Electric Vehicle PMSM Inverter Fault Detection with 9 Fault Classes

A pre-processed dataset for training models to detect faults in the inverters of electric vehicle Permanent Magnet Synchronous Motors (PMSM). The data is structured for classification tasks across nine distinct fault classes. The dataset originates from Kaggle, but specific author, organization, and temporal details are not provided.

Use Cases

• Train a multi-class classifier for inverter fault diagnosis based on the 9 fault classes mentioned.
• Develop anomaly detection models for PMSM inverter health monitoring.
• Benchmark time-series classification algorithms on pre-processed engineering sensor data.
• Research feature importance for different electrical and mechanical fault signatures in EV components.

Strengths

• Data is described as pre-processed and ready for model training, reducing initial preparation effort.
• The dataset is structured for a specific classification task involving 9 fault classes.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file format, and data size are unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Kaggle

Collection Method

Likely contains simulated or collected sensor data from electric vehicle PMSM inverter systems.

Time Range

null

Freshness

Last update date is unknown; freshness unverified.

Geography

null