《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (1): 59-68.DOI: 10.11772/j.issn.1001-9081.2024010043

• 人工智能 • 上一篇    下一篇

基于多路层次化混合专家模型的轴承故障诊断方法

徐欣然1,2, 张绍兵1,2,3(), 成苗1,2,3, 张洋1,2,3, 曾尚1,2   

  1. 1.中国科学院 成都计算机应用研究所,成都 610213
    2.中国科学院大学 计算机科学与技术学院,北京 100049
    3.深圳市中钞科信金融科技有限公司,广东 深圳 518206
  • 收稿日期:2024-01-17 修回日期:2024-03-21 接受日期:2024-03-21 发布日期:2024-05-09 出版日期:2025-01-10
  • 通讯作者: 张绍兵
  • 作者简介:徐欣然(1997—),男,四川成都人,硕士研究生,主要研究方向:人工智能、预测性维护、工业大数据;
    成苗(1983—),男,四川成都人,高级工程师,硕士,主要研究方向:人工智能、机器视觉;
    张洋(1985—),男,河南平顶山人,高级工程师,硕士,主要研究方向:机器视觉、人工智能、大数据;
    曾尚(1995—),男,湖北荆门人,博士研究生,主要研究方向:人工智能。
  • 基金资助:
    四川省科技成果转移转化示范项目(2023ZHCG0005);四川省科技计划项目(2023YFG0113)

Bearings fault diagnosis method based on multi-pathed hierarchical mixture-of-experts model

Xinran XU1,2, Shaobing ZHANG1,2,3(), Miao CHENG1,2,3, Yang ZHANG1,2,3, Shang ZENG1,2   

  1. 1.Chengdu Institute of Computer Application,Chinese Academy of Sciences,Chengdu Sichuan 610213,China
    2.School of Computer Science and Technology,University of Chinese Academy of Sciences,Beijing 100049,China
    3.Shenzhen CBPM-KEXIN Banking Technology Company Limited,Shenzhen Guangdong 518206,China
  • Received:2024-01-17 Revised:2024-03-21 Accepted:2024-03-21 Online:2024-05-09 Published:2025-01-10
  • Contact: Shaobing ZHANG
  • About author:XU Xinran, born in 1997, M. S. candidate. His research interests include artificial intelligence, predictive maintenance, industrial big data.
    CHENG Miao, born in 1983, M. S., senior engineer. His research interests include artificial intelligence, machine vision.
    ZHANG Yang, born in 1985, M. S., senior engineer. His research interests include machine vision, artificial intelligence, big data.
    ZENG Shang, born in 1995, Ph. D. candidate. His research interests include artificial intelligence.
  • Supported by:
    Sichuan Provincial Science and Technology Achievement Transfer and Transformation Demonstration Project(2023ZHCG0005);Sichuan Science and Technology Program(2023YFG0113)

摘要:

针对滚动轴承故障诊断中处理复杂工况准确率较低的问题,提出一个多任务学习(MTL)模型,即多路层次化混合专家(MHMoE)模型,以及对应的层次化训练模式。该模型结合多阶段、多任务联合训练,实现了层次化的信息共享模式,并在普通MTL模式的基础上进一步提升了模型的泛化性和故障识别准确率,使模型能同时在复杂与简单的数据集上出色地完成任务,同时,结合一维ResNet的瓶颈层结构,在保证网络深度的同时,也规避梯度爆炸与梯度消失等问题,从而能充分地提取数据集的相关特征。以帕德博恩大学轴承故障数据集(PU)为测试数据集设计的实验的结果表明,在不同工况复杂度下,与不使用MTL的单任务混合专家单元结构(OMoE)-ResNet18模型相比,所提模型的准确率提升5.45~9.30个百分点;而与集成经验模态分解的Hilbert谱变换方法(EEMD-Hilbert)、MMoE (Multi-gate Mixture-of-Experts)和多尺度多任务注意力卷积神经网络(MSTACNN)等模型相比,所提模型的准确率至少提升3.21~16.45个百分点。

关键词: 轴承故障诊断, 预测性维护, 多任务学习, 深度学习, 卷积神经网络

Abstract:

In response to the issue of low accuracy in handling complex work conditions in rolling bearing fault diagnosis, a Multi-Task Learning (MTL) model naming as Multi-pathed Hierarchical Mixture-of-Experts (MHMoE), and the corresponding hierarchical training mode were proposed. In this model, by combining multi-stage, multi-task joint training, a hierarchical information sharing mode was achieved. The model's generalization and fault recognition accuracy were further improved on the basis of the ordinary MTL mode, enabling the model to perform tasks on both complex and simple datasets excellently. Meanwhile, by incorporating the bottleneck layer structure of one-dimensional ResNet, the depth of the network was ensured while avoiding issues such as vanishing and exploding gradients, so as to extract relevant features of the dataset fully. Experimental results on the Paderborn University bearing fault dataset (PU) as the test dataset demonstrate that under varying degrees of working complexity, compared to the OMoE (One-gate Mixture-of-Experts) -ResNet18 model without MTL, the proposed model has the accuracy improved by 5.45 to 9.30 percentage points. Compared to the models such as Ensemble Empirical Mode Decomposition Hilbert spectral transform (EEMD-Hilbert), MMoE (Multi-gate Mixture-of-Experts), and Multi-Scale multi-Task Attention Convolutional Neural Network (MSTACNN), the proposed model has the accuracy improved by 3.21 to 16.45 percentage points at least.

Key words: bearing fault diagnosis, Predictive Maintenance (PdM), Multi-Task Learning (MTL), deep learning, Convolutional Neural Network (CNN)

中图分类号: