用户名 密 码 登 录 忘记密码 免费注册

网站会员中心 | 设为首页 | 加入收藏

 您好，欢迎光临中国轴承行业网         退出登陆

会员中心 | 设为首页 | 加入收藏

协会

统计

国外文献 > 1.轴承 > 7. 智能方法与计算 > 神经网络

关键词搜索：

标 题： A Convolutional Neural Network with Hybrid Loss Function for Bearing Fault Diagnosis
基于混合损失函数的卷积神经网络轴承故障诊断
作 者： Dongrui Lv
Zhenbao Fu
Zhengchang Su
Haitao Ni
Huaqing Wang
Liuyang Song
关键词： Convolutional neural network  [卷积神经网络]
Cross-entropy margin loss  [ 交叉熵边际损失]
Bearing fault diagnosis  [ 轴承故障诊断]
出版年： 2024
文集名： Proceedings of the UNIfied Conference of DAMAS, IncoME and TEPEN Conferences (UNIfied 2023), Volume 1
联合国大会、DAMAS、IncoME和TEPEN会议联合会议（UNIfied 2023）会议纪要，第一卷
会议名： 2023 International Conference on Damage Assessment of Structures (DAMAS 2023), 2023 International Conference on Maintenance Engineering (IncoME 2023), 2023 Conference of the Efficiency and Performance Engineering Network (TEPEN 2023)
2023年结构损伤评估国际会议（DAMAS 2023）、2023年维护工程国际会议（IncoME 2023）、2023年效率与性能工程网络会议（TEPEN 2023）
届 次：
摘 要： This paper proposes a method based on a convolutional neural network (CNN) with hybrid loss function to solve a problem of bearing fault diagnosis in different working conditions. Firstly, Swish activation function is introduced into traditional convolutional neural network which can reduce the effect of vanishing gradient and optimize the training process. Secondly, a hybrid loss function combines margin loss function and cross entropy loss function (CEM) is used to improve the classification effect. Finally, the parameters of shallow layers pre-trained are frozen and the weights of deep layers are fine-tuned to accommodate the target domain sample distribution. In this paper, gray scale image is taken as the network input. The proposed method is experimentally verified on the bearing data of Case Western Reserve University and the fan data set. The experiment results show that the proposed method outperforms the comparing methods.

本文提出一种基于卷积神经网络（CNN）的混合损失函数方法，用于解决不同工况下轴承故障诊断的问题。首先，将Swish激活函数引入传统卷积神经网络中，以减少梯度消失的影响并优化训练过程。其次，采用混合损失函数，结合置信度损失函数与交叉熵损失函数（CEM），以提升分类效果。最后，冻结浅层参数，对深层权重进行微调，以适应目标域样本的分布特征。本文以灰度图像作为网络输入，将所提方法在Case Western Reserve大学的轴承数据集和风扇数据集上进行了实验验证。实验结果表明，该方法优于现有对比方法。
中文内容为网络AI翻译，专业术语未必准确，仅供参考。

VIP网员

联系我们 | 登录注册 | 会员管理 | 网员服务 | 服务信息 | 广告服务 | 高薪聘请 | 法律声明 | 关于我们

京公网安备 11010202009788    京ICP备2021015424号