基于二叉树型分层的广义混合模糊系统推理规则数的缩减.pdfVIP

控制理论与应用 第 30 卷第 6 期 Vo l. 30No. 6 2013 年6 月 Control ηleory Applications Jun.2013 DOI: 10.764 l1CTA.20 13.21 162 基于二叉树型分层的广义混合模糊系统推理规则数的缩减 杨阳，王贵君↑，杨永强 (天津师范大学数学科学学院，天津3∞387) 摘要:为避免广义混合模糊系统因输入变量个数的增加而引起规则爆炸现象应用二叉树型分层方法给出混合 推理规则，进而对广义混合模糊系统的输入实施二叉树型分层，从理论上获得了该系统分层后的输入输出表达式和 推理规则总数的计算公式.此外，通过实例对该系统分层和不分层的规则总数进行了比较和分析，结果表明分层后 广义混合模糊系统可大幅度缩减推理规则总数，并可有效地避免规则爆炸. 关键词:二叉树型分层;混合推理规则;广义混合模糊系统;推理规则总数 中图分类号: TP183 , 0159 文献标识码:A Reducing the number of inference rules for generalized hybrid fuzzy systems based on binary tree-type hierarchy t YANG Yang , WANG Gui才un ， YANG Yong-qiang (Sch∞1 of Mathematics Sciences, Tianjin Normal Universit弘 Tianjin 3∞387， China) Abstract: To avoid the rule-explosion in the generalized hybrid fuzzy systems with increasing number of input variables , we put the hybrid inference rules in a binary tree-type hierarchy, and hierarchize the input variables of the generalized hybrid fuzzy system in the binary trees to obtain expressions for the input and output of the generalized hybrid fuzzy system, and theoretically derive a formula for computing the total number of inference rules. In an example, we analyze and comp缸E 由e total number of fuzzy inference rules between the hierarchical and non-hierarchical generalized hybrid fuzzy systems. It shows that the total number of inference rules in hierarchical generalized hybrid fuzzy system is reduced greatly, effectively avoiding the rules explosion. Key words: binary tree-type hierarchy; hybrid inference rule; generalized hybrid fuzzy systems; inference rules 1 引言(Introduction)