Recursive Parameter Identification of Lithium-Ion Battery for EVs Based on Equivalent Circuit Model

Battery modeling is one of the critical parts of the development of battery management systems (BMS). A good model on the one hand can track the dynamics of the battery and thereby facilitate the energy management, and on the other hand is very helpful for state estimation algorithm design. So far the most popular and successful model is the equivalent circuit model (ECM). However, in real applications, the model parameters are time-variant, and changing with temperature, current, SOC and ageing process, this affects the performance of the model dramatically. Therefore, to enhance the performance of the ECM, an online model parameter identification technique is developed based on the recursive least square (RLS) algorithm. The ECM used in this study consists of three resistors, two capacitors and one SOC-controlled voltage source. Detailed information on the parameter initialization, the online RLS based parameter identification process and validation experiments are discussed. With two different carefully designed experiments, the factors affecting the performance of the algorithm are analyzed as well. Testing results with real vehicular current profile show that, the proposed technique can identify the model parameters online with a high accuracy, which makes the ECM more adaptive in EV applications.