文献类型：会议论文

英文题名：A Fast Intuitionistic Fuzzy Support Vector Machine Algorithm and Its Application in Wind Turbine Gearboxes Fault Diagnosis

作者：Jiao, Bin[1];Zhang, Qing

机构：[1]Shanghai DianJi Univ, Shanghai 200240, Peoples R China; East China Univ Sci & Technol, Shanghai 200237, Peoples R China

会议论文集：Asia Simulation Conference/International Conference on System Simulation and Scientific Computing (AsiaSim and ICSC 2012)

会议日期：OCT 27-30, 2012

会议地点：Shanghai, PEOPLES R CHINA

语种：英文

外文关键词：wind turbine gearbox; fault diagnosis; intuitionistic fuzzy; support vector machine

摘要：Support vector machine has been successfully applied to the fault diagnosis field, but there are still some problems in practical applications. In this paper we proposed an improved algorithm which reduces the number of support vectors through the reduction of the sample spae to improve the efficiency of the algorithm. As the traditional fuzzy support vector machine cannot classify the sample with the same membership, so we use intuition index to lower the probability of the sample to get the same membership. Here we improve the accuracy of the algorithm through properly redefine the fuzzy membership and intuition index. Finally, we use the improved algorithm to build a multi-classifier based on one against one principle and the voting rules, and apply the multi-classification algorithm to the wind turbine gearbox fault diagnosis. The diagnose results prove that the improved algorithm we proposed can properly resolve the problem of wind turbine gearboxes fault diagnosis.