基于无参数监督核局部保持映射降维的故障诊断  

Fault diagnosis based on parameter-less supervised kernel locality preserving projections for dimension reduction

作　　者：汤永清 张新红 陈虹微[1,2] 王悦新 Tang Yongqing;Zhang Xinhong;Chen Hongwei;Wang Yuexin(School of Physics&Mechanical Engineering,Longyan University,Longyan 364000,China;Special Vehicle Industry Technology Development Base,Longyan 364000,China)

机构地区：[1]龙岩学院物理与机电工程学院,龙岩364000 [2]福建省专用车行业技术开发基地,龙岩364000

出　　处：《电子测量与仪器学报》2019年第7期159-165,共7页Journal of Electronic Measurement and Instrumentation

基　　金：福建省自然科学基金(2019J01798,2017J01707);龙博基金(LB2014007)资助项目

摘　　要：针对局部保持映射(LPP)应用于故障诊断存在识别精度不高的问题,提出了基于无参数监督核局部保持映射(PSKLPP)降维的故障诊断新方法。PSKLPP采用对离群数据更为鲁棒的余弦距离对LPP中的欧氏距离进行替换,并融入样本标签信息构造无参数近邻图,利用核方法将提取的高维故障特征映射到一个高维线性空间再进行降维,避免了相似矩阵计算过程中人为选择选择参数的问题,能够获得更有效的低维流形。电机轴承故障诊断的准确率达到了99.05%,相比于改进前和其他几种方法有了较大幅度提升。Aiming at the problem that accuracy of orthogonal locality preserving projections(LPP)for fault diagnosis is not high enough,a fault diagnosis method based on parameter-less supervised kernel locality preserving projections(PSKLPP)for dimension reduction is proposed.In PSKLPP,firstly,by changing the Euclidean distance to the Cosine distance which is more robust to outline,and constructing a parameter-less nearest-neighbor graph which combined sample label information.And then use the nonlinear mapping to map the high dimension fault feature into an implicit feature space to dimension reduction.Thus,a linear transformation is performed to preserve locality geometric structures of the fault feature,which solves the difficulty of parameter selection in computing affinity matrix,as a result,better fault diagnosis accuracy can be achieved.Motor bearing fault diagnosis accuracy is 99.05%,which is greatly improved compared with previous improvement and several other methods.

关 键 词：故障诊断 局部保持映射 无参数 监督 电机轴承 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] TM307[自动化与计算机技术—控制科学与工程]