摘要
往复机械振动信号非常复杂,通常存在较强的非平稳、非线性特征,使得对其进行振动信号分析、故障识别存在困难。对此提出一种基于改进的总体经验模态分解(CEEMDAN)与奇异值结合的故障特征识别方法,对原始信号进行CEEMDAN分解,得到本征模式函数的奇异值,将得到的奇异值作为特征向量输入支持向量机进行特征分类,从而实现故障模式的识别。通过对实验室模拟故障与往复泵动力端故障模式识别实例分析来论证方法有效性。研究结果表明,该方法适用于提取往复机械振动信号冲击特征和多故障模式识别。
The vibration signals of the reciprocating machinery are very complex with the characteristics of strong nonlinearity and non-stability.Therefore,it is difficult to diagnose the faults of the reciprocating machinery through the vibration signals.Aiming at this problem,a fault diagnosis method is proposed based on the combination of CEEMDAN method and singular value decomposition.The vibration signals collected from the reciprocating machinery are decomposed with CEEMDAN method,and the singular value series is calculated through singular value decomposition from the intrinsic mode function of the CEEMDAN decomposition.Then,the singular value series is used as the feature vector for the support vector machine(SVM)classifier.Finally,the fault pattern recognition of the reciprocating machines is realized.The simulation analysis and experimental test are carried out as the verification for the method.It shows that the method is effective in extracting the vibration characteristics of the reciprocating machinery vibration signals and multi-fault pattern recognition.
作者
别锋锋
徐鹏青
裴峻峰
张仕佳
BIE Fengfeng;XU Pengqing;PEI Junfeng;ZHANG Shijia(College of Mechanical Engineering,Changzhou University,Changzhou 213164,Jiangsu China)
出处
《噪声与振动控制》
CSCD
2018年第4期180-185,207,共7页
Noise and Vibration Control
基金
国家自然科学基金资助项目(51175051)
江苏省2016高校优秀中青年教师和校长境外研修计划资助项目
关键词
振动与波
往复机械
总体经验模态分解
奇异值
支持向量机
vibration and wave
reciprocating machinery
ensemble empirical mode decomposition
singular value
support vector machine