Page 97 - 《软件学报》2021年第8期

软件学报 ISSN 1000-9825, CODEN RUXUEW E-mail: jos@iscas.ac.cn
Journal of Software,2021,32(8):2379−2390 [doi: 10.13328/j.cnki.jos.006188] http://www.jos.org.cn
©中国科学院软件研究所版权所有. Tel: +86-10-62562563

∗
基于 1D-CNN 联合特征提取的轴承健康监测与故障诊断

1
1
1
刘 立 , 朱健成 , 韩光洁 , 毕远国 2
1
(河海大学 信息学部物联网工程学院,江苏 常州 213022)
2
(东北大学 计算机科学与工程学院,辽宁 沈阳 110169)
通讯作者: 韩光洁, E-mail: hanguangjie@gmail.com

摘 要: 针对特定机械设备构建数据驱动的故障诊断模型缺乏泛化能力,而轴承作为各型机械的共有核心部件,
对其健康状态的判定对不同机械的衍生故障分析具有普适性意义.提出了一种基于 1D-CNN(one-dimensional
convolution neural network)联合特征提取的轴承健康监测与故障诊断算法.算法首先对轴承原始振动信号进行分区
裁剪,裁剪获得的信号分区作为特征学习空间并行输入 1D-CNN 中,以提取各工况下的代表性特征域.为了避免对故
障重叠信息的处理,优先使用对健康状态敏感的特征域构建轴承健康状态判别模型,若健康状态判别模型识别轴承
未处于健康状态,特征域将与原始信号联合重构,通过耦合自动编码器开展故障模式判定.使用凯斯西储大学(Case
Western Reserve University)的轴承数据开展实验,结果表明,该算法继承了深层学习模型的准确性和鲁棒性,具有较
高的故障诊断精度和较低的诊断时延.
关键词: 工业物联网;故障诊断;轴承;一维卷积神经网络;联合特征
中图法分类号: TP181

中文引用格式: 刘立,朱健成,韩光洁,毕远国.基于 1D-CNN 联合特征提取的轴承健康监测与故障诊断.软件学报,2021,32(8):
2379−2390. http://www.jos.org.cn/1000-9825/6188.htm
英文引用格式: Liu L, Zhu JC, Han GJ, Bi YG. Bearing health monitoring and fault diagnosis based on joint feature extraction in
one-dimensional convolution neural network. Ruan Jian Xue Bao/Journal of Software, 2021,32(8):2379−2390 (in Chinese).
http://www. jos.org.cn/1000-9825/6188.htm

Bearing Health Monitoring and Fault Diagnosis Based on Joint Feature Extraction in 1D-
CNN

1
1
1
LIU Li , ZHU Jian-Cheng , HAN Guang-Jie , BI Yuan-Guo 2
1 (College of Internet of Things Engineering, Hohai University, Changzhou 213022, China)
2 (School of Computer Science and Engineering, Northeastern University, Shenyang 110169, China)
Abstract: Data-driven fault diagnosis models for specific mechanical equipment lack generalization capabilities. As a core component
of various types of machinery, the health status of bearings makes sense in analyzing derivative failures of different machinery. This study
proposes a bearing health monitoring and fault diagnosis algorithm based on 1D-CNN (one-dimensional convolution neural network) joint
feature extraction. The algorithm first partitions the original vibration signal of the bearing in segmentations. The signal segmentations are
used as feature learning spaces and input into the 1D-CNN in parallel to extract the representative feature domain under each working
condition. To avoid processing overlapping information generated by faults, a bearing health status discriminant model is built in advance
based on the feature domain sensitive to health status. If the health model recognizes that the bearing is not in a healthy state, the feature
domain will be reconstructed jointly with the original signal and coupled with an automatic encoder for failure mode classification.

∗ 基金项目: 国家重点研发计划(2017YFE0125300); 江苏省重点研发计划(BE2019648)
Foundation item: National Key Research and Development Program of China (2017YFE0125300); Key Research and Development
Program of Jiangsu Province (BE2019648)
本文由“泛在嵌入式智能系统”专题特约编辑郭兵教授、王泉教授、邓庆绪教授、陈铭松教授、张凯龙副教授推荐.
收稿时间: 2020-07-20; 修改时间: 2020-09-07; 采用时间: 2020-11-02; jos 在线出版时间: 2021-02-07