Simulated Turbofan Engine Degradation Data from NASA C-MAPSS
Updated 10mo ago
Available on 1 platform
Sign in to view source links and access this dataset
Description
C-MAPSS Aircraft Engine Simulator Data contains simulated flight recordings for commercial turbofan engines under nominal and fault conditions. The data was generated by NASA using the C-MAPSS (Commercial Modular Aero-Propulsion System Simulation) Matlab Simulink tool, with each flight sampled at 1 Hz and consisting of 30 engine and flight condition parameters. The intent is to identify when and in which flight an injected fault occurred.
Use Cases
Train anomaly detection models on 30 engine and flight condition parameters to identify the flight and timestamp of injected faults.
Analyze health indicators, measurement temperatures, and pressure measurements to predict engine degradation over simulated flights.
Build time-series classifiers using 1 Hz sampled flight condition data to distinguish between faults in the fan, low/high pressure compressor, or low/high pressure turbine.
Strengths
Data generated by a high-fidelity system-level simulation tool (C-MAPSS) from NASA Glenn Research Center.
Each flight recording contains 30 distinct engine and flight condition parameters sampled at 1 Hz.
Faults are simulated across five critical engine components: fan, low/high pressure compressor, and low/high pressure turbine.
Limitations
Dataset is currently unavailable for download pending a management review at Glenn Research Center.
Data is entirely simulated, not from physical engines, which may limit real-world model generalization.
Specific row count, file formats, and sample data details are unknown from the provided input.
Provenance
Source
National Aeronautics and Space Administration (NASA)
Collection Method
Generated using the C-MAPSS (Commercial Modular Aero-Propulsion System Simulation) Matlab Simulink tool.
Time Range
null
Freshness
null
Geography
null
The dataset is currently unavailable for download. A review by Glenn Research Center management is ongoing, and updates will be posted on the source website. Contact Jonathan Litt ([email protected]) for questions.