NTRS - NASA Technical Reports Server

Back to Results
Robust Fault Detection for Aircraft Using Mixed Structured Singular Value Theory and Fuzzy LogicThe purpose of fault detection is to identify when a fault or failure has occurred in a system such as an aircraft or expendable launch vehicle. The faults may occur in sensors, actuators, structural components, etc. One of the primary approaches to model-based fault detection relies on analytical redundancy. That is the output of a computer-based model (actually a state estimator) is compared with the sensor measurements of the actual system to determine when a fault has occurred. Unfortunately, the state estimator is based on an idealized mathematical description of the underlying plant that is never totally accurate. As a result of these modeling errors, false alarms can occur. This research uses mixed structured singular value theory, a relatively recent and powerful robustness analysis tool, to develop robust estimators and demonstrates the use of these estimators in fault detection. To allow qualitative human experience to be effectively incorporated into the detection process fuzzy logic is used to predict the seriousness of the fault that has occurred.
Document ID
Document Type
Conference Paper
Collins, Emmanuel G. (Florida Agricultural and Mechanical Univ. Tallahassee, FL United States)
Date Acquired
August 20, 2013
Publication Date
August 1, 2000
Publication Information
Publication: HBCUs/OMUs Research Conference Agenda and Abstracts
Subject Category
Quality Assurance and Reliability
Distribution Limits
Work of the US Gov. Public Use Permitted.

Related Records

IDRelationTitle20010012143Analytic PrimaryHBCUs/OMUs Research Conference Agenda and Abstracts