摘要
针对旋转机械故障信号呈非平稳时变并伴随有强烈的背景噪声的特点,利用经验模态分解获得振动信号的本征模态函数,并对本征模态函数进行系数-能量计算,提取系统的特征信息,并针对EMD中的端点效应问题,提出了基于网格搜索和交叉验证法的最小二乘支持向量机预测方法;在此基础上将能量容差概念引入ART-1神经网络,进而提出了基于EMD与ART-1神经网络相结合的故障分类方法;以离心式风机的故障为例进行分析,实验结果表明该方法在故障信息诊断方面是可行的和有效的,并能够提高故障检测的可靠性。
The fault signal of rotating machinery is proved to be non--stationary and carries intense background noise. The vibration sig- nal was firstly decomposed into intrinsic mode function (IMF) by the empirical mode decomposition (EMD) method. Then the fault informa- tion diagnosis of the rotating machinery vibration signals could be extracted from the coefficient--energy value of intrinsic mode function. Ai- ming at the problem of end effect on EMD, a predictive method of least square support vector machines (LSS--VM) based on grid--search and cross-- validation method is put forward. The conception of energy tolerance was imported to ART-- 1 neural network based on the above theory, then a new fault classification method based on combining EMD and ART-- 1 neural network is forwarded. The multiple faults diag- nosing of a blower fan is illustrated by using the above method, which shows that the method is feasible and effective on multiple faults diag- nosis.
出处
《计算机测量与控制》
CSCD
北大核心
2012年第8期2061-2064,共4页
Computer Measurement &Control