Handling of inconsistent rules with an extended model of fuzzy reasoning

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, t he rule-base is usually tuned through modification of membership functions. Effect of changing a membership function is global in the sense that it in fluences all rules that involve the membership function. Here we propose a very effective extension of the conventional fuzzy reasoning system with in corporation of an importance factor for each rule. This factor allows tunin g of the system at the rule level. Of course, one can still tune the member ship functions. It enables the system to cope with incorrect and/or incompa tible rules and thereby enhances the robustness, flexibility and system mod eling capability. The proposed model is quite general and can be used in di fferent applications including control. In the present investigation, we de monstrate with extensive simulation how for a control application inconsist ent rules can be dealt with.