非完备数据驱动的装备复合故障智能解耦方法

doi:10.3901/JME.2024.24.045

机械工程学报 ›› 2024, Vol. 60 ›› Issue (24): 45-55.doi: 10.3901/JME.2024.24.045

扫码分享

非完备数据驱动的装备复合故障智能解耦方法

李巍华^1,2, 蓝昊¹, 陈祝云^1,2, 黄如意^2,3

1. 华南理工大学机械与汽车工程学院广州 510641;
2. 人工智能与数字经济广东省实验室广州广州 510335;
3. 华南理工大学吴贤铭智能工程学院广州 511442

收稿日期:2024-01-15 修回日期:2024-10-16 出版日期:2024-12-20 发布日期:2025-02-01
作者简介:李巍华，男，1973年出生，博士，教授，博士研究生导师。主要研究方向为工业智能、工业大数据、数字孪生、装备智能运维、预测性维护与健康管理、汽车智能驾驶(环境感知、路径规划与决策)。E-mail：whlee@scut.edu.cn;蓝昊，男，1998年出生，博士研究生。主要研究方向为装备智能故障诊断(复合故障智能诊断、可解释智能诊断)。E-mail：lanhao2021@163.com;黄如意(通信作者)，男，1992年出生，博士，助理研究员。主要研究方向为装备智能故障诊断与预测性维护(复合故障智能解耦、装备剩余寿命预测)、工业大数据(多源信息融合)。E-mail：snowxiaoyu@hotmail.com
基金资助:
国家自然科学基金联合基金(U23B6001)、国家重点研发计划(2023YFF0713400)和“十四五”装备预研共用技术(50910050302)资助项目。

Incomplete Data Driven Intelligent Compound Fault Diagnosis Method for Machinery

LI Weihua^1,2, LAN Hao¹, CHEN Zhuyun^1,2, HUANG Ruyi^2,3

1. School of Mechanical & Automotive Engineering, South China University of Technology, Guangzhou 510641;
2. Guangdong Artificial Intelligent and Digital Economy Laboratory Guangzhou, Guangzhou 510335;
3. Shien-Ming Wu School of Intelligent Engineering, South China University of Technology, Guangzhou 511442

Received:2024-01-15 Revised:2024-10-16 Online:2024-12-20 Published:2025-02-01

摘要/Abstract

摘要： 数据驱动的故障诊断方法在海量数据处理、判别特征学习和精确模式识别等方面具有显著优势，既为装备智能运维带来新的机遇，也面临着严峻的挑战：① 当前数据驱动方法的可靠性与有效性依赖于完备的带标签故障数据。然而工业场景下，装备监测数据存在标签信息缺失、故障数据欠完备、数据可利用率低等问题；② 复合故障是装备的典型故障。在缺乏复合故障数据的条件下，复合故障智能诊断的难度急剧增加，且鲜有针对复合故障智能诊断模型可解释性的研究。为应对上述挑战，提出一种非完备数据驱动的装备复合故障智能解耦方法。首先，结合小波核卷积层可从振动信号中提取可解释性特征的优势以及胶囊层在复合故障智能解耦方面的能力，构建小波胶囊网络；在此基础上，以非完备故障数据(仅含正常和单一故障样本, 缺乏复合故障样本)为输入，训练所提小波胶囊网络；最后，以某车型汽车五档变速器为研究对象，验证所提方法在复合故障智能解耦方面的有效性和可靠性，并对小波胶囊网络模型所提取的特征展开可解释性分析，从而提高模型诊断结果的可信度。试验结果表明，所提方法取得了较高的诊断精度，其习得的特征具有一定的可解释性，可为非完备数据下装备的复合故障诊断提供新的途径。

