BATTERY MANAGEMENT SYSTEM FOR ELECTRIC VEHICLES WITH AI/ML POWERED STATE OF HEALTH AND STATE OF CHARGE PREDICTION SYSTEM

Authors:

Suvansh Ritesh Malik, Dipesh

ABSTRACT

Electric Vehicles (EVs) have emerged as a sustainable alternative to conventional transportation systems due to increasing environmental concerns, rising fuel costs, and the global shift toward clean energy technologies. The performance, safety, and reliability of EVs are heavily dependent on the efficiency of their Battery Management Systems (BMS). Accurate estimation of battery parameters such as State of Charge (SOC) and State of Health (SOH) is essential for ensuring optimal battery utilization, enhanced lifespan, and operational safety.

This thesis presents the design and implementation of a low-cost intelligent Battery Management System prototype using an ESP32 microcontroller, INA219 current sensor, ADS1115 high-resolution ADC, thermistor-based temperature monitoring, and machine learning-assisted SOH prediction techniques. The proposed system continuously monitors voltage, current, temperature, and battery health parameters in real time. A web-based HTML dashboard is developed for live visualization of battery metrics, graphical analysis, and fault indication.

The system incorporates protection mechanisms including overvoltage, undervoltage, overcurrent, and thermal cutoff using MOSFET-based switching and buzzer alerts. Experimental data logging is performed using a Micro SD card module, while AI/ML-based prediction models are developed using publicly available battery degradation datasets.

The prototype demonstrates a scalable and modular architecture suitable for educational EV research, embedded system applications, and intelligent battery diagnostics. The proposed design offers a cost-effective proof-of-concept platform for future advancements in smart electric mobility systems.

KEYWORDS

Battery Management System, Electric Vehicle, State of Charge, State of Health, ESP32, Machine Learning