Fault diagnosis method for reciprocating machine based on local wave theory and neural network

被引：0

作者：

Bie Fengfeng ^{[1
]}

Ma Xiaojiang ^{[1
]}

机构：

[1] Dalian Univ Technol Precis & Nontradit Machining, Key Lab, Minist Educ, Dalian 116023, Peoples R China

来源：

Proceedings of the First International Symposium on Test Automation & Instrumentation, Vols 1 - 3 | 2006年

关键词：

local wave; fault diagnosis; IMF; neural network;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A reciprocating mechanical fault diagnosis method is proposed combining local wave method (planar) and an RBF neural network. For the plant mechanical signal, the local wave time-frequency image can reflect the energy change in time and frequency axis. Different fault signals contain distinctly different time-frequency images. In order to extract and classify, the fault information from the IMF from the two-dimension local wave, the RBF neural network is more excellent in terms of its approach and classifying ability. Several representative faults of some reciprocating compressor are investigated applying the proposed method. The result illustrates that it is, to a certain extent effective and practical.

引用

页码：1060 / 1064

页数：5