Title :
Fault modelling using a mixture of conditional Gaussian Transitions
Author :
Dejan P. Jovanović;Ross S. McVinish;Philip K. Pollett
Author_Institution :
Department of Mathematics, The University of Queensland, 4072 AUSTRALIA
fDate :
6/1/2011 12:00:00 AM
Abstract :
To model a fault that can be caused by more than one source, a mixture of conditional Gaussian transitions is proposed. The conditional means are modelled by recurrent neural networks. An expectation-maximization (EM) algorithm is used to estimate model parameters. By grouping known types of faults it is possible to form a bank of different fault models.
Keywords :
"Mathematical model","Markov processes","Recurrent neural networks","Predictive models","Time series analysis","Training","Equations"
Conference_Titel :
Control & Automation (MED), 2011 19th Mediterranean Conference on
Print_ISBN :
978-1-4577-0124-5
DOI :
10.1109/MED.2011.5983194