关键词: 复合故障故障诊断, 非完备数据, 小波核卷积, 胶囊网络, 可解释性分析

Abstract: Data driven fault diagnosis methods have significant advantages in mass data processing, discriminative feature learning and precision pattern recognition, which bring not only new opportunities for intelligent maintenance of machinery but also unsolved challenges：① The effectiveness and reliability of current data driven methods depend on the completeness of labelled fault data, whereas, in industrial scenarios, it is the often case that the label or repair information is missing, the fault data is lacking and the availability of the monitoring data is low；② Compound faults are common faults in equipment, resulting in the difficulty of intelligent fault diagnosis increasing dramatically when the compound fault data are unavailable, but there are few studies that focus on the interpretability of intelligent compound fault diagnosis models. To address the above challenges, an incomplete data driven intelligent compound fault diagnosis method, named wavelet capsule network, is proposed for industrial equipment. First, the wavelet capsule network is constructed by combining the wavelet kernel convolutional layer (WKL) and the capsule layers. Specifically, WKL is used to endow the model with the ability to extract interpretable features from vibration signals, and the capsule layers are introduced to decouple compound faults. Second, the wavelet capsule network is trained by incomplete fault data (just contains normal and single fault samples, and lacks compound fault samples). Finally, an experiment carried out on a five-speed transmission is used to validate the effectiveness of the proposed method. Furthermore, the interpretability analysis is performed on the features extracted by the proposed model to improve the credibility of the diagnosis result. The experimental results show that the proposed method achieves higher diagnosis accuracy than compared methods and the features extracted by the model are interpretable to a certain extent which might provide a novel solution to decoupling the compound fault with incomplete data.

Key words: compound fault diagnosis, incomplete data, wavelet kernel convolution, capsule network, interpretability analysis

中图分类号:

TH17
TP183

李巍华, 蓝昊, 陈祝云, 黄如意. 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, CHEN Zhuyun, HUANG Ruyi. Incomplete Data Driven Intelligent Compound Fault Diagnosis Method for Machinery[J]. Journal of Mechanical Engineering, 2024, 60(24): 45-55.

