基于Hilbert-Huang变换与人工神经网络的风机故障诊断研究

英文篇名：Study on Fan Fault Diagnosis Based on Hilbert-Huang Transform and Artificial Neural Network
中文刊名：发电设备
英文刊名：Power Equipment
作者：王磊 ; 纪国宜
英文作者：WANG Lei ; JI Guo-yi Institute of Vibration Engineering ; Nanjing University of Aeronautics and Astronautics ; Nanjing 210016 ; China
中文关键词：风机 ; Hilbert-Huang变换 ; 改进的BP神经网络 ; 故障诊断
英文关键词：fan ; Hilbert-Huang transform ; improved BP neural network ; fault diagnosis
出版日期：2012-03-15
机构：南京航空航天大学振动工程研究所;
年：2012
期：02
出版单位：发电设备

摘要

对风机的振动信号进行Hilbert-Huang变换并得到边际谱,以边际谱中各故障频段的能量比为元素构造风机振动信号的特征向量,利用动量法和学习速率自适应改进的BP神经网络模型对风机转子不对中、轴裂纹等故障进行诊断。结果表明该诊断方法是有效的。
Hilbert-Huang transform is applied to fan vibration signal and a marginal spectrum is obtained.Eigenvector of vibration signal of wind turbines is constructed by taking energy ratio of fault bands in the marginal spectrum as elements.Faults such as wind turbine rotor misalignment and axis cracks are diagnosed by momentum method and BP neural network model improved by adaptive learning rate method.The results prove this fault diagnosis method to be effective.

引文

[1]周丽霞,杨荣松,熊瑞平,等.改进的BP神经网络在风机故障诊断中的应用[J].机床与液压,2007,35(2):218-223.
    [2]田松峰,游洋.基于EMD和Hilbert-Huang理论的给水泵振动故障诊断研究[J].发电设备,2011,25(3):152-155.
    [3]刘峡,张学民.小波变换与地震信号特性分析[J].地震,2002,22(3):51-57.
    [4]HUANG N E,SHEN Z,LONG S R.The Empirical Mode Decomposition and Hilbert Spectrum for Nonlinear and Non-stationary Time Series Analysis[J].Proceedings of the Roy-al Society A,1998(454):903-995.
    [5]米江,纪国宜.基于改进BP神经网络的风机故障诊断研究[J].发电设备,2010,24(5):430-434.
    [6]于德介,程军圣,杨宇.机械故障诊断的Hilbert-Huang变换方法[M].北京:科学出版社,2006.
    [7]钟佑明,秦树人,汤宝平.一种振动信号新变换法的研究[J].振动工程学报,2002,15(2):233-238.