Multi-source Thermal and Performance Data for Marine Diesel Engine Fault Diagnosis

A multi-source dataset for thermal fault diagnosis and predictive health management of marine diesel engines. It comprises four subsets including simulation, bench-measured, real-vessel telemetry, and superimposed fault data, covering 24-dimensional parameters across one healthy and ten fault modes. The dataset was created by Zaimi Xie and last updated on April 13, 2026.

Use Cases

• Developing multi-source information fusion algorithms based on the four distinct data subsets.
• Benchmarking noise-robust feature extraction methods based on data with industrial background noise and sea-state disturbances.
• Training unsupervised domain adaptation models based on the virtual-real fusion fault data.
• Building diagnostic classifiers for ten thermal fault modes based on 24-dimensional gas-path and in-cylinder parameters.

Strengths

• Includes four distinct data subsets (simulation, bench-measured, real-vessel, and fault data) for method validation.
• Covers 24-dimensional high-frequency transient and steady-state performance parameters.
• Encompasses one healthy state and ten representative thermal fault modes for comprehensive testing.
• Simulation model was calibrated against real-vessel telemetry to ensure physical fidelity.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is small at 688.1 KB, indicating limited scope.

Provenance

Source

figshare

Collection Method

Data comprises calibrated simulation, bench measurements, real-vessel telemetry, and virtual-real fusion fault generation.

Time Range

null

Freshness

Last updated 2026-04-13 10:27:44; freshness should be verified.

Geography

null