基于小波分析的风机故障诊断  

Sintering fan faults diagnosis based on wavelet analysis

作　　者：胡汉辉[1] 杨洪[1] 谭青[1] 易念恩[1] 

机构地区：[1]中南大学机电工程学院

出　　处：《中南大学学报（自然科学版）》2007年第6期1169-1173,共5页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(50675227)

摘　　要：根据故障信号特征和小波变换多尺度分解性质选取小波分解层次,得到能正确地反映风机运行状态的特征向量;参照特征向量的组成方法,提出并构建基于小波分析的韶钢4号风机典型故障特征表。对待检信号选用db10小波进行6层小波分解,利用待检状态的特征向量与典型故障特征表,通过模糊模式识别方法进行风机故障诊断。结合傅里叶分析方法进一步找出风机存在的倍频微弱信号。实际诊断结果表明:振动信号与故障特征表中典型不平衡故障的模糊贴近度达到0.958,从而诊断出实例中风机存在不平衡故障;风机存在0.5倍频微弱信号,据此有利于发现风机与该频率相关的早期微弱故障征兆。Wavelet decomposition levels were selected according to the characteristics of fault signal and wavelet transform mulfiscale decomposition property, and the feature vector was obtained that can be used to reflect the running status of the sintering fan. According to the feature vector composition method, wavelet analysis method was used to deal with the fault diagnosis of the 4th sintering fan in Shaogang Steel Group, and a feature table of typical fault was built. Detecting signal with dbl0 wavelet six layers wavelet decomposition can reflect the nature of the fan failure. Fourier＇s analysis method was further used to discover frequency multiplication weak signal. The actual diagnosis result shows that using the feature vector typical characteristic fault, imbalance fault reaches 0.958 through the fuzzy pattern recognition, showing that there exists fan＇s imbalance fault. 0.5 frequency multiplication weak signal occurs in the fan, which is useful to discover early weak fault indication that relates to this frequency.

关 键 词：小波分析 烧结风机 故障诊断 

分 类 号：TH165.3[机械工程—机械制造及自动化]