The project introduces a real-time multi-sensor fusion framework combining temperature, voltage, current, pressure, and State-of-Health sensors to detect early faults in EV lithium-ion batteries. It uses fusion algorithms and machine learning to improve detection accuracy (96.8%) and response time (<120 ms), enhancing battery safety and lifespan.

• Design real-time multi-sensor monitoring framework

• Develop fusion algorithm for accurate battery health

• Detect early faults: voltage imbalance, short circuits, thermal hotspots, gas generation, swelling

• Use signal processing & machine learning for anomaly detection

• Provide early-warning alerts and preventive actions

• Reduce false alarms with multi-sensor data

• Support fault-response mechanisms (load reduction, cooling, shutdown)

• Extend battery lifespan via predictive diagnostics

• Create scalable, cost-effective framework

• Validate system with simulations and experiments

1. Voltage Sensors: Monitor cell & pack voltage for faults

2. Current Sensors: Detect current spikes & load behavior

3. Temperature Sensors: Track heat, hotspots, prevent thermal runaway

4. Pressure Sensors: Sense gas build-up & swelling

5. Gas Sensors: Detect electrolyte vapors indicating chemical faults

6. Strain/Stress Sensors: Monitor mechanical deformation & structural stress