Title :
On-line training of adaptive neural network controllers
Author :
Lima, A.M.N. ; Cavalcanti, J.H.F. ; Deep, G.S.
Author_Institution :
Dept. de Engenharia Eletrica, Univ. Federal da Paraiba, Joao Pessoa, Brazil
Abstract :
In this paper, the experimental results obtained with a new algorithm for online training of multilayer neural nets for adaptive speed control of a DC motor are presented. The effect of the number of neurons in the hidden layer on the duration of training phase is discussed. The proposed algorithm does not require any off-line training for the nets
Keywords :
DC motors; adaptive control; angular velocity control; learning (artificial intelligence); machine control; neurocontrollers; DC motor; adaptive neural network controllers; adaptive speed control; multilayer neural nets; online training; Adaptive control; Adaptive systems; Artificial neural networks; Digital control; Jacobian matrices; Neural networks; Neurons; Performance gain; Programmable control; Velocity control;
Conference_Titel :
Industrial Electronics, Control and Instrumentation, 1994. IECON '94., 20th International Conference on
Conference_Location :
Bologna
Print_ISBN :
0-7803-1328-3
DOI :
10.1109/IECON.1994.397998
