Two step design methodology of near optimal fuzzy logic controller

Autor: Naser Prljaca, Jakub Osmic
Rok vydání: 2003
Předmět:
Zdroj: IFAC Proceedings Volumes. 36:257-265
ISSN: 1474-6670
DOI: 10.1016/s1474-6670(17)32544-2
Popis: In this paper a two-step design methodology of near optimal Mamdani type fuzzy logic controller (FLC) of an nonlinear system is proposed and analysed. In the first step tuning/learning process of a data base /knowledge base of the FLC is based on the model of the system using Genetic Algorithms (GA). The achieved solution is optimal with regard to the model of the system. In the second step experiments were performed on the real system in the vicinity of optimum (achieved by GA) and using Response Surface Methodology (RSM) control parameters were readjusted in order to achieve a near optimal solution. The proposed two step methodology gives a systematic way of near optimal FLC tuning/learning when confronted with the real system.
Databáze: OpenAIRE