Fuzzy Cognitive Networks: Adaptive Network Estimation and Control Paradigms.

Fuzzy Cognitive Networks (FCN) constitutes an operational extension of Fuzzy Cognitive Maps (FCM), which assume that they always reach equilibrium points during their operation. Moreover, they are in continuous interaction with the system they describe and may be used to control it. FCN are capable of capturing steady state operational conditions of the system they describe and associate them with input values and appropriate weight sets. In the sequence they store the acquired knowledge in fuzzy rule based data bases, which can be used in determining subsequent control actions. This chapter presents basic theoretical results related to the existence and uniqueness of equilibrium points in FCN, the adaptive weight estimation based on system operation data, the fuzzy rule storage mechanism and the use of the entire framework to control unknown plants. The results are validated using well known control benchmarks. [ABSTRACT FROM AUTHOR]