<![CDATA[The growing demand for energy around the world is driving the development of renewable resources, and batteries are the primary choice for energy storage. To carry out effective energy management, State of Charge (SOC) estimation of Lithium-ion batteries is essential. The development of an Adaptive Neuro-Fuzzy Inference System (ANFIS) model for SOC estimation using LG 18650 HG2 battery dataset is the objective of this research. It is tested with two parameters, namely two inputs consisting of voltage and current; and three inputs consisting of voltage, current, and temperature. The shape of the membership function, number of nodes, and epochs are some of the indicators tested to find the best configuration. The results show that the three-input configuration with generalized-bell membership function (Gbell MF), five nodes, and 100 epochs has the smallest Root Mean Square Error (RMSE), which is 0.0317, compared to the best two-input configuration, which has an RMSE of 0.0527. Since the three-input configuration takes longer to train, further improvements are needed for real-time implementations such as in electric vehicle battery management systems.]]>
Copyrights © 2025
