NASA Logo, National Aeronautics and Space Administration
Intelligent Systems Division Banner

Prognostics Posters at Data Mining Conference

Researchers from the Prognostics Center of Excellence will present three posters at the Conference on Intelligent Data Understanding to be held September 9–10 at NASA Headquarters in Washington, DC.

  • B. Goebel, S. Saha, A. Saxena, and J. Celaya, “Evaluating Data-driven Prognostic Techniques”
Abstract: Traditional damage propagation algorithms rely on physics-based failure mechanisms. Alternatively, one can employ data-driven approaches when sufficient test data are present that map out the damage space. In this investigation, different algorithms are evaluated for their suitability in these situations. Specific interest is in assessing the trade-offs that arise from the amount of data needed, the computational speed exhibited, the robustness of the algorithm to input space perturbations, the ability to support uncertainty management, and the accuracy of the predictions.
  • A. Saxena and K. Goebel, “Generating Simulation Data for Turbofan Engine Degradation using CMAPSS for Prognostics Algorithm Development”
Abstract: CMAPPS is a turbofan engine model (developed at NASA Glenn) that has capabilities to simulate various fault conditions. Several sets of degradation profiles have been simulated with increasing level of difficulty to progressively simulate realistic scenarios. The first release of this dataset was done for the PHM Data Challenge at the upcoming PHM08 conference. Results from participants will be evaluated using a standardized procedure. This will help evaluate existing prognostic algorithms available in the research community on a common ground. These datasets are now accessible through the conference web site.
  • K. Goebel, A. Saxena, J. Celaya, E. Balaban, B. Saha, and S. Saha, “Representing the Prognostics Problem: A Notational Framework”
Abstract: There is significant disagreement on prognostics definitions and evaluation metrics. Requirements are different for different applications and hence a common consensus on evaluation standards has not been reached. A comprehensive review and classification of such applications will help set standardized procedures among the community. This will help define validation standards for prognostics technologies to aid in their fielded applications.

COLLABORATORS: Kai Goebel (NASA), Edward Balaban (NASA), Jose Celaya (RIACS), Bhaskar Saha (MCT), Sankalita Saha (RIACS), Abhinav Saxena (RIACS

NASA PROGRAM FUNDING: ARMD/AVSP/IVHM

Contact: Kai Goebel

09/09/2008

+NASA Home

+Ames Home

Prognostics Posters at Data Mining Conference
+ Home + Organization + News + Research Areas + Publications + Software + Internal Systems Help
First Gov logo
NASA Logo - nasa.gov