Data Acquisition of Battery Variables and Estimation of Battery State of Health

Batteries are the most crucial part of the many electrical devices. In transportation sector, Electric Vehicles (EV) are powered by batteries which degrade over the charge discharge cycling. So it is important to keep track of battery variables to monitor the health of the batteries. Acquiring the data of battery variables (such as voltage, current and temperature) can help in identifying potential issues with battery and optimize its performance. Estimating the State of Health (SOH) of EV batteries is crucial for ensuring the efficient and reliable operation of the vehicle. In this work, firstly, we have developed an Arduino based data acquisition system for the battery variables by integrating Arduino with voltage, current and temperature sensors. Through the developed hardware, We have been able to collect real-time data of different variables of a battery. Apart from data acquisition, it is successful in identifying abnormalities in the system like excess current and perform control actions for the safe operation of the battery. Secondly, using publicly available battery degradation data from NASA Prognostic Center of Excellence (PCoE) data repository, we developed a machine learning model that can estimate the battery SOH. The model takes time stamped values of voltage, current and temperature of the battery and provides its SOH. We compared Polynomial Regression, Support Vector Regression, Ridge Regression, Decision Tree Regression and Lasso Regression based algorithms. Although all the algorithms achieved over 90% accuracy, the Polynomial Regression algorithm outperforms others in terms of Mean Absolute Error, Root Mean Squared Error and R2 Score.