Estimation Capacity Of Ternary Lithium Battery Based On Extended Kalman Filter

In order to ensure the safety of electric vehicle operation,reliability,long-term,power battery performance accurate detection becomes urgent,especially to the actual capacity detection,which is of great significance for its research.In view of the traditional battery capacity detection method is complex and time-consuming problem.Based on the relationship of different aging curves,and the connection between the charged data segment and the complete charge date,an extended Kalman filter estimation method which is convenient and fast is proposed.The characteristics of the battery and charging curve are introduced.Taking the degree of capacity decay as a criterion,the characteristic extraction scheme is designed in paper,and the trend of the inflection points under the aging characteristic are analyzed.By comparing the charging curve under different aging,the charging curve has nothing to do with aging is proved,and the charging cutoff time is shortened along with the charging curve changing in parallel.On this basis,the method of using the relationship of different aging curves to update the curve model is proposed.Paper fully considers the influence of the charging magnification,the sampling interval and the temperature on the charging curve.By analyzing the curve inflection point,the estimation condition test scheme is designed.At the same time,the intrinsic relationship between the charging data segment and the complete charging curve is explored,then charging data segment is a small fragment of the complete charging curve is concluded.A variety of charging curve models are compared,and the Gaussian fifth order model is selected as the charge curve model.At the end,an extended Kalman filter estimation method is prop osed.Based on the battery performance tester on the ternary lithium battery for long cycle charge and discharge testing,the capacity of the experimental data is estimated.With the capacity estimation error chart,it is concluded that when the actual capacity is greater than 90% of the rated capacity,the estimation error can be controlled within 5%.In the charging curve platform area,the true voltage and the estimated value can be controlled at 0.01 V,which has guiding significance on the online estimation.