Data-Driven Detection of Physical Faults and Cyber Attacks in Dual-Motor EV Powertrains

In the last decades, with the pervasive utilization of digital control units and communication networks in modern electric vehicle powertrains, such safety-critical systems have become highly vulnerable to potential cyber threats. Current research primarily focuses on aggressive attacks, which usually cause drastic changes and disturbances to the systems. However, little research has addressed how to detect more stealthy attacks targeting electric vehicle powertrains and distinguish between such attacks and common physical faults. This paper bridges this gap by proposing a data-driven approach to detecting and diagnosing hidden attacks and common physical faults in the dual-motor electric vehicle powertrain. The proposed method achieves promising performance in detecting and diagnosing cyber-attacks and physical faults. It reaches an accuracy of nearly 100% on detecting anomalies and above 90% on distinguishing stealthy attacks from common physical faults.