actual tested curves to compare the results before and after the improvement,and the results prove the accuracy and high precision of the improved PNGV model.To address the problems of large error,low estimation accuracy and difficulty in real-time estimation when estimating the charge state,this paper decides to use the equivalent circuit model with higher accuracy,i.e.,the improved PNGV model,combined with the traceless Kalman filter algorithm to estimate the charge state,and the maximum error of estimating the charge state of the battery by the traceless Kalman filter method under constant current discharge is 2%,which makes the estimation of the charge state with higher accuracy and the algorithm estimation stability.This results in a high accuracy of the charge state estimation and a good stability of the algorithm estimation.Then,the experimental effects are analyzed under the dynamic stress test condition,and the model is simulated and verified based on the experimental data,and the accuracy of the traceless Kalman filter in estimating the charge state is better than that of the time-integral method and the extended Kalman filter method.This also demonstrates the superiority of the traceless Kalman filter in estimating the charge state of Li-ion batteries.