感知代理学习开集域泛化机械故障智能诊断方法

doi:10.3901/JME.260696

机械工程学报 ›› 2026, Vol. 62 ›› Issue (13): 220-230.doi: 10.3901/JME.260696

扫码分享

感知代理学习开集域泛化机械故障智能诊断方法

江星星^1,2, 宋秋昱², 邢黄坤², 张龙¹, 叶小芬³, 朱忠奎²

1. 华东交通大学山区土木工程安全与韧性全国重点实验室南昌 330013;
2. 苏州大学轨道交通学院苏州 215131;
3. 中车戚墅堰机车车辆工艺研究所股份有限公司常州 213011

收稿日期:2025-06-06 修回日期:2025-12-15 发布日期:2026-08-28
作者简介:江星星，男，1989年出生，博士，教授，博士研究生导师。主要研究方向为机械装备故障诊断、载运工具智能运维。E-mail：jiangxx@suda.edu.cn;朱忠奎(通信作者)，男，1974年出生，博士，教授，博士研究生导师。主要研究方向为车辆系统动力学与控制、振动测试和信号处理、载运工具智能运维。E-mail：zhuzhongkui@suda.edu.cn
基金资助:
华东交通大学山区土木工程安全与韧性全国重点实验室开放课题(HJGZ2023112)，国家自然科学基金(52172406、52575132)，江西省自然科学基金重点(20224ACB204017)资助项目。

Perception Poxy Learning-based Open Set Domain Generalization Method for Intelligent Diagnosis of Mechanical Faults

JIANG Xingxing^1,2, SONG Qiuyu², XING Huangkun², ZHANG Long¹, YE Xiaofen³, ZHU Zhongkui²

1. State Key Laboratory of Safety and Resilience of Civil Engineering in Mountain Area, East China Jiaotong University, Nanchang 330013;
2. School of Rail Transportation, Soochow University, Suzhou 215131;
3. CRRC Qishuyan Institute Co., Ltd., Changzhou 213011

Received:2025-06-06 Revised:2025-12-15 Published:2026-08-28

摘要/Abstract

摘要： 机械设备运行工况的复杂多变性不仅会导致状态数据分布存在显著的域偏移问题，更会给诊断模型带来未知工况下未知新型故障诊断的巨大挑战。因此，提升智能诊断模型在开集场景下的泛化能力对实际工程应用中模型在线诊断的有效决策具有重要价值和意义。受人类学习新事物方式的启发，该论文提出了一种感知代理学习开集域泛化机械智能故障诊断方法，旨在提高未知工况下机械故障智能诊断的有效性。具体地，一方面，该方法通过已知类感知对抗学习策略摆脱训练样本被预设为未知类别的假设，从而构建已知与未知故障类别间的决策边界。另一方面，该方法构建了未知类代理，设计了开放式代理对比学习策略以增强样本特征的类内聚集性和类间判别性。基于两组机械故障数据集的实验案例验证了提出方法在开集域泛化故障诊断任务中相比现有先进的对比方法的优越性能。

关键词: 对抗学习, 对比学习, 机械故障诊断, 开集域泛化

Abstract: The complex and variable operating conditions of mechanical equipment not only lead to significant domain shift in the distribution of state data, but also pose great challenges for diagnostic models to identify unknown faults under unseen conditions. Therefore, enhancing the generalization ability of intelligent diagnostic models in open-set scenarios is of great importance for effective real-time diagnosis in practical engineering applications. Inspired by how humans learn new things, this paper proposes a perception proxy-based approach for open-set domain generalization in mechanical intelligent fault diagnosis, aiming to improve the effectiveness of diagnosing mechanical faults under unknown conditions. Specifically, the method employs known class perception adversarial learning strategies to eliminate the assumption that training samples are pre-set as unknown categories, thereby establishing decision boundaries between known and unknown fault classes. Additionally, the approach constructs unknown class proxies and designs an open-set proxy contrastive learning strategy to enhance sample feature compactness within classes and discriminability between classes. Experimental results on two mechanical fault datasets demonstrate that the proposed method outperforms existing advanced contrastive techniques in open-set domain generalization fault diagnosis tasks.

Key words: adversarial learning, contrastive learning, mechanical fault diagnosis, open set domain generalization

中图分类号:

TH17

江星星, 宋秋昱, 邢黄坤, 张龙, 叶小芬, 朱忠奎. 感知代理学习开集域泛化机械故障智能诊断方法[J]. 机械工程学报, 2026, 62(13): 220-230.

JIANG Xingxing, SONG Qiuyu, XING Huangkun, ZHANG Long, YE Xiaofen, ZHU Zhongkui. Perception Poxy Learning-based Open Set Domain Generalization Method for Intelligent Diagnosis of Mechanical Faults[J]. Journal of Mechanical Engineering, 2026, 62(13): 220-230.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260696

