Maximization Aggregation Attention Convolutional Capsule Network: Intelligent Decoupling Diagnosis Method for Compound Faults in Electro-hydrostatic Actuators

doi:10.3901/JME.260131

Abstract

Abstract: Electro-hydrostatic actuators(EHAs), as complex mechatronic-hydraulic integrated systems, play a significant role in aerospace and related fields. However, due to their intricate internal structures and harsh working environments, compound faults are among the most common types of failures. These faults exhibit concealment, coupling effects, uncertainty, and complex causal relationships, often being mistaken for single faults, which can lead to missed or incorrect alarms. Considering the high cost of collecting data on compound faults in hydraulic systems, the limited perception of single-sensor information, and the weak fault characteristics, an intelligent compound fault decoupling diagnosis method for EHAs based on a maximized aggregation attention convolutional capsule network(MAACCN) is proposed . By leveraging feature-level fusion of multidimensional sensing information and deep feature extraction, the proposed model can accurately identify and decouple compound faults even when trained using only single-fault data. Verification using an EHA fault dataset shows that the proposed method achieves a subset accuracy of 97.4% with a Hamming loss as low as 0.025.

Key words: electro-hydrostatic actuator, composite fault, maximizing aggregation, attention convolutional capsule network, decoupling classification

CLC Number:

TG156

ZHAO Xiaoli, HU Yuanhao, SUN Hui, DENG Wenxiang, HU Jian, YAO Jianyong, LI Yang, SHAO Haidong. Maximization Aggregation Attention Convolutional Capsule Network: Intelligent Decoupling Diagnosis Method for Compound Faults in Electro-hydrostatic Actuators[J]. Journal of Mechanical Engineering, 2026, 62(4): 355-365.

References

[1] XU J， JIN W，GUO H，et al. Modeling and analysis of oil frictional loss in wet-type permanent magnet synchronous motor for aerospace electro-hydrostatic actuator[J]. Chinese Journal of Aeronautics，2023，36(11)：328-341.
[2] 张军辉，倪小昊，纵怀志，等. 机器人用电静液作动器研究现状与发展趋势[J]. 机械工程学报，2025，61(4)：273-289. ZHANG Junhui，NI Xiaohao，ZONG Huaizhi，et al. Research status and development trend of electrohydrostatic actuators for robots[J]. Journal of Mechanical Engineering，2025，61(4)：273-289.
[3] MIAO J，WANG J，WANG D，et al. Experimental investigation on electro-hydraulic actuator fault diagnosis with multi-channel residuals[J]. Measurement，2021，180：109544.
[4] MA Z，LIAO H，GAO J，et al. Physics-informed machine learning for degradation modeling of an electrohydrostatic actuator system[J]. Reliability Engineering & System Safety，2023，229：108898.
[5] HUANG K，WU S，LI F，et al. Fault diagnosis of hydraulic systems based on deep learning model with multirate data samples[J]. IEEE Transactions on Neural networks and learning systems，2021，33(11)：6789-6801.
[6] 吕飞，徐兵，张军辉. 转速对EHA泵柱塞副柱塞位姿及泄漏量影响仿真分析[J]. 机械工程学报，2018，54(20)：123-130. LÜ Fei，XU Bing，ZHANG Junhui. Simulative analysis of piston posture and piston/cylinder interface leakage of EHA pumps by the influence of rotating speed[J]. Journal of Mechanical Engineering，2018，54(20)：123-130.
[7] LI J，WANG W，ZHONG S，et al. HeMTAN：Hybrid task-adapted experts-based multi-task attention network for unseen compound fault decoupling diagnosis of rotating machinery[J]. Expert Systems with Applications，2024，252：124189.
[8] GÜLTEKIN Ö，CINAR E，ÖZKAN K，et al. Multisensory data fusion-based deep learning approach for fault diagnosis of an industrial autonomous transfer vehicle[J]. Expert Systems with Applications，2022，200：117055.
[9] 李文华，牛国波，刘羽佳. 基于遗传算法优化BP神经网络的液压系统故障诊断[J]. 机床与液压，2023，51(8)：159-164. LI Wenhua，NIU Guobo，LIU Yujia. Hydraulic system fault diagnosis based on genetic algorithm optimized BP neural network[J]. Machine Tool & Hydraulics，2023，51(8)：159-164.
[10] 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.
[11] CAO H，SU S，JING X，et al. Vibration mechanism analysis for cylindrical roller bearings with single/multi defects and compound faults[J]. Mechanical Systems and Signal Processing，2020，144：106903.
[12] HUANG W，SONG Z，ZHANG C，et al. Multi-source fidelity sparse representation via convex optimization for gearbox compound fault diagnosis[J]. Journal of Sound and Vibration，2021，496：115879.
[13] LIANG P，DENG C，WU J，et al. Intelligent fault diagnosis of rotating machinery via wavelet transform，generative adversarial nets and convolutional neural network[J]. Measurement，2020，159：107768.
[14] LIU J，HU Y，ZHU X，et al. Intelligent fault diagnosis for electro-hydrostatic actuator based on multisource information convolutional residual network[J]. Measurement Science and Technology，2024，35(6)：066114.
[15] ZHAO X，ZHU X，LIU J，et al. Model-assisted multi-source fusion hypergraph convolutional neural networks for intelligent few-shot fault diagnosis to electro-hydrostatic actuator[J]. Information Fusion，2024，104：102186.
[16] YANG C，CAI B，WU Q，et al. Digital twin-driven fault diagnosis method for composite faults by combining virtual and real data[J]. Journal of Industrial Information Integration，2023，33：100469.
[17] LIU J，XU H，PENG X，et al. Reliable composite fault diagnosis of hydraulic systems based on linear discriminant analysis and multi-output hybrid kernel extreme learning machine[J]. Reliability Engineering & System Safety，2023，234：109178.
[18] HUANG R，LI J，LIAO Y，et al. Deep adversarial capsule network for compound fault diagnosis of machinery toward multidomain generalization task[J]. IEEE Transactions on Instrumentation and Measurement，2020，70：1-11.
[19] SABOUR S，FROSST N，HINTON G E. Dynamic routing between capsules[J]. Advances in neural information processing systems，2017，30：09829.
[20] WANG Z，ZHENG L，DU W，et al. A novel method for intelligent fault diagnosis of bearing based on capsule neural network[J]. Complexity，2019，2019(1)：6943234.
[21] HINTON G E，SABOUR S，FROSST N. Matrix capsules with EM routing[C]// International conference on learning representations. Vancouver，OpenReview.net：2018.
[22] LI，X，YU S，LEI Y，et al. Dynamic vision-based machinery fault diagnosis with cross-modality feature alignment[J]. IEEE/CAA Journal of Automatica Sinica，2024，11(10)：2068-2081.
[23] HUANG R，LIAO Y，ZHANG S，et al. Deep decoupling convolutional neural network for intelligent compound fault diagnosis[J]. IEEE Access，2018，7：1848-1858.
[24] ZHAO Y，ZHOU M，ZHANG N，et al. Fault diagnosis of gas turbine based on matrix capsules with EM routing[J]. Systems Science & Control Engineering，2021，9(Suppl1.)：96-102.