Artificial Intelligence Approach to State of Charge Estimation for Smart Battery Management Systems

Lithium-ion batteries have been used widely in multiple sectors due to their versatility. However, these batteries are expensive thus require intensive care to avoid any possible damage. Some of the major reasons that affect the health of the battery are overcharging and discharging. Therefore, it is essential to estimate the state of charge of the battery accurately and in a time efficient manner. This paper proposes deep learning approach to estimate the state of charge of simulated as well as experimental data. Through an extensive literature review, the proposed models to be used in this paper were GRU, LSTM, and FNN. Experimental data was generated using a laboratory setup where the proposed deep learning technique was trained and tested. The models used show fast and accurate results. This demonstrates the effectiveness and potential of the proposed artificial intelligence technique in state of charge estimation.