Optimizing Electric Vehicle Battery Performance: A Comparative Analysis of ANFIS and AUKFM for SOC and SOH Estimation

The rising prevalence of electric vehicles (EVs) underscores the need for effective and dependable strategies concerning the charging and discharging of the EV batteries. This proposal offers a comprehensive overview of diverse methodologies and techniques utilized for optimizing EV battery charging and discharging. Among the various approaches discussed, particular emphasis is placed on two prominent methods: the Adaptive Neuro Fuzzy Inference System (ANFIS) and the Adaptive Unscented Kalman Filter Method (AUKFM). ANFIS employs fuzzy logic to dynamically regulate charging and discharging parameters based on battery conditions and external factors. Conversely, AUKFM employs a recursive estimation algorithm to gauge battery status and fine-tune the charging and discharging processes. Both methodologies have exhibited promising outcomes in augmenting battery performance, extending battery lifespan, and attaining optimal energy management. By applying these two techniques for SOC and SOH estimation, the relative efficiency is attained by analyzing simulation results and observations.