Model-based SOC Estimation And Equalization Control Study Of Lithium Batteries

As an important part of energy storage system,lithium batteries have the advantages of low self-discharge rate,high energy density and long service life,and has been widely used in many electrification fields such as electric vehicles,nuclear power plants,microgrids and many other electrification fields.The battery state of charge,residual life prediction and energy balance as the key performance of the battery management system,which have important research value.This article focuses on the estimation of the state of charge,residual life prediction,and energy balance of lithium batteries,in order to improve the application layer development of the battery management system.The details of the research are as follows:Firstly,a method for estimating the charge state of lithium batteries based on an improved federal extended Kalman filter algorithm is proposed to address the problem that the traditional Kalman filter algorithm uses only one sensor to collect data resulting in low accuracy.The federal extended Kalman filter algorithm is improved to use recursive least squares with forgetting factor for online identification of lithium battery parameters to improve the adaptability of the system to parameter changes.The D-S evidence theory is used to update the information fusion weights of the filtering algorithm in real time,which overcomes the drawback that the traditional federal extended Kalman filtering algorithm is empirically given and invariant when assigning weights to the fused information.The D-S evidence theory uses "interval estimation" rather than "point estimation",which provides great flexibility in distinguishing between unknown and uncertain aspects and in accurately reflecting evidence collection.Simulation results under both conditions show that the improved algorithm in this paper has higher estimation accuracy and robustness compared with the adaptive unscented Kalman filter algorithm and the adaptive extended Kalman filter algorithm.Secondly,a method for predicting the remaining life of lithium batteries based on an improved particle filtering algorithm is proposed for the problem of particle depletion that easily occurs in the iterations of the traditional particle filtering algorithm for lithium battery life prediction.For the particle depletion problem,an arithmetic optimization algorithm is used to optimize the particle distribution in order to improve the particle diversity.The optimization search process of the arithmetic optimization algorithm is mainly divided into two stages: exploration and development,and these two stages are switched at any time according to the optimization progress with strong optimization capability.The particle set in the particle filtering algorithm is regarded as an operator in arithmetic for optimization,which makes the optimized particle population move to the high-likelihood region,ensures the diversity of particles in each iteration of the algorithm,suppresses the phenomenon of particle depletion,and improves the prediction accuracy of the remaining lifetime of lithium batteries.The simulation results show that compared with the traditional particle filtering method,the prediction error of the proposed fusion algorithm is only 4 cycles and 6 cycles at the 80 th cycle and 60 th cycle,which has a higher prediction accuracy of the remaining life of Li-ion battery.Finally,a fixed time distributed equilibrium control strategy for the charge state of lithium battery packs based on a finite time disturbance observer is proposed to address the issue of inconsistent charge states of each individual lithium battery during use.In the physical layer,considering the impact of input disturbances on the equilibrium results of the equilibrium system,a model of the equilibrium system with disturbances is established,and a finite time disturbance observer is constructed to estimate the disturbances in real-time.In the information layer,the lithium battery pack is regarded as a multi-agent system,and the communication between single lithium batteries is described by a directed topology.Design a fixed time distributed collaborative balancing strategy to ensure that the state of charge of individual lithium batteries tends to be consistent within a fixed time.The stability proof of the disturbance observer and closed-loop system is provided.Simulation experiments have verified the effectiveness of the proposed method.