A Sensor-Fault-Estimation Method for Lithium-Ion Batteries in Electric Vehicles

Area of Science:

• Electrical Engineering
• Materials Science
• Automotive Engineering

Background:

• Electric vehicles (EVs) and lithium-ion batteries (LIBs) are rapidly advancing.
• Battery safety and reliability are critical concerns for LIBs.
• Battery management systems (BMS) rely on accurate sensor data for optimal performance.

Purpose of the Study:

• To develop validated electrical and thermal models for LIBs.
• To establish a model-based fault diagnosis scheme for battery sensors.
• To achieve early-stage detection of sensor faults in LIBs.

Main Methods:

• Utilized descriptor proportional and derivative observer systems for sensor diagnosis.
• Developed validated electrical and thermal models of LIBs using experimental data.
• Implemented real-time estimation of voltage, current, and temperature sensor faults.

Main Results:

• Successfully simulated various fault types to verify the proposed diagnosis scheme.
• Demonstrated the capability of the observer systems to estimate sensor faults in real-time.
• Validated the effectiveness of the model-based approach using experimental battery data.

Conclusions:

• The proposed model-based fault diagnosis scheme effectively detects sensor faults in LIBs.
• Early fault detection enhances the safety and reliability of battery management systems.
• This approach contributes to the advancement of safer and more dependable electric vehicles.