Title :
Fault detection for jump discrete systems
Author :
Tzes, Anthony ; Le, Ke
Author_Institution :
Dept. of Mech. Eng., Polytech. Univ., Brooklyn, NY, USA
Abstract :
The development of a fault detection scheme for linear jump discrete systems is addressed. The system´s parameter vector is identified through a sliding-window orthotopic set membership identification scheme. A recursive algorithm predicts the interval of the system´s output based on the identified parameter bounds. The fault is detected when the actual system output is not within the predicted `worst case´ interval output, or when the actual parameter vector is not within the identified interval. Simulation and experimental studies are offered to illustrate the scheme´s performance
Keywords :
discrete time systems; fault diagnosis; identification; linear programming; linear systems; stochastic systems; SISO systems; fault detection; identification; jump discrete systems; linear programming; linear systems; orthotopic set membership; parameter bounds; recursive algorithm; Fault detection; Fault diagnosis; Filters; Gaussian noise; Mechanical engineering; Noise measurement; Prediction algorithms; Technological innovation; Uncertainty; Vectors;
Conference_Titel :
American Control Conference, 1999. Proceedings of the 1999
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-4990-3
DOI :
10.1109/ACC.1999.786436