State Of Charge And State Of Health Estimation Of Li-ion Batteries Based On Adaptive Square-root Unscented Kalman Filters

With the rapid development of science and technology,as well as the overall construction of ecological civilization,electric vehicle industry has acquired a fast growth.Compared with other storage batteries,lithium-ion battery has higher voltage and energy density,and its self-discharge rate is lower,which has become mainstream power source for electric vehicles,and also a hotspot in the field of energy research.State of charge(SOC)and state of health(SOH)of lithium-ion battery directly affect the safety and economic efficiency of electric vehicle.It is necessary to design management system to monitor the states of the battery.In this paper,Considering that the static and dynamic performances of Lithium-ion battery in charge/discharge process is complex,a second-order RC model is established.Due to the change of noise covariance of battery system,the accuracy of SOC estimation is not satisfactory.After studying the main factors affecting battery state estimation comprehensively,on the basis of Sage-Husa adaptive filter,the traditional square-root unscented Kalman filter(SRUKF)is modified and an adaptive SRUKF(ASRUKF)is proposed.And then discharge experiments are designed.Compared with SRUKF,this method reduces the model error and algorithm error in SOC estimation greatly,and has higher reliability and convergence.Finally,in order to realize the prediction of the SOH,extended Kalman filter(EKF)is used to estimate the capacity and ohmic resistance of battery online in the process of SOC estimation.Real-time estimation of SOC as model parameter estimates the two parameters,which achieves real-time update more precisely,and provides the necessary information for the prediction of SOH.By means of using the environment of Matlab simulated the proposed method,the preliminary experimental results show that the estimation strategy not only makes the estimated value of SOC more close to the true value,but also provides the real time estimation of the battery capacity and the ohmic resistance,realizing the prediction of SOH.