基于图正则化堆叠自编码器的风机轴承故障诊断方法  

Bearing Faults Diagnosis Method Based on Stacked Auto-Encoder With Graph Regularization for Wind Turbines

作　　者：刘展 刘健洵 包琰洋 李大字[2] LIU Zhan;LIU Jianxun;BAO Yanyang;LI Dazi(Beijing Pukang Measurement and Monitoring Tech Co.,Ltd.,Fengtai District,Beijing 100070,China;College of Information Science and Technology,Beijing University of Chemical Technology,Chaoyang District,Beijing 100029,China)

机构地区：[1]北京能高普康测控技术有限公司,北京市丰台区100070 [2]北京化工大学信息科学与技术学院,北京市朝阳区100029

出　　处：《发电技术》2024年第6期1146-1152,共7页Power Generation Technology

基　　金：国家自然科学基金项目(62273026);工信部高技术船舶科研项目(MC-202025-S02)。

摘　　要：【目的】为解决风电机组轴承故障诊断中的故障特征提取效率低、特征表示不够精准以及现有方法难以适应复杂信号需求的问题,提出一种基于图正则化的故障诊断方法,提升了对振动信号的分析能力,从而实现对不同故障类型的准确分类和可靠诊断。【方法】采用基于图正则化自编码器的技术,结合图嵌入思想,指导堆叠监督自编码器进行特征提取。在故障特征提取阶段,首先对诊断信号进行图表示,然后在堆叠自编码器中添加图正则化项,以确保嵌入后的低维特征保持流形结构,从而提取数据深层的复杂几何特征。【结果】所提取的特征能够实现对不同故障类型的准确分类,展现出在故障特征捕捉方面的显著优势。实验结果表明,该方法在实际风场数据的应用中,具有更高的诊断精度和可靠性,有效提高了故障特征的提取效率和分类准确性。【结论】所提方法在风电轴承故障诊断领域表现出显著的有效性和优越性,为实际应用提供了可靠的技术支持。[Objectives]In order to solve the problems of low efficiency of fault feature extraction,inaccurate feature representation,and difficulty in adapting existing methods to complex signal requirements in wind turbine bearing fault diagnosis,a fault diagnosis method based on graph regularization was proposed.The method helps to improve the analysis ability of vibration signals,thereby achieving accurate classification and reliable diagnosis of different fault types.[Methods]The technology based on graph regularization auto-encoder was adopted,combined with the idea of graph embedding,to guide the stack-supervised autoencoder to carry out feature extraction.In the fault feature extraction stage,the diagnostic signal was first graphically represented,and then graph regularization terms were added to the stacked auto-encoder to ensure that the embedded lowdimensional features maintain the manifold structure,thereby extracting complex geometric features deep in the data.[Results]The extracted features can accurately classify different fault types,showing significant advantages in fault feature capture.Experimental results show that this method has higher diagnosis accuracy and reliability in the application of actual wind farm data,and effectively improves the extraction efficiency and classification accuracy of fault features.[Conclusions]The proposed method has shown significant effectiveness and advantages in the field of wind power bearing fault diagnosis,and provides reliable technical support for practical applications.

关 键 词：风电机组 轴承 故障诊断 图正则化 自编码器 图嵌入 特征提取 振动信号 

分 类 号：TK83[动力工程及工程热物理—流体机械及工程]