Real-Time Monitoring and Optimization of Two-Wheeler Battery Life

Electric two-wheelers have gained popularity because of their eco-friendly nature and their ability to decrease urban emissions. Yet, battery degradation is still a serious challenge, greatly influencing the efficiency, performance, and life of these vehicles. The current research work proposes the design of a real-time battery monitoring system that integrates machine learning methods to forecast battery health and enhance charging habits. The system makes use of voltage, current, and temperature sensors in monitoring the conditions under which a battery operates constantly. The information is displayed using an LCD interface while also sent to a nearby processing server to be analyzed. The battery life is predicted by the use of a K-Nearest Neighbors (KNN) algorithm, which had an accuracy of around 83 % based on historical as well as real-time information. The model has been trained using a dataset with 2,600 entries characterizing key battery parameters, enhanced by another 100 data points centered on battery status and voltage. The outlined system has the potential to enhance battery reliability, lower maintenance, and prolong the lifespan of the battery, ensuring the sustainable accommodation of electric cars in contemporary mobility systems.