Pattern-based fault diagnosis using neural networksAn architecture for a real-time pattern-based diagnostic expert system capable of accommodating noisy, incomplete, and possibly erroneous input data is outlined. Results from prototype systems applied to jet and rocket engine fault diagnosis are presented. The ability of a neural network-based system to be trained via the presentation of behavioral patterns associated with fault conditions is demonstrated.
Document ID
19890040231
Acquisition Source
Legacy CDMS
Document Type
Conference Paper
Authors
Dietz, W. E. (Tennessee Univ. Tullahoma, TN, United States)
Kiech, E. L. (Tennessee Univ. Tullahoma, TN, United States)
Ali, M. (Tennessee, University Tullahoma, United States)
Date Acquired
August 14, 2013
Publication Date
January 1, 1988
Subject Category
Cybernetics
Meeting Information
Meeting: International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems