Knowledge-Based Adaptive Identification for Process Control and Modelling

Identification methods mainly based on recursive least-squares algorithms have been well developed in recent years. However, almost all existing identification, control and modelling techniques are completely based upon mathematical computations that cannot thoroughly describe many characteristics of a real physical system such as qualitative features. Furthermore, these mathematical algorithms must satisfy theoretical assumptions and pre-conditions that are not necessarily available in practice at all.

[1]  George A. Bekey.  Knowledge Based Systems in Modeling, Simulation and Identification , 1988 .

[2]  D. Riordan,et al.  An intelligent tuning control technique , 1990, Appl. Artif. Intell..

[3]  D. J. Shakley,et al.  Real-time expert systems for control using parallel processors , 1988, Proceedings of the 27th IEEE Conference on Decision and Control.

[4]  B. Bavarian,et al.  Introduction to neural networks for intelligent control , 1988, IEEE Control Systems Magazine.

[5]  M. Monsion,et al.  An Expert System for Industrial Process Identification , 1988 .

[6]  Rolf Isermann,et al.  Parameter-adaptive control with configuration aids and supervision functions , 1985, Autom..

[7]  Karl-Erik Årzén,et al.  Expert control , 1986, at - Automatisierungstechnik.