Title :
An investigation into the feasibility of using neural networks to control turbogenerators
Author :
Shepstone, NM ; Harley, RG ; Jennings, G. ; Rodgerson, J.
Author_Institution :
Dept. of Electr. Eng., Natal Univ., Dalbridge, South Africa
Abstract :
This paper reports on the feasibility of using an online learning neural network as an adaptive turbogenerator controller to replace the automatic voltage regulator and the turbine governor. Results are presented which show that the neural network can control the turbogenerator at least as well as the conventional controller, but more importantly, its performance does not degrade when system conditions change
Keywords :
learning (artificial intelligence); machine control; neurocontrollers; power engineering computing; turbogenerators; adaptive turbogenerator controller; neural networks; online learning neural network; turbogenerators control; Adaptive control; Adaptive systems; Automatic control; Automatic voltage control; Control systems; Neural networks; Programmable control; Regulators; Turbines; Turbogenerators;
Conference_Titel :
AFRICON, 1996., IEEE AFRICON 4th
Conference_Location :
Stellenbosch
Print_ISBN :
0-7803-3019-6
DOI :
10.1109/AFRCON.1996.563003