http://www.cjmenet.com.cn/CN/Y2026/V62/I13/220

参考文献

[1] 李巍华，蓝昊，陈祝云，等. 非完备数据驱动的装备复合故障智能解耦方法[J]. 机械工程学报，2024，60(24)：45-55. LI Weihua，LAN Hao，XHEN Zhuyun，et al. Incomplete data driven intelligent compound fault diagnosis method for machinery[J]. Journal of Mechanical Engineering，2024，60(24)：45-55.
[2] 邵海东，颜深，刘政武，等. 面向小样本场景的风机多增量故障诊断[J]. 机械工程学报，2025，61(10)：230-240. SHAO Haidong，YAN Shen，LIU Zhengwu，et al. Multi-incremental fault diagnosis of wind turbine towards small sample scenario[J]. Journal of Mechanical Engineering，2025，61(10)：230-240.
[3] Rombach K，Michau G，Fink O. Controlled generation of unseen faults for partial and open-partial domain adaptation[J]. Reliability Engineering & System Safety，2023，230：108857.
[4] Peng Peng，Lu Jiaxun，Xie Tingyu，et al. Open-set fault diagnosis via supervised contrastive learning with negative out-of-distribution data augmentation[J]. IEEE Transactions on Industrial Informatics，2022，19(3)：2463-2473.
[5] Yu Xiaolei，Zhao Zhibin，Zhang Xingwu，et al. Deep-learning-based open set fault diagnosis by extreme value theory[J]. IEEE Transactions on Industrial Informatics，2022，18(1)：185–196.
[6] 雷亚国，杨彬，李乃鹏，等. 跨设备的机械故障靶向迁移诊断方法[J]. 机械工程学报，2022，58(12)：1-9. LEI Yaguo，YANG Bin，LI Naipeng，et al. Targeted transfer diagnosis method across different machines[J]. Journal of Mechanical Engineering，2022，58(12)：1-9.
[7] 佘博，梁伟阁，秦奋起，等. 基于双识别器对抗的开放域自适应故障诊断方法[J]. 仪器仪表学报，2023，44(7)：325-334. SHE Bo，LIANG Weige，QIN Fwngqi，et al. Open set domain adaptation method based on adversarial dual classifiers for fault diagnosis[J]. Chinese Journal of Scientific Instrument，2023，44(7)：325-334.
[8] Zhu Jun，Huang Chenggeng，Shen Changqing，et al. Cross-domain open-set machinery fault diagnosis based on adversarial network with multiple auxiliary classifiers[J]. IEEE Transactions on Industrial Informatics，2022，18(11)：8077–8086.
[9] DENG Yafei，LV Jun，HUANG Delin，et al. Combining the theoretical bound and deep adversarial network for machinery open-set diagnosis transfer[J]. Neurocomputing，2023，548：126391.
[10] SU Zuqiang，JIANG Weilong，CHEN Kai，et al. Multi-adversarial deep transfer network for multi-source open-set fault diagnosis of rotating machinery with category shift[J]. Knowledge-based Systems，2023，282：111106.
[11] Mo Zhenling，Zhang Zijun，Miao Qiang，et al. Sparsity-constrained invariant risk minimization for domain generalization with application to machinery fault diagnosis modeling[J]. IEEE Transactions on Cybernetics，2024，54(3)：1547–1559.
[12] Zhao Chao and Shen Weiming. Adaptive open set domain generalization network：Learning to diagnose unknown faults under unknown working conditions[J]. Reliability Engineering & System Safety，2022，226：108672.
[13] Yao Xufeng，Bai Yang，Zhang Xinyun，et al. PCL：Proxy-based contrastive learning for domain generalization[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022：7097–7107.
[14] Song Qiuyu，Jiang Xingxing，Liu Jie，et al. Contrast-assisted domain-specificity-removal network for semi-supervised generalization fault diagnosis[J]. IEEE Transactions on Neural Networks and Learning，2025，36(3)：5403–5416.
[15] Liu Dongong，Cui Lingli，Cheng Weidong. A review on deep learning in planetary gearbox health state recognition：Methods，applications，and dataset publication[J]. Measurement Science and Technology[J]. 2023，35(1)：012002.
[16] Wang Xiran，Zhang Jian，Qi Lei，et al. Generalizable decision boundaries：dualistic meta-learning for open set domain generalization[C]// Proceedings of the IEEE/CVF International Conference on Computer Vision 2023：11564–11573.
[17] Jiang Xingxing，Li Xuegang，Wang Qian，et al. Multi-sensor data fusion-enabled semi-supervised optimal temperature-guided PCL framework for machinery fault diagnosis[J]. Information Fusion，2024，101：102005.
[18] Wilcoxon F. Individual comparisons by ranking methods[J]. Biometrics Bulletin, 1945，1(6)：80-83.

感知代理学习开集域泛化机械故障智能诊断方法

Perception Poxy Learning-based Open Set Domain Generalization Method for Intelligent Diagnosis of Mechanical Faults

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

本文评价

[1]	林天骄, 宋浏阳, 崔玲丽, 王华庆. 非理想数据下基于时序对比注意力模型的寿命预测方法[J]. 机械工程学报, 2025, 61(18): 27-37.
[2]	邵海东, 肖一鸣, 钟翔, 韩特. 面向可信机械故障诊断的模型校准方法[J]. 机械工程学报, 2025, 61(17): 114-123.
[3]	赵轲, 叶敏, 王瑞欣, 陆海, 刘孟孟, 邵海东. 基于模糊域自适应的源自由域旋转机械故障诊断方法[J]. 机械工程学报, 2024, 60(18): 43-52.
[4]	陈彬强;张周锁;訾艳阳;何正嘉. 机械故障诊断的衍生增强离散解析小波分析框架[J]. , 2014, 50(17): 77-86.
[5]	秦毅;王家序;汤宝平. 基于迭代Hilbert变换的多分量信号解调方法研究及应用[J]. , 2009, 45(8): 37-44.
[6]	陈敏;胡茑庆;秦国军;安茂春. 参数调节随机共振在机械系统早期故障检测中的应用[J]. , 2009, 45(4): 131-135.
[7]	程军圣;于德介;杨宇. 基于EMD的能量算子解调方法及其在机械故障诊断中的应用[J]. , 2004, 40(8): 115-118.