摘要
实现了基于多重分形的往复压缩机振动信号的故障特征提取。针对往复压缩机振动信号的非线性和非平稳性,使用多重分形谱和广义维数对压缩机振动信号进行分析,从中提取可识别的故障特征。分析结果发现多重分形谱中的Δα值和广义维数Dq作为故障特征能够很好地反映往复压缩机的工作状态,为往复压缩机的故障特征识别提供了必要依据。
The fault feature extraction of reciprocating compressor vibration signals based on multifractal theory is presented in this study. Aiming at that the reciprocating compressor vibration signals are nonlinear and non- stationary, multifractal spectrum and general dimension are applied to analyze compressor vibration signals and extract fault feature which can be identified. The analysis results show that multifractal spectrum and general dimension give a good presentation for reciprocating compressor working condition and provide necessary evidence for reciprocating compressor fault feature identification.
出处
《压缩机技术》
2008年第1期12-14,共3页
Compressor Technology
关键词
往复压缩机
故障诊断
多重分形谱
广义维数
reciprocating compressor
fault diagnosis
multifractal spectrum
general dimension
