State of charge estimation for a group of lithium-ion batteries using long short-term memory neural network

The present paper estimates for the first time State of Charge (SoC) a high capacity grid-scale lithium-ion battery storage system used to improve power profile in distribution network. proposed long short-term memory (LSTM) neural network model can overcome problems associated with nonlinear and adapt complexity uncertainty estimation process. accuracy developed was compared results obtained from Feed-Forward Neural Network (FFNN) topology Deep-Feed-Forward (DFFNN) under three different series. trained using real data Al-Manara PV plant. LSTM learn-and-adapt-to-train-date properties, as well idea “forget gate,” shows exceptional ability determine SoC various ID data. properly calculated all three-time models maximum standard error (MSE) less than 0.62%, while FFNN DFFNN provided fair estimate MSEs 5.37 9.22% 4.03 7.37%, respectively. promising lead excellent monitoring control management systems.