Modelling supervisory control systems using fuzzy cognitive maps

Abstract Modelling complex systems and their supervisior has attracted the high interest of many scientists and engineers. There has been a need for highly sophisticated Autonomous Intelligent Systems. A very promising methodology to model the Supervisor of a plant is the use of Fuzzy Cognitive Maps (FCM). FCM are a combination of Fuzzy Logic and Neural Networks. A new mathematical model for FCMs is proposed and its representation is examined in this paper. FCM construction is presented through the development of the model for a simple control process problem. Then, issues for the application of FCM as the model of the supervisor of a complex system are addressed and a hierarchical two-level structure is proposed.