By Topic

An efficient Novelty Detector for online fault diagnosis based on Least Squares Support Vector Machines

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)

A paradigm shift in the standard operating procedures (SOP) is underway in the reliability and health management industry. As the community transitions from traditional preventive maintenance procedures to modern predictive or health-based management systems, areas such as efficient online monitoring and diagnosis schemes based on real-time observations have emerged as key research subjects for engineers. Most diagnostic systems require data from both healthy and faulty conditions in order to properly train their classification algorithms. However, in many situations, normal signals are acquired easily while fault samples are difficult to be gained. In this paper, we present a diagnosis scheme based on the least squares support vector machines (LS-SVM). Our modified LS-SVM algorithm is a one-class novelty detector which can differentiate between a normal and faulty condition based only on the normal samples which are easily available. We diagnose a growing crack fault on a planetary gear plate mounted aboard a UH-60 Blackhawk aircraft using this approach. Comparisons drawn with other contemporary approaches lean favorably towards the viability of the suggested novelty detector.

Published in:


Date of Conference:

8-11 Sept. 2008