航空学报 > 2024, Vol. 45 Issue (15): 630407-630407   doi: 10.7527/S1000-6893.2024.30407

基于注意力卷积胶囊网络的电液比例伺服阀故障诊断

胡渊豪1, 宋艺博1, 刘家辉1, 赵孝礼1,2(), 邓文翔1, 胡健1, 姚建勇1   

  1. 1.南京理工大学 机械工程学院,南京 210094
    2.重庆大学 高端装备机械传动全国重点实验室,重庆 400044
  • 收稿日期:2024-03-16 修回日期:2024-04-09 接受日期:2024-05-11 出版日期:2024-06-06 发布日期:2024-06-03
  • 通讯作者: 赵孝礼 E-mail:xlzhao@njust.edu.cn
  • 基金资助:
    国家重点研发计划(2021YFB2011300);国家自然科学基金(52205062);江苏省自然科学基金(BK20220950);中央高校基本科研业务费专项资金(30922010701);中国博士后科学基金(2022M711624);流体动力与机电系统国家重点实验室开放基金(GZKF-202218);机械传动国家重点实验室开放基金(SKLMT-MSKFKT-202220);智控实验室开放基金(ICL-2023-0305)

Fault diagnosis of electro-hydraulic proportional servo valves based on attention convolutional capsule networks

Yuanhao HU1, Yibo SONG1, Jiahui LIU1, Xiaoli ZHAO1,2(), Wenxiang DENG1, Jian HU1, Jianyong YAO1   

  1. 1.School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing  210094,China
    2.State Key Laboratory of Mechanical Transmission for Advanced Equipment,Chongqing University,Chongqing  400044,China
  • Received:2024-03-16 Revised:2024-04-09 Accepted:2024-05-11 Online:2024-06-06 Published:2024-06-03
  • Contact: Xiaoli ZHAO E-mail:xlzhao@njust.edu.cn
  • Supported by:
    National Key R&D Program of China(2021YFB2011300);National Natural Science Foundation of China(52205062);Natural Science Foundation of Jiangsu Province(BK20220950);Fundamental Research Funds for the Central Universities(30922010701);China Postdoctoral Science Foundation(2022M711624);Open Foundation of the State Key Laboratory of Fluid Power and Mechatronic Systems(GZKF-202218);the State Key Laboratory of Mechanical Transmissions(SKLMT-MSKFKT-202220);the Open Fund of Intelligent Control Laboratory(ICL-2023-0305)

摘要:

针对目前基于模型的比例伺服阀故障诊断方法存在参数未知、不易建模的问题,首次提出了注意力卷积胶囊网络(ACCN)算法,并设计了基于注意力卷积胶囊网络的比例伺服阀故障诊断新方法。该算法首先使用卷积神经网络和高效通道注意力机制,对原始的多通道一维时域信号进行特征提取与融合,然后将融合结果输入到后续的胶囊网络中进行特征再提取和故障分类,输出对应的诊断结果。比例伺服阀故障模拟试验台验证了所提算法与诊断方法的有效性,提出的方法在测试集上取得了100%的准确率,且通道注意力机制和胶囊网络的加入使模型的诊断效果得到了不同程度的提升。

关键词: 注意力机制, 卷积神经网络, 胶囊网络, 比例伺服阀, 故障诊断

Abstract:

To address the issues of unknown parameters and difficult modeling in current model-based proportional servo valve fault diagnosis methods, we propose the Attention Convolution Capsule Network (ACCN) algorithm for the first time and design a new proportional servo valve fault diagnosis method based on this network. This algorithm first uses convolutional neural networks and an efficient channel attention mechanism to extract and fuse features from the original multi-channel one-dimensional time-domain signals. The fused results are then input into the subsequent capsule network for further feature extraction and fault classification, yielding the corresponding diagnosis results. The effectiveness of the proposed algorithm and diagnostic method was validated on a proportional servo valve fault simulation test bench. The proposed method achieved 100% accuracy on the test set, and the inclusion of the channel attention mechanism and capsule network enhanced the diagnostic performance of the model to varying degrees.

Key words: attention mechanism, convolutional neural networks, capsule network, proportional servo valve, fault diagnosis

中图分类号: