DocumentCode :
2133213
Title :
Dynamic modelling using a multiple neural network architecture
Author :
Rivas, Carlos ; Willis, M.J. ; Peel, C. ; Hartmann, G.
Author_Institution :
Newcastle upon Tyne Univ., UK
Volume :
2
fYear :
1994
fDate :
21-24 March 1994
Firstpage :
977
Abstract :
A multiple neural network architecture has been introduced. The methodology makes use of the ´K´ means clustering algorithm in order to differentiate between different process operating regions thus allowing an improved utilisation of the training data set. Indeed the ability of the technique to focus on specific operating regions appears to enhance characterisation of process behaviour when compared to a single network trained over a large region. An additional advantage of the technique, when compared to the standard feedforward ANN, is that the reliability of the network prediction may be monitored. Finally the advantages of the proposed technique are highlighted by application to two simulated nonlinear systems: a binary distillation column; and a continuous stirred tank reactor.
Keywords :
chemical technology; feedforward neural nets; nonlinear control systems; pattern recognition; K means clustering algorithm; binary distillation column; continuous stirred tank reactor; dynamic modelling; feedforward ANN; multiple neural network architecture; network prediction; process behaviour; process operating regions; reliability; simulated nonlinear systems;
fLanguage :
English
Publisher :
iet
Conference_Titel :
Control, 1994. Control '94. International Conference on
Conference_Location :
Coventry, UK
Print_ISBN :
0-85296-610-5
Type :
conf
DOI :
10.1049/cp:19940267
Filename :
327333
Link To Document :
بازگشت