A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault Diagnosis
To address the limited generalization capability of data-driven fault diagnosis models caused by scarce gearbox fault samples in engineering practice, this paper proposes a hierarchical attention-guided data–knowledge dual-driven fusion network for intelligent fault diagnosis under few-shot conditio...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Machines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1702/13/6/486 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839653489109630976 |
|---|---|
| author | Xin Feng Tianci Zhang |
| author_facet | Xin Feng Tianci Zhang |
| author_sort | Xin Feng |
| collection | DOAJ |
| description | To address the limited generalization capability of data-driven fault diagnosis models caused by scarce gearbox fault samples in engineering practice, this paper proposes a hierarchical attention-guided data–knowledge dual-driven fusion network for intelligent fault diagnosis under few-shot conditions. Distinct from traditional single data-driven paradigms, this method breaks through the constraints of limited samples through the synergy of prior knowledge and monitoring data. First, domain knowledge of gearbox fault diagnosis is utilized to construct prior features of monitoring data. Second, a deep convolutional neural network is designed to hierarchically capture abstract features from monitoring data. Subsequently, a hierarchical attention module is proposed to realize adaptive fusion of prior features and abstract features through hierarchical feature weight allocation, generating highly discriminative fused features for accurate gearbox fault identification. Experimental results on gearbox fault data demonstrate that the proposed method achieves 0.9880 recognition accuracy with less than 10% of the training samples, significantly outperforming purely data-driven models such as MGAN and CNET, thus verifying its superior generalization ability to train despite data scarcity. This approach establishes a novel data–knowledge dual-driven fusion paradigm for intelligent fault diagnosis of mechanical equipment under few-shot conditions. |
| format | Article |
| id | doaj-art-8d78239aa29047d9b602edb33173f16d |
| institution | Matheson Library |
| issn | 2075-1702 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machines |
| spelling | doaj-art-8d78239aa29047d9b602edb33173f16d2025-06-25T14:07:12ZengMDPI AGMachines2075-17022025-06-0113648610.3390/machines13060486A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault DiagnosisXin Feng0Tianci Zhang1AECC ZhongChuan Transmission Machinery Co., Ltd., Changsha 410200, ChinaState Key Laboratory of Precision Manufacturing for Extreme Service Performance, Central South University, Changsha 410083, ChinaTo address the limited generalization capability of data-driven fault diagnosis models caused by scarce gearbox fault samples in engineering practice, this paper proposes a hierarchical attention-guided data–knowledge dual-driven fusion network for intelligent fault diagnosis under few-shot conditions. Distinct from traditional single data-driven paradigms, this method breaks through the constraints of limited samples through the synergy of prior knowledge and monitoring data. First, domain knowledge of gearbox fault diagnosis is utilized to construct prior features of monitoring data. Second, a deep convolutional neural network is designed to hierarchically capture abstract features from monitoring data. Subsequently, a hierarchical attention module is proposed to realize adaptive fusion of prior features and abstract features through hierarchical feature weight allocation, generating highly discriminative fused features for accurate gearbox fault identification. Experimental results on gearbox fault data demonstrate that the proposed method achieves 0.9880 recognition accuracy with less than 10% of the training samples, significantly outperforming purely data-driven models such as MGAN and CNET, thus verifying its superior generalization ability to train despite data scarcity. This approach establishes a novel data–knowledge dual-driven fusion paradigm for intelligent fault diagnosis of mechanical equipment under few-shot conditions.https://www.mdpi.com/2075-1702/13/6/486gearboxfault diagnosisprior knowledgefew-shot conditionattention mechanism |
| spellingShingle | Xin Feng Tianci Zhang A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault Diagnosis Machines gearbox fault diagnosis prior knowledge few-shot condition attention mechanism |
| title | A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault Diagnosis |
| title_full | A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault Diagnosis |
| title_fullStr | A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault Diagnosis |
| title_full_unstemmed | A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault Diagnosis |
| title_short | A Hierarchical Attention-Guided Data–Knowledge Fusion Network for Few-Shot Gearboxes’ Fault Diagnosis |
| title_sort | hierarchical attention guided data knowledge fusion network for few shot gearboxes fault diagnosis |
| topic | gearbox fault diagnosis prior knowledge few-shot condition attention mechanism |
| url | https://www.mdpi.com/2075-1702/13/6/486 |
| work_keys_str_mv | AT xinfeng ahierarchicalattentionguideddataknowledgefusionnetworkforfewshotgearboxesfaultdiagnosis AT tiancizhang ahierarchicalattentionguideddataknowledgefusionnetworkforfewshotgearboxesfaultdiagnosis AT xinfeng hierarchicalattentionguideddataknowledgefusionnetworkforfewshotgearboxesfaultdiagnosis AT tiancizhang hierarchicalattentionguideddataknowledgefusionnetworkforfewshotgearboxesfaultdiagnosis |