摘要
轴承故障诊断对及时发现安全隐患以保证生产的正常运作具有重大意义。但大多数故障诊断关注的重点是诊断精度,却忽略了实际工厂最关心的诊断时效性、诊断过程占用资源的代价等指标。针对该问题,文章提出了一种基于边缘计算的轴承故障诊断方法。该方法使边缘计算和轴承故障诊断相结合,利用边缘计算框架的优势将复杂度低的任务分配到离数据源较近的边缘端,甚至设备端,复杂度高的任务交给云端处理。并采用CNN和oneclasssvm联合处理的方法,进行故障诊断。实验结果表明,该方法相对于单独使用CNN或oneclasssv更有优势,在保证诊断精度的同时,也能保证诊断的时效性。
Bearing fault diagnosis is of great significance for timely detection of potential safety hazards to ensure the normal operation of production.However,most of the fault diagnosis focuses on the accuracy of the diagnosis,but ignores the indicators such as the timeliness of diagnosis and the cost of resources occupied by the diagnosis process.Aiming at the problem,this paper proposes a bearing fault diagnosis method based on edge computing.This method combines edge computing and bearing fault diagnosis,and uses the advantages of the edge computing framework to distribute low-complexity tasks to the edge end closer to the data source,or even the device side.High-complexity tasks are delivered to the cloud for processing.And adopt the combined processing method of cnn and oneclasssvm for fault diagnosis.Experimental results show that this method has advantages over using cnn or oneclasssv alone,while ensuring the accuracy of diagnosis,it can also ensure the timeliness of diagnosis.
作者
彭超
唐向红
陆见光
PENG Chao;TANG Xiang-hong;LU Jian-guang(Key Laboratory of Advanced Manufacturing Technology,Ministry of Education,Guizhou University,Guiyang 550025,China;School of Mechanical Engineering,Guizhou University,Guiyang 550025,China;Guizhou Provincial Key Lab of Public Big Data,Guiyang 550025,China)
出处
《组合机床与自动化加工技术》
北大核心
2020年第12期52-55,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
贵州省公共大数据重点实验室开放基金资助项目(2017BDKFJJ019)
贵州大学引进人才基金资助项目(贵大人基合字(2016)13号)
贵州省留学回国人员科技活动择优资助项目-优秀类项目(2018.0002)。
作者简介
彭超(1995-),男,重庆万州人,贵州大学硕士研究生,研究方向为轴承故障诊断,(E-mail)873283941@qq.com;通讯作者:唐向红(1979-),男,湖南永州人,贵州大学教授,硕士生导师,研究方向为大数据与智能制造,(E-mail)txhwuhan@163.com。
