NASA Logo, National Aeronautics and Space Administration

+NASA Home

+Ames Home

Chetan Kulkarni and External Partners Release New Turbofan Engine Degradation Simulation Dataset
Intelligent Systems Division Banner

Chetan Kulkarni and External Partners Release New Turbofan Engine Degradation Simulation Dataset

Intelligent Systems Division Diagnostics and Prognostics group and Prognostics Center of Excellence (PCoE) member Chetan Kulkarni, along with external partners Manuel Arias Chao and Olga Fink (ETH Zurich) and Kai Goebel (PARC), have released a new Turbofan Engine Degradation Simulation Dataset. The dataset was generated with the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dynamical model. A previously released engine degradation dataset has more than 65K downloads and more than 30 publications comparing different fault diagnostics and prognostics algorithms.

The generation of data-driven prognostics models requires significant run-to-failure data. The newly released dataset provides synthetic run-to-failure degradation trajectories of a small fleet comprising nine turbofan engines with unknown and different initial health conditions. Real flight conditions as recorded onboard a commercial jet were taken as input to the C-MAPSS model. The diversity of perspectives from the inter-organization team supported the creation of a robust and complete dataset incorporating a diverse set of failure modes relevant to the broader Prognostics & Health Management (PHM) community.

The damage propagation modelling used for the generation of this synthetic dataset builds on the modeling strategy from previous work done within the group. Two distinctive failure modes are present in the available Dataset (D). Units 2, 5, and 10 have failure modes of an abnormal High-Pressure Turbine (HPT) efficiency degradation. Units 16, 18, and 20 are subject to a more complex failure mode that affects the Low-Pressure Turbine (LPT) efficiency and flow in combination with the HPT efficiency degradation. Test units are subjected to the same complex failure mode.

The new Turbofan Engine Degradation Simulation Dataset is available as dataset #17 on the PCoE Website.

BACKGROUND: The Prognostics Data Repository is a collection of open-source datasets that have been generated within the PCoE or donated by various universities, agencies, and companies. The data repository focuses exclusively on prognostic data sets, i.e., data sets that can be used for development of prognostic algorithms and compared with algorithms developed by other researchers. Mostly these are data time series from some nominal state to a failed state. The collection of data in this repository is an ongoing process. Some of the datasets have been used for Data Challenge competitions at the annual PHM conferences. Publications making use of databases obtained from this repository are requested to acknowledge both the assistance received by using this repository and the data donors. This will help others to obtain the same data sets and replicate experiments. There have been more than 250 publications from researchers in the area of prognostics using these open-source datasets.

NASA PROGRAM FUNDING: System Wide Safety (SWS) project, Aeronautics Research Mission Directorate (ARMD)

TEAM: Chetan S. Kulkarni, Manuel Arias Chao & Olga Fink (ETH Zurich), and Kai Goebel (PARC)

POINT OF CONTACT: Chetan S. Kulkarni,

First Gov logo
NASA Logo -