电子测量与仪器学报

韩延,李超,黄庆卿,文瑞,张焱.小样本下时序注意力边界增强原型网络的齿轮箱故障诊断方法[J].电子测量与仪器学报,2023,37(2):90-98

小样本下时序注意力边界增强原型网络的齿轮箱故障诊断方法

Boundary-enhanced prototype network with time-series attention forgearbox fault diagnosis under limited samples

DOI：

英文关键词:fault diagnosis few-shot task prototype network metrics learning time-series attention

基金项目:国家重点研发计划项目(2021YFB3301000)、国家自然科学基金(51605065)、中国博士后科学基金(2022MD713687)、重庆市博士后科学基金项目(cstc2021jcyj-bshX0094)资助

作者	单位
韩延	1. 重庆邮电大学自动化学院,2. 重庆邮电大学工业互联网研究院
李超	1. 重庆邮电大学自动化学院,2. 重庆邮电大学工业互联网研究院
黄庆卿	1. 重庆邮电大学自动化学院,2. 重庆邮电大学工业互联网研究院
文瑞	1. 重庆邮电大学自动化学院
张焱	1. 重庆邮电大学自动化学院,2. 重庆邮电大学工业互联网研究院

Author	Institution
Han Yan	1. Automation College, Chongqing University of Posts and Telecommunications,2. Institute of Industrial Internet, Chongqing University of Posts and Telecommunications
Li Chao	1. Automation College, Chongqing University of Posts and Telecommunications,2. Institute of Industrial Internet, Chongqing University of Posts and Telecommunications
Huang Qingqing	1. Automation College, Chongqing University of Posts and Telecommunications,2. Institute of Industrial Internet, Chongqing University of Posts and Telecommunications
Wen Rui	1. Automation College, Chongqing University of Posts and Telecommunications
Zhang Yan	1. Automation College, Chongqing University of Posts and Telecommunications,2. Institute of Industrial Internet, Chongqing University of Posts and Telecommunications

摘要点击次数: 264

全文下载次数: 707

中文摘要:

针对小样本条件下原型网络在提取特征过程中会丢失振动数据的时序特征,且未修正样本在度量空间中的分布导致模型精度低的问题,提出一种时序注意力边界增强原型网络的齿轮箱故障诊断方法。首先,通过构建时间序列注意力模块,建立通道间的时序特征依赖,获得通道时序融合特征;然后,在计算类原型之后,增加邻边界损失以修正度量空间中的故障特征类内和类间分布,明确类原型的表征边界。最后,通过计算测试样本与类原型的欧氏距离,输出故障诊断结果。实验表明,在小样本条件下本文所提方法相比其他方法具有更高的故障诊断精度。

英文摘要:

To address the problem that the time-series characteristics of vibration data are lost in the process of feature extraction in the prototype network, and the distribution of samples in the metric space is not corrected which results in low model accuracy under few-shot task, this paper proposes a new boundary-enhanced prototype network with time-series attention for gearbox fault diagnosis. First, the time-series fusion features of the channels are obtained by building a time-series attention module to establish the time-series feature dependencies between channels. Then, after calculating the class prototypes, the near-neighbor boundary loss is added to correct the intra- and inter-class distributions of the fault features in the metric space to clarify the representation boundaries of the class prototypes. Finally, the fault diagnosis results are output by calculating the Euclidean distance between the test sample and the class prototype. The experiments show that the proposed method in this paper has higher fault diagnosis accuracy compared with other methods under small sample conditions.

查看全文查看/发表评论下载PDF阅读器