参考文献

[1] 雷亚国，贾峰，孔德同，等. 大数据下机械智能故障诊断的机遇与挑战[J]. 机械工程学报，2018，54(5)：94-104. LEI Yaguo，JIA Feng，KONG Detong，et al. Opportunities and challenges of machinery intelligent fault diagnosis in big data era[J]. Journal of Mechanical Engineering，2018，54(5)：94-104.
[2] LI W，HUANG R，LI J，et al. A perspective survey on deep transfer learning for fault diagnosis in industrial scenarios：Theories，applications and challenges[J]. Mechanical Systems and Signal Processing，2022，167：108487.
[3] 陈是扦，彭志科，周鹏. 信号分解及其在机械故障诊断中的应用研究综述[J]. 机械工程学报，2020，56(17)：91-107.CHEN Shiqian，PENG Zhike，ZHOU Peng. Review of signal decomposition theory and its applications in machine fault diagnosis[J]. Journal of Mechanical Engineering，2020，56(17)：91-107.
[4] LI N，HUANG W，GUO W，et al. Multiple enhanced sparse decomposition for gearbox compound fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement，2020，69(3)：770-781.
[5] ZHAO D，CHENG W，GAO R X，et al. Generalized vold–Kalman filtering for nonstationary compound faults feature extraction of bearing and gear[J]. IEEE Transactions on Instrumentation and Measurement，2020，69(2)：401-410.
[6] CHEN Guohua，QIE Longfei，ZHANG Aijun，et al. Improved CICA algorithm used for single channel compound fault diagnosis of rolling bearings[J]. Chinese Journal of Mechanical Engineering，2016，29(1)：204-211.
[7] TANG G，WANG Y，HUANG Y，et al. Compound bearing fault detection under varying speed conditions with virtual multichannel signals in angle domain[J]. IEEE Transactions on Instrumentation and Measurement，2020，69(8)：5535-5545.
[8] CUI L，SUN Y，ZHANG J，et al. Adapted dictionary-free orthogonal matching pursuit and 0-1 programming to solve the isolation and diagnosis of bearing and gear compound faults[J]. Measurement，2021，178：109331.
[9] SUN G，WANG Y，SUN C，et al. Intelligent detection of a planetary gearbox composite fault based on adaptive separation and deep learning[J]. Sensors (Basel，Switzerland)，2019，19(23)：5222.
[10] WANG S，XIANG J，ZHONG Y，et al. A data indicator-based deep belief networks to detect multiple faults in axial piston pumps[J]. Mechanical Systems and Signal Processing，2018，112：154-170.
[11] SHAO H，LIN J，ZHANG L，et al. A novel approach of multisensory fusion to collaborative fault diagnosis in maintenance[J]. Information Fusion，2021，74：65-76.
[12] SHAO H，JIANG H，ZHANG H，et al. Rolling bearing fault feature learning using improved convolutional deep belief network with compressed sensing[J]. Mechanical Systems and Signal Processing，2018，100：743-765.
[13] HUANG R，LI W，CUI L. An intelligent compound fault diagnosis method using one-dimensional deep convolutional neural network with multi-label classifier[C/OL]//2019 IEEE International Instrumentation and Measurement Technology Conference (I2MTC). Auckland，New Zealand：IEEE，2019：1-6.
[14] LIANG P，DENG C，WU J，et al. Compound fault diagnosis of gearboxes via multi-label convolutional neural network and wavelet transform[J]. Computers in Industry，2019，113：103132.
[15] JIN Y，QIN C，HUANG Y，et al. Actual bearing compound fault diagnosis based on active learning and decoupling attentional residual network[J]. Measurement，2021，173：108500.
[16] LEI Y，YANG B，JIANG X，et al. Applications of machine learning to machine fault diagnosis：A review and roadmap[J]. Mechanical Systems and Signal Processing，2020，138：106587.
[17] HUANG R，XIA J，ZHANG B，et al. Compound fault diagnosis for rotating machinery：State-of-the-art，challenges，and opportunities[J]. Journal of Dynamics，Monitoring and Diagnostics，2023，2(1)：13-29.
[18] ZHAO K，JIANG H，WANG K，et al. Joint distribution adaptation network with adversarial learning for rolling bearing fault diagnosis[J]. Knowledge-Based Systems，2021，222：106974.
[19] ZHOU F，YANG S，FUJITA H，et al. Deep learning fault diagnosis method based on global optimization GAN for unbalanced data[J]. Knowledge-Based Systems，2020，187：104837.
[20] LIU J，ZHANG C，JIANG X. Imbalanced fault diagnosis of rolling bearing using improved MsR-GAN and feature enhancement-driven CapsNet[J]. Mechanical Systems and Signal Processing，2022，168：108664.
[21] LI T，ZHAO Z，SUN C，et al. WaveletKernelNet：An interpretable deep neural network for industrial intelligent diagnosis[J]. IEEE Transactions on Systems，Man，and Cybernetics：Systems，2022，52(4)：2302-2312.
[22] HUANG R，LIAO Y，ZHANG S，et al. Deep decoupling convolutional neural network for intelligent compound fault diagnosis[J]. IEEE Access，2019，7：1848-1858.
[23] 赵学智，陈统坚，叶邦彦. 奇异值分解对连续Morlet小波变换的压缩和提纯[J]. 机械工程学报，2015，51(16)：57-70.ZHAO Xuezhi，CHEN Tongjian，YE Bangyan. Purification and compression of continuous morlet wavelet transform based on singular value decomposition[J]. Journal of Mechanical Engineering，2015，51(16)：57-70.
[24] 陈祝云，钟琪，黄如意，等. 基于增强迁移卷积神经网络的机械智能故障诊断[J]. 机械工程学报，2021，57(21)：96-105.CHEN Zhuyun，ZHONG Qi，HUANG Ruyi，et al. Intelligent fault diagnosis for machinery based on enhanced transfer convolutional neural network[J]. Journal of Mechanical Engineering，2021，57(21)：96-105.
[25] SABOUR S，FROSST N，HINTON G E. Dynamic routing between capsules[J]. arXiv，2017：3859-3869.
[26] ZHANG W，PENG G，LI C，et al. A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals[J]. Sensors，2017，17(2)：425.
[27] SELVARAJU R R，COGSWELL M，DAS A，et al. Grad-CAM：Visual explanations from deep networks via gradient-based localization[J]. International Journal of Computer Vision，2020，128(2)：336-359.

