Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual Network
In ship propulsion, accurately diagnosing faults in permanent magnet synchronous motor is essential but challenging due to limitations in the intuitive characterization and feature extraction of fault signals. This study presents an innovative approach to motor fault detection by integrating phase-c...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Journal of Marine Science and Engineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-1312/13/1/70 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588199916994560 |
|---|---|
| author | Wei Yuan Julong Chen Xingji Yu |
| author_facet | Wei Yuan Julong Chen Xingji Yu |
| author_sort | Wei Yuan |
| collection | DOAJ |
| description | In ship propulsion, accurately diagnosing faults in permanent magnet synchronous motor is essential but challenging due to limitations in the intuitive characterization and feature extraction of fault signals. This study presents an innovative approach to motor fault detection by integrating phase-contrastive current dot patterns with an enhanced residual network, enhancing the diagnostic effect. Initially, the research involves creating a dataset that simulates stator currents. It is achieved through mathematical modeling of two common faults in permanent magnet synchronous motors: inter-turn short circuits and demagnetization. Subsequently, the parameters of the phase-contrastive current dot pattern are optimized using the Hunter-Prey Optimization technique to convert the three-phase stator currents of the motor into grayscale images. Lastly, a residual network, which includes a Squeeze-and-Excitation module, is engineered to boost the identification of crucial fault characteristics. The experimental results show that the proposed method achieves a high accuracy rate of 98.5% in the fault diagnosis task of motors, which can accurately identify the fault information and is significant in enhancing the reliability and safety of ship propulsion systems. |
| format | Article |
| id | doaj-art-5c5a664c27144b77bd039146609a9828 |
| institution | Kabale University |
| issn | 2077-1312 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Marine Science and Engineering |
| spelling | doaj-art-5c5a664c27144b77bd039146609a98282025-01-24T13:36:45ZengMDPI AGJournal of Marine Science and Engineering2077-13122025-01-011317010.3390/jmse13010070Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual NetworkWei Yuan0Julong Chen1Xingji Yu2College of Automation, Jiangsu University of Science and Technology, Zhenjiang 212100, ChinaCollege of Automation, Jiangsu University of Science and Technology, Zhenjiang 212100, ChinaDepartment of Architecture and Technology, Norwegian University of Science and Technology, Alfred Getz’vei 1, 7034 Trondheim, NorwayIn ship propulsion, accurately diagnosing faults in permanent magnet synchronous motor is essential but challenging due to limitations in the intuitive characterization and feature extraction of fault signals. This study presents an innovative approach to motor fault detection by integrating phase-contrastive current dot patterns with an enhanced residual network, enhancing the diagnostic effect. Initially, the research involves creating a dataset that simulates stator currents. It is achieved through mathematical modeling of two common faults in permanent magnet synchronous motors: inter-turn short circuits and demagnetization. Subsequently, the parameters of the phase-contrastive current dot pattern are optimized using the Hunter-Prey Optimization technique to convert the three-phase stator currents of the motor into grayscale images. Lastly, a residual network, which includes a Squeeze-and-Excitation module, is engineered to boost the identification of crucial fault characteristics. The experimental results show that the proposed method achieves a high accuracy rate of 98.5% in the fault diagnosis task of motors, which can accurately identify the fault information and is significant in enhancing the reliability and safety of ship propulsion systems.https://www.mdpi.com/2077-1312/13/1/70ship propulsion systempermanent magnet synchronous motorsfault diagnosisresidual networkphase-contrastive current dot pattern |
| spellingShingle | Wei Yuan Julong Chen Xingji Yu Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual Network Journal of Marine Science and Engineering ship propulsion system permanent magnet synchronous motors fault diagnosis residual network phase-contrastive current dot pattern |
| title | Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual Network |
| title_full | Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual Network |
| title_fullStr | Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual Network |
| title_full_unstemmed | Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual Network |
| title_short | Research on Fault Diagnosis of Ship Propulsion System Based on Improved Residual Network |
| title_sort | research on fault diagnosis of ship propulsion system based on improved residual network |
| topic | ship propulsion system permanent magnet synchronous motors fault diagnosis residual network phase-contrastive current dot pattern |
| url | https://www.mdpi.com/2077-1312/13/1/70 |
| work_keys_str_mv | AT weiyuan researchonfaultdiagnosisofshippropulsionsystembasedonimprovedresidualnetwork AT julongchen researchonfaultdiagnosisofshippropulsionsystembasedonimprovedresidualnetwork AT xingjiyu researchonfaultdiagnosisofshippropulsionsystembasedonimprovedresidualnetwork |