Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networks

Abstract This paper propose a new fault identification method based on variable pooling multiscale CNN (VPMCNN), which solves the bearing industrial problem of huge variable features and inherent multiscale characteristics in acoustic emission (AE) signals. First, the pooling projection components (...

Full description

Saved in:
Bibliographic Details
Main Authors: Yue Zhang, Yang Yu, Zheng Yang, Qiang Liu
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-00573-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850190590901223424
author Yue Zhang
Yang Yu
Zheng Yang
Qiang Liu
author_facet Yue Zhang
Yang Yu
Zheng Yang
Qiang Liu
author_sort Yue Zhang
collection DOAJ
description Abstract This paper propose a new fault identification method based on variable pooling multiscale CNN (VPMCNN), which solves the bearing industrial problem of huge variable features and inherent multiscale characteristics in acoustic emission (AE) signals. First, the pooling projection components (PPCs) of the signals are obtained through the variable pooling layer. The PPCs consider the curse of invariant feature weight in traditional CNN pooling layer, and select the more weighted features to enhance the classifying quality. Second, an improved multiscale fusion feature module is introduced to further extract the hidden features, which is called fused components (FCs). The FCs aims to automatically extract multiple scale features using different filter sizes from raw acoustic signals. Then the GAP (Global Average Pooling) layer is performed to realize classification. Finally, the fault identification using the proposed algorithm is performed by testing the bearing AE signals with single operating condition and variable operating conditions, and the results show the effectiveness of the proposed method, compared with existing AE based bearing fault identification methods.
format Article
id doaj-art-8b442aef1d2e4a7c807a2d1cd89156f8
institution OA Journals
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-8b442aef1d2e4a7c807a2d1cd89156f82025-08-20T02:15:15ZengNature PortfolioScientific Reports2045-23222025-05-0115111610.1038/s41598-025-00573-7Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networksYue Zhang0Yang Yu1Zheng Yang2Qiang Liu3College of Information Science and Engineering, Shenyang University of TechnologyCollege of Information Science and Engineering, Shenyang University of TechnologyCollege of Information Science and Engineering, Shenyang University of TechnologyCollege of Information Science and Engineering, Shenyang University of TechnologyAbstract This paper propose a new fault identification method based on variable pooling multiscale CNN (VPMCNN), which solves the bearing industrial problem of huge variable features and inherent multiscale characteristics in acoustic emission (AE) signals. First, the pooling projection components (PPCs) of the signals are obtained through the variable pooling layer. The PPCs consider the curse of invariant feature weight in traditional CNN pooling layer, and select the more weighted features to enhance the classifying quality. Second, an improved multiscale fusion feature module is introduced to further extract the hidden features, which is called fused components (FCs). The FCs aims to automatically extract multiple scale features using different filter sizes from raw acoustic signals. Then the GAP (Global Average Pooling) layer is performed to realize classification. Finally, the fault identification using the proposed algorithm is performed by testing the bearing AE signals with single operating condition and variable operating conditions, and the results show the effectiveness of the proposed method, compared with existing AE based bearing fault identification methods.https://doi.org/10.1038/s41598-025-00573-7Convolutional neural networkFault IdentificationMultiscale FusionAcoustic emissionPooling projection components, Variable operating conditions
spellingShingle Yue Zhang
Yang Yu
Zheng Yang
Qiang Liu
Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networks
Scientific Reports
Convolutional neural network
Fault Identification
Multiscale Fusion
Acoustic emission
Pooling projection components, Variable operating conditions
title Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networks
title_full Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networks
title_fullStr Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networks
title_full_unstemmed Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networks
title_short Rolling bearing fault identification with acoustic emission signal based on variable-pooling multiscale convolutional neural networks
title_sort rolling bearing fault identification with acoustic emission signal based on variable pooling multiscale convolutional neural networks
topic Convolutional neural network
Fault Identification
Multiscale Fusion
Acoustic emission
Pooling projection components, Variable operating conditions
url https://doi.org/10.1038/s41598-025-00573-7
work_keys_str_mv AT yuezhang rollingbearingfaultidentificationwithacousticemissionsignalbasedonvariablepoolingmultiscaleconvolutionalneuralnetworks
AT yangyu rollingbearingfaultidentificationwithacousticemissionsignalbasedonvariablepoolingmultiscaleconvolutionalneuralnetworks
AT zhengyang rollingbearingfaultidentificationwithacousticemissionsignalbasedonvariablepoolingmultiscaleconvolutionalneuralnetworks
AT qiangliu rollingbearingfaultidentificationwithacousticemissionsignalbasedonvariablepoolingmultiscaleconvolutionalneuralnetworks