非完备数据驱动的装备复合故障智能解耦方法

Incomplete Data Driven Intelligent Compound Fault Diagnosis Method for Machinery

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	李响, 陈欣瑞, 雷亚国, 李乃鹏, 杨彬, 俞舒鹏. 动态视觉赋能的非接触式装备迁移诊断[J]. 机械工程学报, 2024, 60(24): 1-10.
[2]	史丽晨, 史炜椿, 王海涛, 李金阳, 刘亚雄. 基于DRSN-BiLSTM模型的刀具磨损预测方法研究[J]. 机械工程学报, 2024, 60(24): 66-74.
[3]	段荣, 江志刚, 张华, 龚青山, 李明遥. 机电装备全工作域能效运维的难点与挑战[J]. 机械工程学报, 2024, 60(23): 354-364.
[4]	凡云峰, 俞铭宏, 王浩, 马波, 冯坤, 江志农, 胡明辉. 面向全恢复系数的碰摩故障接触力分析与试验研究[J]. 机械工程学报, 2024, 60(22): 403-414.
[5]	严彤彤, 王冬, 彭志科, 雷亚国. 基于谱幅融合广义健康指数的可解释装备退化评估优化模型研究进展[J]. 机械工程学报, 2024, 60(18): 1-16.
[6]	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.
[7]	李志鹏, 马天雨, 刘金平, 向青松, 唐俊杰. 基于非对称性对抗训练的多源域自适应智能故障诊断方法[J]. 机械工程学报, 2024, 60(18): 76-88.
[8]	任元强, 严嘉慧, 袁慎芳, 张巾巾. 时变影响下导波健康HMM的损伤定量诊断方法[J]. 机械工程学报, 2024, 60(16): 54-61.
[9]	严如强, 商佐港, 王志颖, 许文纲, 赵志斌, 王诗彬, 陈雪峰. 可解释人工智能在工业智能诊断中的挑战和机遇：先验赋能[J]. 机械工程学报, 2024, 60(12): 1-20.
[10]	严如强, 周峥, 杨远贵, 李亚松, 胡晨烨, 陶治宇, 赵志斌, 王诗彬, 陈雪峰. 可解释人工智能在工业智能诊断中的挑战和机遇：归因解释[J]. 机械工程学报, 2024, 60(12): 21-40.
[11]	袁静, 任港星, 蒋会明, 赵倩, 魏臣隽, 朱骏. 基于多元提升核神经网络的机械故障诊断方法及其特征提取可解释性研究[J]. 机械工程学报, 2024, 60(12): 51-64.
[12]	陈钱, 陈康康, 董兴建, 皇甫一樊, 彭志科, 孟光. 一种面向机械设备故障诊断的可解释卷积神经网络[J]. 机械工程学报, 2024, 60(12): 65-76.
[13]	王义, 丁嘉凯, 孙浩然, 秦毅, 汤宝平. 基于深度信号分离的航空发动机可解释智能诊断方法[J]. 机械工程学报, 2024, 60(12): 77-89.
[14]	王诗彬, 王世傲, 陈雪峰, 黄海, 安波涛, 赵志斌, 刘永泉, 李应红. 可解释性智能监测诊断网络构造及航空发动机整机试车与中介轴承诊断应用[J]. 机械工程学报, 2024, 60(12): 90-106.
[15]	余晓霞, 汤宝平, 魏静, 张志刚. 大转数波动条件下机理约束图权重增强网络的航空发动机附件机匣故障诊断方法[J]. 机械工程学报, 2024, 60(12): 126-136.