This paper proposes an integrated approach to rule structure and parameter identification for fuzzy systems. The rule structure problem is formulated as a structure reduction process of the neuro-fuzzy network used to model a fuzzy system and is solved through an iterative algorithm aiming at selecting the minimal number of rules for the problem at hand. The parameter identification problem is solved, as an optimization problem, by means of a genetic algorithm. The integrated algorithm allows manipulation of a fuzzy system to minimize its complexity and to preserve its level of accuracy. Experimental results demonstrate the algorithm’s effectiveness in identifying reduced fuzzy systems with equivalent performance to the original one.
An iterative Two-Phase approach to Fuzzy System Design
ABBATTISTA, Fabio;CASTELLANO, GIOVANNA;FANELLI, Anna Maria
1998
Abstract
This paper proposes an integrated approach to rule structure and parameter identification for fuzzy systems. The rule structure problem is formulated as a structure reduction process of the neuro-fuzzy network used to model a fuzzy system and is solved through an iterative algorithm aiming at selecting the minimal number of rules for the problem at hand. The parameter identification problem is solved, as an optimization problem, by means of a genetic algorithm. The integrated algorithm allows manipulation of a fuzzy system to minimize its complexity and to preserve its level of accuracy. Experimental results demonstrate the algorithm’s effectiveness in identifying reduced fuzzy systems with equivalent performance to the original one.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.