IoT-based Battery Health Monitoring and Its Remaining Useful Life Prediction using Artificial Neural Network

Electric vehicles nowadays face issues with battery downgrading and sometimes explosions. The predictive maintenance and preventive alerts for hazardous situations regarding batteries will help the user take necessary actions. In this study, the condition of a Lithium-ion battery was monitored on the ThingSpeak cloud platform using an Internet of Things setup. An Artificial Neural Network (ANN) model was developed using NASA's battery dataset for discharge cycles. The model was used for predicting the Remaining Useful Life (RUL) of a battery at its current operating cycle number. The study showed degradation in battery capacity over a no, of cycles. The trained ANN model was able to predict the RUL of a battery before its End of Life (EoL).