Title :
On-line fault detection in multi-output systems using Kalman filter and neural network
Author :
Chowdhury, Fahmida N.
Author_Institution :
Dept. of Electr. Eng., Michigan Technol. Univ., Houghton, MI, USA
fDate :
29 June-1 July 1994
Abstract :
Describes a hybrid scheme for fault detection in multi-output systems. The scheme consists of a Kalman filter and a three-layer neural network. The neural net mimics a team of experts with several ´consultants´ and a ´leader´. The idea is tested on a three-phase voltage model, and simulation results are presented.
Keywords :
Kalman filters; decision theory; electrical engineering; electrical engineering computing; fault diagnosis; fault location; multivariable control systems; neural nets; stochastic systems; Kalman filter; consultants; hybrid scheme; leader; multi-output systems; neural network; online fault detection; three-phase voltage model; Decision making; Electrical fault detection; Fault detection; Fault diagnosis; Intelligent networks; Neural networks; Neurons; Stochastic systems; Testing; Voltage;
Conference_Titel :
American Control Conference, 1994
Print_ISBN :
0-7803-1783-1
DOI :
10.1109/ACC.1994.752368
