Fuzzy Logics Simulation Technology in General Strategy of Intelligent Systems Designing - Fuzzy Information Processing Society, 1996. NAFIPS., 1996 Biennial Conference of the North America

This research suggests a new type of inte1,ligent system based on new approaches to generation and selection of family of fuzzy logics. It is based on new generalizedl strategy of designing of fuzzy intelligent systems. Family of fuzzy logics can be generated by T-norms axiomatic system to choose the logic, that is the most suitable to expert way of thinking. Then the selected fuzzy logic is tested by simulation model of inference process in intelligent system. The designed logic is the input of inference module of intelligent system. 1. Bottom-up and top-down strategies of designing of fuzzy intelligent systems We suggest two approaches for designing of fuzzy intelligent systems. The first approach (we call it bottom-up designing) consists in designing fuzzy intelligent systems on the basis of possible hardware realization for a kinowledge based processor. For example, such a system can be combination of fuzzy controller and fuzzy neural network. Then we design all possible fuzzy intelligent systems, based on this architectural decision. This is the case of soft computing. The second approach (we call it top-down designing) consists in selecting one of the possible algebras of knowledge in semiotic systems fuzzy algebra, soft algebra,, uncertainty algebra, rough algebra etc. Let it be a fuzzy algebra. On the basis of such fuzzy algebra we can generate various fuzzy * This work is supported by Russian Fund of Fundamental Researches (grant N 95-01-01030 "Simulation of fuzzy expert systems" 0-7803-3225-3-6/96 $5.00