基于EW-型贴近度的云模型相似性度量方法  

Cloud Model Similarity Measure Based on EW-type Closeness

作　　者：黄琼桃 刘瑞敏[1] HUANG Qiongtao;LIU Ruimin(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《贵州大学学报（自然科学版）》2021年第2期44-49,共6页Journal of Guizhou University:Natural Sciences

基　　金：国家自然科学基金资助项目(61863018)。

摘　　要：针对现有的相似性度量方法中存在区分度不高、结果不稳定等问题,提出了一种基于EW-型贴近度的云模型相似性度量方法。该方法利用正态云模型的扩展模型三角云为研究对象,分别把三角云的期望曲线及最大边界曲线看作三角模糊数,通过计算三角模糊数的EW-型贴近度来度量云模型的相似性,充分考虑了期望曲线和最大边界曲线的特点,定义了一种综合的求两云模型相似度的计算方法。通过仿真实验可以看出,提出的EMTCM方法具有一定的区分度;在Synthetic Control Chart Dataset数据集上的分类对比实验表明,EMTCM方法的分类精度明显优于先前的LICM、ECM、MCM方法,验证了EMTCM方法有一定的可行性及有效性。Aiming at the problems of low discrimination and unstable results in existing similarity measurement methods,a cloud model similarity measurement method based on EW-type closeness is proposed.This method uses the extended model triangle cloud of the normal cloud model as the research object,regards the expected curve and the maximum boundary curve of the triangle cloud as the triangle fuzzy number,and measures the similarity of the cloud model by calculating the EW-type closeness of the triangle fuzzy number.It fully considers the characteristics of the expected curve and the maximum boundary curve,and defines a comprehensive calculation method for calculating the similarity of the two cloud models.It can be seen from the simulation experiment that the proposed EMTCM method has a certain degree of discrimination.The classification comparison experiment on the Synthetic Control Chart Dataset data set shows that the classification accuracy of the EMTCM method is significantly better than the previous LCM,ECM,and MCM methods.The EMTCM method has certain feasibility and effectiveness.

关 键 词：三角云 EW-型贴近度 期望曲线 最大边界曲线 相似性度量 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]