文献类型：期刊文献

英文题名：Study on fault diagnosis of turbine using an improved cosine similarity measure for vague sets

作者：Shi, L.L.[1]; Ye, J.[1]

机构：[1] Department of Electrical and Information Engineering, Shaoxing University, Shaoxing, China

年份：2013

卷号：13

期号：10

起止页码：1781

外文期刊名：Journal of Applied Sciences

收录：EI(收录号：20134116844441)、Scopus(收录号：2-s2.0-84885085333)

语种：英文

外文关键词：Failure analysis - Membership functions - Turbines

外文摘要：Aiming at complexity and uncertainty of relation between vibration and fault types of turbine, a new method of fault diagnosis of turbine was proposed based on an improved vague cosine similarity measure. Compared with the previous cosine similarity measures for vague sets, the improved cosine similarity measure has more information to deal with vagueness and uncertainty problems by considering truth-membership functions, false-membership functions and hesitancy degree of vague sets and then it can overcome the undefined problem when the degree of membership and degree of non-membership are zero, respectively. Then, the cosine similarity measure was applied to fault diagnosis of turbine. For this fault diagnosis, the matter-element models of the turbine fault were built according to diagnostics derived from specialists' knowledge of practical experience and then through the vague cosine similarity measure between a fault-testing sample and fault knowledge samples, the vibration fault is determined according to the maximum cosine similarity measure value. The fault-diagnosis example of the turbine shows that the proposed method is simple and effective. ? 2013 Asian Network for Scientific Information.