摘要
根据离心泵故障振动信号的特点,提出了利用小波包分解、重构技术进行消噪处理及频带能量特征提取,并以“能量”为元素,构造离心泵振动信号的特征向量,通过对神经网络和模糊系统的结合方式的研究,提出了一种基于小波包和模糊神经网络的离心泵轴系故障诊断方法,实验分析结果表明,该方法可以有效地对离心泵轴系振动信号进行诊断。
According to the characteristics of fault vibration signals of centrifugal pump,a method of killing the noise and extracting frequency band energy feature by means of wavelet package decomposition and reconstruction technique is presented.The conception of 'energy' is proposed based on the theory that signals energy in all frequency can be affected by faults considerably.To construct feature vectors of Centrifugal Pump vibration signals,a centrifugal pump fault diagnosis method base on wavelet package and fuzzy neural network is put forward and applied to the fault diagnosis of centrifugal pump axis.The experimental result indicates that this method can efficiently process the fault vibration signals of the centrifugal pump axis.
出处
《华电技术》
CAS
2006年第9期1-5,共5页
HUADIAN TECHNOLOGY
关键词
小波包
神经网络
模糊神经网络
特征提取
故障诊断
wavelet packet
neural network
fuzzy neural network
feature extraction
fault diagnosis
