摘要
针对旋转机械复合故障振动信号中的非线性、非平稳特征,提出了一种基于GP奇异谱分解(GP-SSD)的故障特征提取方法。奇异谱分解(SSD)是一种新的针对非线性非平稳信号的自适应信号处理方法,但其具有主观选取嵌入维数的缺点。GPSSD方法基于GP算法能根据嵌入维数与关联维数的关系自适应选取嵌入维数的优势,可以自适应的分解出若干具有物理意义的奇异谱分量(SSC),从而克服了SSD主观选取嵌入维数的缺点。仿真信号的分析结果验证了GP-SSD方法的优越性,在此基础上将GP-SSD应用于旋转机械复合故障诊断中,实验数据的分析结果表明该方法能有效提取旋转机械复合故障的特征。
For the non-linear and non-stationary characteristics of compound fault vibration signal in rotating machinery,a fault feature extraction method based on GP-singular spectrum decomposition( GP-SSD) is proposed. Singular spectrum decomposition( SSD) is a new adaptive signal processing method for nonlinear and non-stationary signal,but it has the disadvantage of selecting embedding dimension subjectively. In the GP-SSD method,the GP algorithm can adaptively select the embedding dimension according to the relationship between the embedding dimension and the correlation dimension to adaptively obtain some singular spectrum components( SSCs) with physical meaning,thus the disadvantage of selecting embedding dimension is overcomed subjectively. The simulation results verify the superiority of the GP-SSD method,then GP-SSD is applied to the compound fault diagnosis of rotating machinery. The experimental results show that the method can effectively extract characteristics of compound faults in rotating machinery.
出处
《电子测量与仪器学报》
CSCD
北大核心
2018年第5期17-24,共8页
Journal of Electronic Measurement and Instrumentation
基金
国家科技支撑计划(2015BAF32B03)
国家自然科学基金(51575168,51375152)
江苏省科技项目(BY2016037-03)资助
关键词
GP算法
奇异谱分解
旋转机械
复合故障诊断
特征提取
GP algorithm
singular spectrum decomposition (SSD)
rotating machinery
compound fault diagnosis
feature extraction
