Performance of Adaptive Neuro-Fuzzy Inference System State-of-Charge Estimation of Lithium-Ion Batteries for Electric Ship

Estimating the State of Charge (SoC) is a crucial part of a battery management system (BMS) for the electric ship. An Adaptive Neuro-Fuzzy Inference System (ANFIS) to predict the SoC using the ANFIS under different ambient temperature variations is proposed. A five-layer Artificial Neural Network for SoC estimation is used. The ANFIS achieves the lowest prediction error compared to the Neural Network, electrical circuit model, coulomb counting, Kalman Filter, Fuzzy Neural Network, and Support Vector Machine method.