Comparative study of Battery Fault Detection using Machine Learning Algorithm

This Paper shows comparative study of individual cell fault detection in battery pack by predicting the fault using real time data from parameter of SOC of each cell that is present in the battery pack. Compare the real time data with the predicted dataset using different Machine learning algorithm such as K-Nearest Neighbours and Random Forest algorithm. The NodeMCU acts as a controller which retrieves the data from voltage sensor of individual cell in the battery pack and given to the cloud. The cloud stores pretrained algorithm of Machine Algorithm Model and dataset of predicted value.by comparing the real time data from the controller and predicted data from dataset from the confusion matrix which shows the accuracy of the algorithm and also results the faulty cell in the cloud the data is returned to the controller and also displays the faulty cell in the LCD display which is connected to controller.