中国机械工程 ›› 2021, Vol. 32 ›› Issue (12): 1456-1461,1478.DOI: 10.3969/j.issn.1004-132X.2021.12.009

• 智能制造 • 上一篇    下一篇

基于弹性核凸包张量机的机械设备热成像故障诊断方法#br#

何知义1,2;邵海东1,2;程军圣1,2;杨宇1,2   

  1. 1.湖南大学汽车车身先进设计制造国家重点实验室,长沙,410082
    2. 湖南大学机械与运载工程学院,长沙,410082
  • 出版日期:2021-06-25 发布日期:2021-07-08
  • 通讯作者: 邵海东(通信作者),男,1990年生,博士后研究人员、讲师。研究方向为机械设备智能故障诊断和寿命预测。发表论文30余篇。E-mail:hdshao@hnu.edu.cn。
  • 作者简介:何知义,男,1989年生,博士。研究方向为机械设备故障诊断与健康管理。发表论文10余篇。E-mail:hezhiyihnu@126.com。
  • 基金资助:
    国家自然科学基金(51905160)

A Fault Diagnosis Method of Mechanical Equipment Based on FKCH-TM and Thermal Images#br#

HE Zhiyi1,2;SHAO Haidong1,2;CHENG Junsheng1,2;YANG Yu1,2   

  1. 1.State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body,Hunan University,Changsha,410082
    2.College of Mechanical and Vehicle Engineering,Hunan University,Changsha,410082
  • Online:2021-06-25 Published:2021-07-08

摘要: 针对机械设备热成像故障诊断中图像样本向量化维数灾难和结构信息破坏的问题,提出了基于弹性核凸包张量机的机械设备热成像故障诊断方法。首先定义样本特征张量空间的弹性核凸包,然后求解弹性核凸包间的最优超平面来构建弹性核凸包张量机模型,最后以机械设备不同故障状态下的热成像图像为输入,实现机械设备的故障诊断。实验结果表明:该方法可以准确、稳定地对机械设备进行故障诊断。

关键词: 机械设备, 故障诊断, 热成像, 弹性核凸包张量机

Abstract:  Aiming at problems of dimensionality disaster and structure information destruction in thermal imaging fault diagnosis of mechanical equipment, a fault diagnosis method of mechanical equipment thermal images was proposed based on FKCH-TM. Firstly, flexible kernel convex hull(FKCH) of each class in feature tensor space was defined. Then, the optimal hyperplane between the FKCH was found, and FKCH-TM model was built. Finally, thermal images of mechanical equipment under different fault conditions were directly used as the input of FKCH-TM for fault diagnosis. Experimental results show that the proposed method may identify and diagnose the faults of mechanical equipment accurately and stably.

Key words: mechanical equipment, fault diagnosis, thermal image, flexible kernel convex hull based on tensor machine(FKCH-TM) 

中图分类号: