Battery technology has advanced due to the need for dependable energy storage in systems like electric vehicles (EVs) and portable electronics. Evaluating battery conditions and improving EV autonomy require precise models and control strategies. Predicting battery performance, however, is difficult because of several variables, including temperature, deterioration, and usage patterns. Understanding battery performance requires a thorough understanding of time series analysis, especially forecasting. This research employs Artificial Neural Networks (ANN) configure to take into account the dynamic of the system, to create a lithium-ion battery (LIB) prediction model. The experimental tests were carried on a 59.2 VDC, 120 Ah LIB, monitoring the voltage and temperature while simulating charging and discharging. The developed model is a hybrid one, that uses clustering to create internal local models. The model forecasts the voltage evolution and achieves a Mean Squared Errors with values between 10 \({ }^{-3}\) and 10 \({ }^{-4}\) . The research concludes that hybrid models are suitable tools for predicting battery behavior, and the authors want to extend this model approach to predict temperature and State of Charge (SOC).