基于局部线性嵌入和云神经网络的转子故障诊断方法

PDF(1645 KB)

振动与冲击 ›› 2012, Vol. 31 ›› Issue (23) : 99-103.

论文

基于局部线性嵌入和云神经网络的转子故障诊断方法

孙斌薛广鑫陈军蒋能飞

作者信息 +

A Method of Rotor Fault Diagnosis Based on Local Linear Embedding and Cloud Neural Network

SUN Bin, XUE Guang-xin, CHEN Jun, JIANG Neng-fei

Author information +

文章历史 +

摘要

提出了一种基于流形学习的汽轮机转子故障诊断方法。利用振动信号构造一个能够表示该信号的矩阵作为流形学习的输入数据，使用局部线性嵌入算法对矩阵进行维数约简，实现了高维数据向低维空间的嵌入，从而有效提取了故障特征。文中使用云神经网络分类器测试LLE算法输出维数大于3时的故障诊断率，并分析了各个参数对诊断率的影响。该方法克服了在样本较少的情况下故障诊断的困难，能在有限的故障数据中发掘故障特征并进行故障诊断。

Abstract

A fault diagnosis method based on local linear embedding (LLE) is proposed in the paper to deal with turbine rotor fault, which is constructing a feature matrix with original vibration signals as the input data of manifold learning, and embedding the high-dimensional data into low-dimensional space, by LLE algorithm for dimensionality reduction. The rate of fault diagnosis is calculated by cloud neural network when the output dimensionality of LLE algorithm is greater than 3, and the relationship between the rate and parameters is discussed in this paper. Although the sample of fault is limited, we are success to complete the fault diagnosis with the method which is represented in this paper.

导出引用

孙斌薛广鑫陈军蒋能飞. 基于局部线性嵌入和云神经网络的转子故障诊断方法[J]. 振动与冲击, 2012, 31(23): 99-103

SUN Bin;XUE Guang-xin;CHEN Jun;JIANG Neng-fei. A Method of Rotor Fault Diagnosis Based on Local Linear Embedding and Cloud Neural Network[J]. Journal of Vibration and Shock, 2012, 31(23): 99-103