Lithium-ion Battery Model Parameter Identification Using Modified Adaptive Forgetting Factor-Based Recursive Least Square Algorithm

A proficient battery management system (BMS) is constantly expected to make an electric vehicle (EV) more dependable. The battery states like state of charge (SOC) and state of health (SOH) estimation are one of the significant functions of BMS. However, the accuracy of the model-based state estimation strategy is profoundly affected by the exhibition of the battery modeling approach. Particularly, in a continuous application, it is constantly needed to utilize a precise online battery model parameters identification algorithm. In this study, a modified adaptive forgetting factor-based recursive least square (MAFF-RLS) algorithm is proposed. Under which, the forgetting factor values are adaptively updated based on the model voltage error. To implement the proposed algorithm, the first-order RC battery model is utilized. The dynamic load profiles suitable for the EV environment are used for the validation of the proposed algorithm. Besides, to demonstrate the predominance of the MAFF-RLS algorithm over the RLS, and FFRLS algorithms, the estimated voltage errors such as Max AE, MAE ad RMSE are analyzed. The results demonstrated that the value of the estimated voltage RMSEs using the MAFF-RLS algorithm is lesser than of the voltage RMSEs using RLS and FFRLS algorithms.