HF-MSCN: a high frequency-multiscale cascade network for bearing fault diagnosis

可解释性 计算机科学 级联 断层(地质) 噪音(视频) 卷积神经网络 模式识别(心理学) 人工智能 块(置换群论) 特征(语言学) 特征提取 深度学习 背景噪声 工程类 电信 数学 语言学 哲学 几何学 化学工程 地震学 图像(数学) 地质学
作者
Alaeldden Abduelhadi,Haopeng Liang,Jie Cao,Peng Chen
出处
期刊:Measurement Science and Technology [IOP Publishing]
卷期号:35 (11): 116120-116120
标识
DOI:10.1088/1361-6501/ad6893
摘要

Abstract In the field of data-driven fault diagnosis (FD), deep learning methods have proven their excellent performance, especially when dealing with complex signals from rotating equipment such as bearings. However, fault features in vibration signals are often mixed with noise features and distributed at different frequency scales, posing challenges for effective feature extraction. In order to solve this problem, this paper proposes a high frequency-multiscale cascade network (HF-MSCN), which enhances the noise suppression and feature learning capability of the model by combining a high-frequency convolutional block (HFCB) with a multi-scale cascade block (MSCB). HFCB effectively suppresses high-frequency noise through wide convolutional layers and self-attention mechanisms while still retaining essential high-frequency fault signals. MSCB enhances the interaction between convolutional layers at different scales by cascading the layers at different scales and strengthens the model’s ability to capture subtle fault features, especially when processing periodic fault pulse signals. Finally, we investigate the internal functioning of the network using time–frequency analysis methods in signal processing to improve the interpretability of deep learning methods in FD applications and further verify the enhanced effect of HFCB and MSCB on feature extraction. We validate the effectiveness of HF-MSCN on the case western reserve university dataset as well as a self-constructed bearing composite fault dataset, and the experimental results demonstrate that the network exceeds the performance of six state-of-the-art fault diagnostic methods in high-noise environments.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI

祝大家在新的一年里科研腾飞
更新
大幅提高文件上传限制,最高150M (2024-4-1)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
無羁完成签到 ,获得积分10
1秒前
nonosense发布了新的文献求助10
5秒前
学无止境完成签到,获得积分10
7秒前
FashionBoy应助斑其采纳,获得10
9秒前
健壮凡桃完成签到,获得积分10
9秒前
10秒前
11秒前
15秒前
Harry完成签到,获得积分10
17秒前
benny279完成签到,获得积分10
18秒前
斯文败类应助honey采纳,获得10
18秒前
皮皮发布了新的文献求助10
18秒前
Harry发布了新的文献求助10
21秒前
健壮凡桃发布了新的文献求助10
22秒前
张陶求完成签到,获得积分20
23秒前
杳鸢应助张陶求采纳,获得10
26秒前
27秒前
27秒前
30秒前
科研通AI2S应助ccalvintan采纳,获得10
31秒前
日子语梦完成签到,获得积分20
33秒前
33秒前
斑其发布了新的文献求助10
34秒前
lucky完成签到,获得积分10
35秒前
日子语梦发布了新的文献求助10
37秒前
41秒前
gzdxcll应助科研通管家采纳,获得20
41秒前
乐乐应助科研通管家采纳,获得10
41秒前
渔舟唱晚应助科研通管家采纳,获得10
41秒前
李爱国应助科研通管家采纳,获得10
41秒前
斑其完成签到,获得积分10
41秒前
Ava应助科研通管家采纳,获得30
41秒前
渔舟唱晚应助科研通管家采纳,获得10
41秒前
田様应助origin采纳,获得10
41秒前
Jasper应助科研通管家采纳,获得10
41秒前
烟花应助科研通管家采纳,获得10
42秒前
科研通AI2S应助丸橙采纳,获得10
46秒前
46秒前
木雨亦潇潇完成签到,获得积分10
48秒前
指导灰完成签到 ,获得积分10
49秒前
高分求助中
Востребованный временем 2500
Aspects of Babylonian celestial divination: the lunar eclipse tablets of Enūma Anu Enlil 1000
Kidney Transplantation: Principles and Practice 1000
Separation and Purification of Oligochitosan Based on Precipitation with Bis(2-ethylhexyl) Phosphate Anion, Re-Dissolution, and Re-Precipitation as the Hydrochloride Salt 500
The Restraining Hand: Captivity for Christ in China 500
Encyclopedia of Mental Health Reference Work 400
Mercury and Silver Mining in the Colonial Atlantic 300
热门求助领域 (近24小时)
化学 医学 生物 材料科学 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 物理化学 催化作用 细胞生物学 免疫学 冶金
热门帖子
关注 科研通微信公众号,转发送积分 3376007
求助须知:如何正确求助?哪些是违规求助? 2992295
关于积分的说明 8750295
捐赠科研通 2676626
什么是DOI,文献DOI怎么找? 1466189
科研通“疑难数据库(出版商)”最低求助积分说明 678137
邀请新用户注册赠送积分活动 669801