Fuzzy Cognitive Maps in modeling supervisory control systems

Article type: Research Article

Authors: Stylios, Chrysostomos D. | Groumpos, Peter P.

Affiliations: Laboratory for Automation and Robotics, Department of Electrical and Computer Engineering, University of Patras, 26500 Rion, Patras, Greece. Tel.: +30 61 997295; Fax: +30 61 997309; E-mail: groumpos@ee.upatras.gr

Abstract: This paper investigates a hybrid methodology that combines fuzzy logic and neural networks, Fuzzy Cognitive Map (FCM), for modeling and controlling Supervisory Control Systems. A mathematical description of Fuzzy Cognitive Maps (FCM) will be presented and new construction methods will be extensively examined. A Fuzzy Cognitive Map will be developed to model and control a process example and the Supervisor-FCM model characteristics will be discussed. There is a going need for more autonomous and intelligent systems, especially in Complex Systems area, the application of Fuzzy Cognitive Maps for modeling the Supervisor may contribute to develop more sophisticated systems.

Journal: Journal of Intelligent and Fuzzy Systems, vol. 8, no. 1, pp. 83-98, 2000