Combined SoC and SoE Estimation of Lithium-ion Battery using Multi-layer Feedforward Neural Network

Abstract

The estimation of state of charge (SoC) and state of energy (SoE) serves the premise of an efficient Battery Management System(BMS). The estimation technique should be able to capture the dynamics the battery is subjected to, along with its inherent non-linear behaviour. This study proposes a combined SoC and SoE estimation framework using multi-layer feedforward neural network. The experimental results validate the higher accuracy and robustness of the proposed method under dynamic driving and temperature conditions. The Mean Square Error(MSE) obtained during the testing of the algorithm with various drive cycles is found to be quite promising.