A Research On Variable Weight Synthesis Method For Fuzzy Reasoning

Fuzzy reasoning has always been concerned by many people, as an important part of fuzzy logic. People had studied fuzzy reasoning theory for several years and had made great achievements, especially on industrial control. Many methods of fuzzy reasoning which can be implemented by computer, had been put forward. To make the reasoning result more reasonable, some knowledge, such as, the principle of variable weighted and interactive rules were introduced. Using the knowledge, it can make the reasoning accord with the man's thought better, and improve the accuracy of reasoning.At first, this thesis investigates a problem in fuzzy inference which handles interacting rules, according to the traditional concept of variable weight and state variable weight, we redefine the concept of variable weight and state variable weight, proposes a method of variable weighted synthesis inference to handle interaction, and proves that the method is consistent under some special conditions. And then, the membership of interval-valued fuzzy sets was easier to assure than fuzzy sets', because its membership function contained not only the positive information but also the negative of variable, it provides more choices for the attribute description of an object and has stronger ability to express uncertainty than ordinary fuzzy sets. Several methods of interval-valued fuzzy reasoning has been discussed in the thesis, the new similarity measures of interval-valued fuzzy sets are defined based on the inclusion degree, then a method of variable weighted synthesis inference of interval-valued fuzzy sets is proposed, the process of reasoning is illustrated with an example. Finally, the thesis proposes a new similarity measure of interval-valued, discusses the problem for interval-valued fuzzy inference with interacting rules, and establishes the interval-valued fuzzy reasoning method which handles interacting rules.