文章摘要
基于多源域混淆蒸馏的变工况轴承故障诊断方法
Fault diagnosis method for variable operating condition bearings based on confusing distillation of multi-source domains
投稿时间:2023-10-11  修订日期:2023-10-11
DOI:
中文关键词: 故障诊断  知识蒸馏  多源域迁移  域混淆  变工况
英文关键词: fault diagnosis  knowledge distillation  Multi-source domain transfer  domain confusion  variable working conditions
基金项目:安徽省自然科学基金“安徽高校自然科学研究重点项目”(KJ2020A0300)
作者单位邮编
丁建建 安徽理工大学 232001
朱晓娟* 安徽理工大学 232001
摘要点击次数: 464
全文下载次数: 0
中文摘要:
      针对轴承工况发生变化时,新工况缺乏足够的标记故障样本和信号分布产生差异等问题,提出了一种基于多源域混淆蒸馏的轴承变工况故障诊断方法。该方法将域混淆的思想引入多教师知识蒸馏方法,首先让每个源域模型互相蒸馏的过程中混淆他们提取的特征,强调提取源域不变信息的同时抑制源域特异信息,再通过向目标域模型蒸馏整合所有源域不变信息,最后使用目标域数据进行模型迁移,有效解决了多个源域迁移时效果不稳定的问题。经实验验证,所提方法可以通过多种已知工况数据实现对目标工况的数据进行故障诊断,且具有较高的准确率。
英文摘要:
      To address the problems such as the lack of sufficient marker fault samples for the new condition and the difference in signal distribution when the bearing operating condition changes, this paper proposes a multi-source domain obfuscated distillation-based bearing fault diagnosis method. In this paper, a bearing fault diagnosis method based on multi-source domain obfuscation distillation for changing working conditions of bearings is proposed. The method introduces the idea of domain obfuscation into the multi-teacher knowledge distillation method, which firstly allows each source domain model to obfuscate their extracted features in the process of distilling each other.Emphasis is placed on extracting source domain invariant information while suppressing source domain specific information, then integrating all source domain invariant information by distilling to the target domain model, and finally using the target domain data for model migration, which effectively solves the problem of unstable effect when migrating multiple source domains. It is experimentally verified that the proposed method can achieve fault diagnosis on the data of the target working condition through multiple known working condition data with high accuracy.
View Fulltext   查看/发表评论  下载PDF阅读器
关闭

手机扫一扫看 分享按钮