Sensorless Drive Diagnosis Data for Fault Detection

UCI Machine Learning Repository provides a dataset for sensorless drive diagnosis. The dataset contains 48 features derived from motor current signals to classify 11 different operating conditions. It is designed for benchmarking fault detection algorithms.

Use Cases

• Classify 11 distinct operating conditions using 48 derived features from motor current signals.
• Train a multi-class classifier to distinguish between healthy and various fault states without physical sensors.
• Benchmark feature selection methods on the 48 high-dimensional features for efficient model deployment.
• Develop anomaly detection models for early fault identification in electric drives.

Strengths

• Dataset contains 48 features for detailed analysis of drive conditions.
• Provides labeled data for 11 distinct classes for supervised learning tasks.

Limitations

• Specific sample size and temporal coverage are unknown.
• Data origin and collection methodology details are not provided.

Provenance

Source

UCI Machine Learning Repository

Collection Method

Features derived from motor current signals, likely from experimental setup.

Time Range

null

Freshness

null

Geography

null