State Of Charge Estimation for Lithium-Ion Battery Using Evolving Local Model Network

this paper proposes a novel method for accuracy estimation of the state of the charge of lithium-ion batteries adapted for Electric Vehicles (EVs). The proposed method is based on Evolving Local Model Network (ELMN), which is a new reconcilable method for time-varying dynamic processes. The main advantages of the proposed method are implementing a high accurate model and its high reliability in various operation conditions. The performance of the proposed method is evaluated based on Root Means Square Error (RMSE) and also Mean Absolute Percentage Error (MAPE). The proposed method is simulated in MATLAB/SIMULINK environment. The simulation results confirm that the accuracy of the proposed ELMN approach is significantly high in comparison with other well-known approaches adapted for this application such as Local Model Network (LMN) and the Adaptive Nero-Fuzzy Inference System (ANFIS).