基于聚类和遗传算法的解释性模糊模型设计  

Design of Interpretable Fuzzy Model Based on Clustering and Genetic Algorithm

作　　者：张永[1] 邢宗义[1] 向峥嵘[1] 胡维礼[1] 

机构地区：[1]南京理工大学自动化学院,南京210094

出　　处：《计算机工程》2007年第8期160-162,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(60474034)

摘　　要：提出了一种基于模糊聚类和遗传算法构建解释性模糊模型的设计方法。定义了模糊模型的精确性指标,给出了模糊模型解释性的必要条件。然后利用模糊聚类算法和最小二乘法辨识初始的模糊模型;采用多目标遗传算法优化模糊模型;为提高模型的解释性,在遗传算法中利用基于相似性的模糊集合和模糊规则的简化方法对模型进行约简。采用该方法对Mackey-Glass系统进行建模,仿真结果验证了该方法的有效性。An approach to constructing interpretable fuzzy system based on fuzzy clustering and genetic algorithm is proposed. The precision index is defined, and the necessary conditions of interpretability are analyzed. The initial fuzzy system is identified using fuzzy clustering algorithm and the least-squares estimator. Genetic algorithm is used to optimize the initial model. Similar fuzzy sets merging and fuzzy rules merging are adopted to reduce the fuzzy model to enhance its interpretability. The proposed approach is applied to the Mackey-Glass system, and the results show its validity.

关 键 词：TS模糊模型 模糊聚类 遗传算法 解释性 精确性 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]