摘要
提出了基于VPMCD(Variable Predictive Model Based Class Discriminate,简称VPMCD)和EMD(Empirical mode decomposition,简称EMD)的齿轮故障诊断方法,并将它应用于齿轮稳态信号的分析。VPMCD方法是一种新的模式识别方法,特别适合于非线性分类问题,它充分利用从原始数据中所提取的特征值之间的相互内在关系建立数学模型,从而进行模式识别。在基于VPMCD和EMD的齿轮故障诊断方法中,首先采用EMD方法将齿轮振动信号自适应地分解为若干个单分量信号,然后提取各个分量的样本熵并将其作为特征值,最后采用VPMCD分类器进行故障识别和分类。结果表明该方法能够有效地突出齿轮故障振动信号的故障特征,提高了齿轮故障诊断的准确性。
A gear fault diagnosis method based on variable predictive model-based class discriminate (VPMCD) and empirical mode decomposition (EMD) was proposed here. It was applied to analyze steady state signals of gears. The method of VPMCD was a new way for pattern recognitions, and specially appropriate to classification of nonlinear problems. It made full use of the inner relations among eigen-values extracted from the original data to recognize models. Using the gear fault diagnosis method based on VPMCD and EMD, gear vibration signals firstly, were adaptively decomposed into several single-component signals with EMD, then the sample entropy of each component was extracted and taken as eigen-values, the VPMCD classifier finally was used to recognize and classify the faults. The results showed that this method can effectively highlight the fault features of gear fault vibration signals, and raise the correctness of gear fault diagnosis.
出处
《振动与冲击》
EI
CSCD
北大核心
2013年第20期9-13,共5页
Journal of Vibration and Shock
基金
国家自然科学基金(51175158
51075131)
湖南省自然科学基金(11JJ2026)
中央高校基本科研业务费专项基金资助项目
关键词
VPMCD
样本熵
齿轮
故障诊断
variable predictive model-based class discriminate (VPMCD)
sample entropy
gear
fault diagnosis