文献类型：期刊文献

英文题名：Multi-classification LSSVM application in fault diagnosis of wind power gearbox

作者：Jiao, Bin[1]; Xu, Zhixiang[1,2]

机构：[1] Electric Engineering School, Shanghai DianJi University, No.690, Jiang Chuan Rd., Min Hang District, Shanghai 200240, China; [2] College of Information Science and Engineering, East China University of Science and Technology, No.130, Mei Long Rd., Shanghai 200237, China

年份：2012

卷号：125 AISC

起止页码：277

外文期刊名：Advances in Intelligent and Soft Computing

收录：EI(收录号：20122015019939)

语种：英文

外文关键词：Decision trees - Computer aided diagnosis - Neural networks - Wind power - Wind turbines - Fault detection - Gears

摘要：For wind turbine gearbox fault diagnosis problem, we propose a multi-classification least squares support vector machines (MCLSSVM) model. According to failure mechanism and vibration characteristics of gearbox, it investigates some formulas of fault diagnosis. Through the combination of voting method and decision tree, it constructs the MCLSSVM decision-making structure, and then it is applied on the fault diagnosis of wind turbine gearbox. Tests show that MCLSSVM can be effectively used in the fault diagnosis of wind turbine gearbox. It solves the studying problem of small sample, and overcomes the shortcoming of artificial neural network (ANN) when it is used in fault diagnosis. ? 2012 Springer-Verlag GmbH Berlin Heidelberg.