Enhancing State of Charge Estimation for Lithium-Ion Batteries in Noisy Environments Using a Hybrid LSTM-EKF Mode

Lithium-ion battery State of Charge (SoC) calculation must be done accurately in order to maximize energy management in energy storage and electric car systems. SoC precision is impacted by issues like measurement noise and environmental influences, though. This research suggests a hybrid method to increase accuracy and robustness in noisy situations by combining Extended Kalman Filtering (LSTM-EKF) with Long Short-Term Memory (LSTM) networks. Through effective data fusion, the LSTM-EKF improves the model's capacity to manage noise, while the LSTM captures nonlinear battery dynamics. The hybrid model performs better than conventional techniques, according to experimental data, providing more dependable battery monitoring and effective energy utilization for electric mobility and sustainable energy.