An unscented Kalman filter in designing dynamic GMDH neural networks for robust fault detection.

Marcin Mrugalski

Published in: Int. J. Appl. Math. Comput. Sci. (2013)

Keyphrases

neural network
unscented kalman filter
robust fault detection
dynamic environments
artificial neural networks
genetic algorithm
neural network model
group method of data handling
real time
back propagation
kalman filter
constraint satisfaction
constant velocity
feature space
fault diagnosis