摘要
基于目前采用遗传算法模糊控制规则优化存在的没有考虑规则的相容性、得到的模糊控制规则难以理解的问题,介绍了以模糊贴近度为评价标准,定量评价模糊规则相容性的方法。在此基础上,提出了新的适应度函数,即遗传算法的适应度函数由控制性能指标和规则相容性指标两部分组成。利用遗传算法生成了模糊规则,同时保证规则之间的相容性。仿真结果表明,生成的模糊规则具有较好的相容性和控制性能。
To the problems of optimizing fuzzy control rule using genetic algorithm-without thinking about the compatibility of rules, and the intfication of the obtained fuzzy control rule, a method to evaluate the compatibility of fuzzy rule in quantity based on approaching degree of fuzzy subsets is introduced. Furthermore, a method is presented to generate fuzzy rules simultaneity keeping compatibility of fuzzy rule by using genetic algorithm with a novel fitness-function, which combines the control performance index with the compatibility of rule index. The simulation resuits show that the fuzzy rules have good compatibility and control performance.
出处
《控制工程》
CSCD
2006年第2期145-148,共4页
Control Engineering of China
基金
西南师范大学基金资助项目(SWNUF2004008)
关键词
模糊规则
完备性
相客性
遗传算法
仿真
fuzzy rules
completeness
compatibility
genetic algorithm
simulation
