Efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics

Abstract Convolutional Neural Networks have been widely applied in fault diagnosis tasks of mechanical systems due to their strong feature extraction and classification capabilities. However, they have limitations in handling global context information. Vision Transformers, by leveraging self-attent...

Full description

Saved in:
Bibliographic Details
Main Authors: Chaoquan Mo, Ke Huang, Houxin Ji, Wenhan Li, Kaibo Xu
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-88668-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850026112032178176
author Chaoquan Mo
Ke Huang
Houxin Ji
Wenhan Li
Kaibo Xu
author_facet Chaoquan Mo
Ke Huang
Houxin Ji
Wenhan Li
Kaibo Xu
author_sort Chaoquan Mo
collection DOAJ
description Abstract Convolutional Neural Networks have been widely applied in fault diagnosis tasks of mechanical systems due to their strong feature extraction and classification capabilities. However, they have limitations in handling global context information. Vision Transformers, by leveraging self-attention mechanisms to capture global dependencies, have shown excellent performance in many visual tasks, but often come with high computational costs. Therefore, this paper proposes a lightweight and efficient intelligent fault diagnosis method based on the fusion of Convolutional Network and Vision Transformer features (FCNVT). This method combines the local feature extraction capability of CNNs with the global dependency capturing ability of ViTs, while maintaining computational efficiency. Random overlapping sampling (ROS) techniques are used to preprocess signals, generating two-dimensional synchronized wavelet transform (SWT) images as inputs to the network. Experimental verification has shown that the proposed method achieves up to 100% classification accuracy, with the model having 7 million parameters and a computational cost of only 0.28 G, outperforming other state-of-the-art methods. Finally, a graphical user interface (GUI)-based mechanical equipment fault detection system was developed using this method, which holds positive implications for advancing the practical application of intelligent fault diagnosis in mechanical equipment.
format Article
id doaj-art-a4c504b0ab144e25b91f174d9e73e5b2
institution DOAJ
issn 2045-2322
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-a4c504b0ab144e25b91f174d9e73e5b22025-08-20T03:00:39ZengNature PortfolioScientific Reports2045-23222025-02-0115111810.1038/s41598-025-88668-zEfficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristicsChaoquan Mo0Ke Huang1Houxin Ji2Wenhan Li3Kaibo Xu4College of Mechanical & Electrical Engineering, Wenzhou UniversityCollege of Mechanical & Electrical Engineering, Wenzhou UniversityCollege of Mechanical & Electrical Engineering, Wenzhou UniversityCollege of Mechanical & Electrical Engineering, Wenzhou UniversityCollege of Mechanical & Electrical Engineering, Wenzhou UniversityAbstract Convolutional Neural Networks have been widely applied in fault diagnosis tasks of mechanical systems due to their strong feature extraction and classification capabilities. However, they have limitations in handling global context information. Vision Transformers, by leveraging self-attention mechanisms to capture global dependencies, have shown excellent performance in many visual tasks, but often come with high computational costs. Therefore, this paper proposes a lightweight and efficient intelligent fault diagnosis method based on the fusion of Convolutional Network and Vision Transformer features (FCNVT). This method combines the local feature extraction capability of CNNs with the global dependency capturing ability of ViTs, while maintaining computational efficiency. Random overlapping sampling (ROS) techniques are used to preprocess signals, generating two-dimensional synchronized wavelet transform (SWT) images as inputs to the network. Experimental verification has shown that the proposed method achieves up to 100% classification accuracy, with the model having 7 million parameters and a computational cost of only 0.28 G, outperforming other state-of-the-art methods. Finally, a graphical user interface (GUI)-based mechanical equipment fault detection system was developed using this method, which holds positive implications for advancing the practical application of intelligent fault diagnosis in mechanical equipment.https://doi.org/10.1038/s41598-025-88668-zMechanical equipment fault diagnosisCNN-transformerRandom overlapping samplingSynchrosqueezed wavelet transformGraphical user interface
spellingShingle Chaoquan Mo
Ke Huang
Houxin Ji
Wenhan Li
Kaibo Xu
Efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics
Scientific Reports
Mechanical equipment fault diagnosis
CNN-transformer
Random overlapping sampling
Synchrosqueezed wavelet transform
Graphical user interface
title Efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics
title_full Efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics
title_fullStr Efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics
title_full_unstemmed Efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics
title_short Efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics
title_sort efficient intelligent fault diagnosis method and graphical user interface development based on fusion of convolutional networks and vision transformers characteristics
topic Mechanical equipment fault diagnosis
CNN-transformer
Random overlapping sampling
Synchrosqueezed wavelet transform
Graphical user interface
url https://doi.org/10.1038/s41598-025-88668-z
work_keys_str_mv AT chaoquanmo efficientintelligentfaultdiagnosismethodandgraphicaluserinterfacedevelopmentbasedonfusionofconvolutionalnetworksandvisiontransformerscharacteristics
AT kehuang efficientintelligentfaultdiagnosismethodandgraphicaluserinterfacedevelopmentbasedonfusionofconvolutionalnetworksandvisiontransformerscharacteristics
AT houxinji efficientintelligentfaultdiagnosismethodandgraphicaluserinterfacedevelopmentbasedonfusionofconvolutionalnetworksandvisiontransformerscharacteristics
AT wenhanli efficientintelligentfaultdiagnosismethodandgraphicaluserinterfacedevelopmentbasedonfusionofconvolutionalnetworksandvisiontransformerscharacteristics
AT kaiboxu efficientintelligentfaultdiagnosismethodandgraphicaluserinterfacedevelopmentbasedonfusionofconvolutionalnetworksandvisiontransformerscharacteristics