Title :
Auto-tuning PID control using neural predictor to compensate large time-delay
Author :
Tan, Yonghong ; De Keyser, Robin
Author_Institution :
Dept. of Autom. Conrol, Gent Univ., Belgium
Abstract :
This paper presents a scheme introducing an auto-tuning PID controller with a neural network based Smith type predictor to control nonlinear systems with large time-delay. To model the dynamic nonlinear systems, a diagonal recurrent neural network is applied. A fast RLS type algorithm is employed to train the weights of the neural model on-line. Finally, the presented control method is used to control some nonlinear systems with large time-delay
Keywords :
delay systems; nonlinear control systems; predictive control; recurrent neural nets; recursive estimation; three-term control; tuning; RLS type algorithm; Smith type predictor; auto-tuning PID control; diagonal recurrent neural network; large time-delay compensation; neural predictor; nonlinear systems; Delay systems; Neural network applications; Nonlinear systems; Predictive control; Proportional control; Recurrent neural networks; Recursive estimation; Tuning;
Conference_Titel :
Control Applications, 1994., Proceedings of the Third IEEE Conference on
Conference_Location :
Glasgow
Print_ISBN :
0-7803-1872-2
DOI :
10.1109/CCA.1994.381315
