面向RR算子的异蕴涵模糊控制器

Different implicational fuzzy controller facing RR operator

从逻辑严谨性而言,三I算法是当前最为合理的模糊推理方法之一,而其改进与推广形式——泛三I算法的提出,则使得其逻辑合理性与实用性得到有机统一。针对泛三I算法尚未解决的较为核心的问题开展进一步研究,面向RR算子提出了基于泛三I算法的多输入单输出异蕴涵模糊控制器,并研究了其在模式识别中的应用。首先,探讨了采取RR算子的泛三I算法的还原性,结果表明其效果良好。其次,面向多输入单输出情形,提出并构建了基于泛三I算法的异蕴涵模糊控制器;并且分析了9种异蕴涵模糊控制器的响应能力,获得了一些理想的模糊控制器。最后,将异蕴涵模糊控制器应用于模式识别领域,给出了极性情感的模式识别算法,取得了良好效果。这些工作有力推动了模糊推理、模糊控制、模式识别等领域的发展。

From the viewpoint of logistic rationality,the triple I method is currently one of the most reasonable fuzzy reasoning methods,and then,as its improvement and generalization,the universal triple I method accomplishes the organic unification of its logistic rationality and practicability.Aiming at some un-solution that vital problems of universal triple I method,further researches are carried through.For the case employing RR operator,the multiple-input single-output differently implicational fuzzy controller（based on the universal triple I method） is presented,and then its application in pattern recognition is investigated.First,the consistency of universal triple I method employing RR operator is discussed,and it is found that its consistency seems excellent.Secondly,the differently implicational fuzzy controller based on universal triple I method is proposed and established for the case of multiple-input single-output,and then the response abilities are analyzed for 9 differently implicational fuzzy controllers,thereby some ideal specific fuzzy controllers are obtained.Finally,the differently implicational fuzzy controller is applied to the field of pattern recognition,and the pattern recognition algorithm of emotion polarity is provided,which verifies its excellent effect.These works forcefully promote the progress of many fields including fuzzy reasoning,fuzzy control,pattern recognition and so on.