On the Sparse Gradient Denoising Optimization of Neural Network Models for Rolling Bearing Fault Diagnosis Illustrated by a Ship Propulsion System

被引:5
|
作者
Wang, Shuangzhong [1 ]
Zhang, Ying [2 ]
Zhang, Bin [2 ]
Fei, Yuejun [3 ]
He, Yong [3 ]
Li, Peng [4 ]
Xu, Mingqiang [5 ]
机构
[1] Shanghai Maritime Univ, Logist Engn Coll, Shanghai 201306, Peoples R China
[2] Shanghai Maritime Univ, Coll Informat Engn, Shanghai 201306, Peoples R China
[3] SOA, East China Sea Ctr Stand & Metrol Technol, Shanghai 201306, Peoples R China
[4] SOA, East China Sea Forecasting Ctr, Shanghai 200136, Peoples R China
[5] CTTIC Big Data Shanghai Technol Co Ltd, Shanghai 201901, Peoples R China
基金
中国国家自然科学基金;
关键词
neural networks; sparse denoising; gradient optimization; rolling bearings fault diagnosis; ALGORITHM;
D O I
10.3390/jmse10101376
中图分类号
U6 [水路运输]; P75 [海洋工程];
学科分类号
0814 ; 081505 ; 0824 ; 082401 ;
摘要
The drive rolling bearing is an important part of a ship's system; the detection of the drive rolling bearing is an important component in ship-fault diagnosis, and machine learning methods are now widely used in the fault diagnosis of rolling bearings. However, training methods based on small batches have a disadvantage in that the samples which best represent the gradient descent direction can be disturbed by either other samples in the opposite direction or anomalies. Aiming at this problem, a sparse denoising gradient descent (SDGD) optimization algorithm, based on the impact values of network nodes, was proposed to improve the updating method of the batch gradient. First, the network is made sparse by using the node weight method based on the mean impact value. Second, the batch gradients are clustered via a distribution-density-based clustering method. Finally, the network parameters are updated using the gradient values after clustering. The experimental results show the efficiency and feasibility of the proposed method. The SDGD model can achieve up to a 2.35% improvement in diagnostic accuracy compared to the traditional network diagnosis model. The training convergence speed of the SDGD model improves by 2.16%, up to 17.68%. The SDGD model can effectively solve the problem of falling into the local optimum point while training a network.
引用
收藏
页数:25
相关论文
共 50 条
  • [1] A Novel Sparse Enhancement Neural Network for Rolling Bearing Fault Diagnosis
    Zhang, Yong
    Ye, Junjie
    Yang, Wenhu
    Shi, Jinwang
    He, Wangpeng
    Cai, Gaigai
    SHOCK AND VIBRATION, 2022, 2022
  • [2] Intelligent fault diagnosis method of rolling bearing based on stacked denoising autoencoder and convolutional neural network
    Che Changchang
    Wang Huawei
    Ni Xiaomei
    Fu Qiang
    INDUSTRIAL LUBRICATION AND TRIBOLOGY, 2020, 72 (07) : 947 - 953
  • [3] Extension neural network learning algorithms and models and their applications in fault diagnosis of rolling bearing
    Su Z.
    Ying Z.
    International Journal of Smart Home, 2016, 10 (03): : 201 - 210
  • [4] Fault diagnosis of rolling bearing based on BP neural network with fractional order gradient descent
    Jiao, Rui
    Li, Sai
    Ding, Zhixia
    Yang, Le
    Wang, Guan
    JOURNAL OF VIBRATION AND CONTROL, 2024, 30 (9-10) : 2139 - 2153
  • [5] A sparse denoising deep neural network for improving fault diagnosis performance
    Funa Zhou
    Tong Sun
    Xiong Hu
    Tianzhen Wang
    Chenglin Wen
    Signal, Image and Video Processing, 2021, 15 : 1889 - 1898
  • [6] A sparse denoising deep neural network for improving fault diagnosis performance
    Zhou, Funa
    Sun, Tong
    Hu, Xiong
    Wang, Tianzhen
    Wen, Chenglin
    SIGNAL IMAGE AND VIDEO PROCESSING, 2021, 15 (08) : 1889 - 1898
  • [7] Fault Diagnosis of Rolling Bearing on the Basis of Wavelet Neural Network
    Lin, Wu Song
    Liu Jianxin
    Lili
    ADVANCED MATERIALS, MECHANICS AND INDUSTRIAL ENGINEERING, 2014, 598 : 244 - 249
  • [8] A review on convolutional neural network in rolling bearing fault diagnosis
    Li, Xin
    Ma, Zengqiang
    Yuan, Zonghao
    Mu, Tianming
    Du, Guoxin
    Liang, Yan
    Liu, Jingwen
    MEASUREMENT SCIENCE AND TECHNOLOGY, 2024, 35 (07)
  • [9] Fault diagnosis of rolling bearing based on BSA neural network
    Du Wenliao
    Huang Chang
    Li Ansheng
    Gong Xiaoyun
    Wang Liangwen
    Wang Zhiyang
    2015 10TH INTERNATIONAL CONFERENCE ON INTELLIGENT SYSTEMS AND KNOWLEDGE ENGINEERING (ISKE), 2015, : 424 - 427
  • [10] Rolling Bearing Fault Diagnosis Based on BP Neural Network
    Yu, Chenglong
    Wang, Hongjun
    PROCEEDINGS OF TEPEN 2022, 2023, 129 : 576 - 595