Research On State Estimation Method Based On Equivalent Temperature Of Lithium-ion Batteries

In recent years,new energy vehicles have developed rapidly,and lithium-ion batteries,as one of their most important components,have the characteristics of large capacity and high power,so it is necessary for battery management system to monitor and estimate the battery state in real time to ensure the normal operation and safety of the battery.However,in the current battery state estimation method,the battery model parameters are usually identified by a large number of offline experiments,and self-correction cannot be performed during battery operation.In addition,under the condition of low temperature and high current rate,the real temperature of the battery is difficult to obtain due to the heating of the battery itself.As an alternative,the surface temperature or ambient temperature of the battery is used for calculation,which leads to large errors in battery state estimation.These problems need to be solved urgently.Aiming at the problems existing in current lithium ion battery state estimation methods,this paper is based on the relationship between battery charge transfer resistance and battery real temperature,and introduces state of charge(SOC)and current rate correction factors to make it suitable for practical conditions.Then it is substituted into the second-order RC equivalent circuit model of lithium ion batteries,and the dual extended Kalman filter algorithm is used to realize the joint on-line estimation of the battery equivalent temperature and SOC.The SOC estimation results of the proposed battery equivalent temperature correction are validated under wide temperature(298K,273K,263K)and different current rate DST conditions.Compared with the traditional SOC estimation method using ambient temperature or battery surface temperature,the accuracy of SOC estimation in this paper is obviously improved,and the error is within ±2%.Especially in the case of low temperature and high current rate DST,the traditional method SOC estimation error increases significantly with discharge,and the SOC estimation result of this method is always extremely accurate.In addition,the joint online estimation results of battery equivalent temperature and SOC are applied to the estimation of battery power state(SOP).The SOP prediction of batteries is realized by using the circuit model modified by the equivalent temperature of batteries and the extended Kalman filter algorithm,which has a very high accuracy in a wide temperature range.