SOC Estimation Of Lithium Battery Based On Dual-filter Structure Of I-STCKF

At present,lithium battery is the main power source of pure electric vehicle.Therefore,in order to ensure the safe and efficient operation of the lithium battery system,it is necessary to use BMS to manage the battery system.SOC estimation is one of the core technologies of BMS.Therefore,this paper analyses and summarizes the problems of SOC estimation with non-linear filtering algorithm such as uncertain statistical characteristics of noise,inaccurate system model and undefined covariance matrix.On this basis,an improved strong tracking filter(I-STF)is proposed to correct the accuracy,as the system model is untimely.Combined with the improved Capture Kalman I-CKF,a battery SOC estimation algorithm with double filter structure is established,which realizes the joint online estimation of model parameters and battery SOC.Firstly,a lithium battery test platform is built.The available capacity and resistance characteristics at different discharge rates and temperatures are tested.The influence of environment temperature,available capacity and discharge rate on the OCV of lithium battery is analyzed.A database of relevant performance parameters is established,which provides data support for subsequent algorithm verification.Secondly,the second-order off-line time equation of ECM is established,and the results of parameter off-line identification are obtained on this basis,which provides a reference for subsequent parameter on-line identification.Subsequently,the existing SOC estimation algorithms are analyzed,and the I-STF is proposed to deal with the time-varying characteristics of noise and model.Combining with I-CKF,the calculation process of covariance matrix is improved,and a SOC estimation algorithm with double filter structure is established,which realizes the joint online estimation of model parameters and battery SOC.Finally,the battery is tested under HPPC,DST and UDDS to verify the accuracy and robustness of the proposed algorithm.