DocumentCode :
288681
Title :
Adaptive control with NeuCOP, the neural control and optimization package
Author :
Graettinger, Timothy J. ; Bhat, Naveen V. ; Buck, Jeffrey S.
Author_Institution :
NeuralWare Inc., Pittsburgh, PA, USA
Volume :
4
fYear :
1994
fDate :
27 Jun-2 Jul 1994
Firstpage :
2389
Abstract :
NeuCOP is the neural control and optimization package that has been jointly developed by Texaco and NeuralWare. It is a state-of-the-art, multivariable, adaptive controller that combines the nonlinear modeling power of neural networks with nonlinear optimization algorithms. The NeuCOP controller belongs to the general class of model predictive controllers. One novel feature of the NeuCOP controller is its use of a nonlinear, neural network process model. We describe the identification subsystem that has been developed. More specifically, we address the issue of system re-identification, after the system is put online. The re-identification process allows the model to adapt to changing process conditions
Keywords :
adaptive control; identification; multivariable control systems; neural nets; neurocontrollers; optimisation; predictive control; software packages; NeuCO; NeuralWare; Texaco; identification; model predictive controllers; multivariable, adaptive controller; neural control; neural networks; nonlinear modeling; optimization; Adaptive control; Economic forecasting; Industrial control; Neural networks; Packaging; Power generation economics; Predictive models; Programmable control; Robust control; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
Type :
conf
DOI :
10.1109/ICNN.1994.374593
Filename :
374593
Link To Document :
بازگشت