Comprehensive dataset for fault detection and diagnosis in inverter-driven permanent magnet synchronous motor systems

This work introduces a new, comprehensive dataset for Fault Detection and Diagnosis (FDD) in inverter-driven Permanent Magnet Synchronous Motor (PMSM) systems. Despite the increasing significance of AI-driven FDD techniques, the domain suffers from a lack of publicly accessible, real-world datasets for algorithm development and evaluation. Our contribution fills this gap by offering a comprehensive, multi-sensor dataset obtained from a bespoke experimental apparatus. The dataset includes different fault cases, such as open-circuit faults, short-circuit faults, and overheating conditions in the inverter switches. The dataset incorporates 8 raw sensor measurements and 15 derived features, recorded at 10 Hz, amounting to 10,892 samples across 9 operational conditions (one normal, eight fault types). By keeping this dataset publicly accessible, we seek to accelerate research in AI-driven fault identification and diagnosis for electric drive systems.