Handling of inconsistent rules with an extended model of fuzzy reasoning

Abstract

It may not always be possible for an expert to provide a set of completely consistent rules. Even if the rules are consistent, all rules may not have equal importance to control the system. Moreover, for a fuzzy controller, the rule-base is usually tuned through modification of membership functions. Effect of changing a membership function is global in the sense that it influences all rules that involve the membership function. Here we propose a very effective extension of the conventional fuzzy reasoning system with incorporation of an importance factor for each rule. This factor allows tuning of the system at the rule level. Of course, one can still tune the membership functions. It enables the system to cope with incorrect and/or incompatible rules and thereby enhances the robustness, flexibility and system modeling capability. The proposed model is quite general and can be used in different applications including control. In the present investigation, we demonstrate with extensive simulation how for a control application inconsistent rules can be dealt with.