文章摘要
李云飞,苏文胜.基于时序灰度图和分组ResNet的回转支承故障诊断[J].机床与液压,2022,50(12):187-191.
LI Yunfei,SU Wensheng.Fault Diagnosis of Slewing Bearing Based on Time Series Gray Image and Grouped ResNet [J].Machine Tool & Hydraulics,2022,50(12):187-191
基于时序灰度图和分组ResNet的回转支承故障诊断
Fault Diagnosis of Slewing Bearing Based on Time Series Gray Image and Grouped ResNet
  
DOI:10.3969/j.issn.1001-3881.2022.12.034
中文关键词: 回转支承  声发射信号  时序二维化  深度学习
英文关键词: Slewing bearing  Acoustic emission signal  Two-dimensional time series  Deep learning
基金项目:
作者单位E-mail
李云飞 江苏省特种设备安全监督检验研究院,国家桥门式起重机械产品质量监督检验中心 yf98@163.com 
苏文胜 江苏省特种设备安全监督检验研究院,国家桥门式起重机械产品质量监督检验中心  
摘要点击次数: 67
全文下载次数: 0
中文摘要:
      为解决大型回转支承转速低、背景噪声大、常规的声发射诊断方法难以适用的问题,提出一种基于灰度图和ResNet模型相结合的声发射信号处理方法。将声发射信号编码为二维灰度图像,并通过ResNet模型识别声发射信号编码得到的灰度图,通过训练模型实现对大型回转支承的故障诊断。对某型号大型回转支承进行试验,结果表明:以时序二维化后的灰度图作为故障诊断依据,可以显著提高回转支承的故障诊断准确率;相比于传统方法,所提方法泛化性能和鲁棒性能更好,可以很好地应用在实际工况中的大型回转支承故障诊断。
英文摘要:
      In order to solve the problems of low speed of large slewing bearings, high background noise, and difficulty in applying conventional acoustic emission diagnostic methods, an acoustic emission signal processing method based on the combination of gray image and ResNet model was proposed. The acoustic emission signal was encoded into a two-dimensional grayscale image, and the grayscale image obtained by the acoustic emission signal encoding was identified through the ResNet model, and the fault diagnosis of the large slewing bearing was realized through the training model.A certain type of large-scale slewing bearing was tested. The results show that the fault diagnosis accuracy of the slewing bearing can be significantly improved by using the two-dimensional gray-scale diagram of the time series as the basis;compared with the traditional method, the proposed method has better generaliztion performance and robustness, and it can be well applied to the fault diagnosis of large slewing bearings in actual working conditions.
查看全文   查看/发表评论  下载PDF阅读器
关闭

分享按钮