A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main Engine
Ensuring operational reliability in machinery requires accurate fault detection. While time-domain vibration pulsation signals are intuitive for pattern recognition and feature extraction, downsampling can reduce analytical complexity, but may result in low-precision data, affecting fault detection...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Journal of Marine Science and Engineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-1312/11/8/1577 |
| _version_ | 1827729497640665088 |
|---|---|
| author | Hong Je-Gal Seung-Jin Lee Jeong-Hyun Yoon Hyun-Suk Lee Jung-Hee Yang Sewon Kim |
| author_facet | Hong Je-Gal Seung-Jin Lee Jeong-Hyun Yoon Hyun-Suk Lee Jung-Hee Yang Sewon Kim |
| author_sort | Hong Je-Gal |
| collection | DOAJ |
| description | Ensuring operational reliability in machinery requires accurate fault detection. While time-domain vibration pulsation signals are intuitive for pattern recognition and feature extraction, downsampling can reduce analytical complexity, but may result in low-precision data, affecting fault detection performance. To address this, we propose time–frequency feature fusion, combining information from both the time and frequency domains for fault detection. Our approach transforms vibrational pulse data into instantaneous revolutions per minute (RPM) and employs statistical analysis for the time-domain features. For the frequency-domain features, we use the combined method of empirical mode decomposition and independent component analysis (EMD-ICA), along with the Wigner bispectrum method to capture the nonlinear characteristics and phase conjugation. Using a deep neural network (DNN), we classify the anomaly states, demonstrating the effectiveness and versatility of our approach in detecting anomalies and improving diagnostic precision. Compared to using time or frequency features alone, our time–frequency feature fusion model achieves higher accuracy, with 100% accuracy at lower downsampling rates and 96.3% accuracy at a downsampling rate of 100×. |
| first_indexed | 2024-03-10T23:49:38Z |
| format | Article |
| id | doaj.art-8ebfd4e59c694b71b6ee4bd85be6f4a8 |
| institution | Directory Open Access Journal |
| issn | 2077-1312 |
| language | English |
| last_indexed | 2024-03-10T23:49:38Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Marine Science and Engineering |
| spelling | doaj.art-8ebfd4e59c694b71b6ee4bd85be6f4a82023-11-19T01:46:05ZengMDPI AGJournal of Marine Science and Engineering2077-13122023-08-01118157710.3390/jmse11081577A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main EngineHong Je-Gal0Seung-Jin Lee1Jeong-Hyun Yoon2Hyun-Suk Lee3Jung-Hee Yang4Sewon Kim5Department of Intelligent Mechatronics Engineering, Sejong University, Seoul 05006, Republic of KoreaDepartment of Intelligent Mechatronics Engineering, Sejong University, Seoul 05006, Republic of KoreaDepartment of Intelligent Mechatronics Engineering, Sejong University, Seoul 05006, Republic of KoreaDepartment of Intelligent Mechatronics Engineering, Sejong University, Seoul 05006, Republic of KoreaSmart Ship Solution Department, Hanwha Ocean Co., Ltd., Seoul 04527, Republic of KoreaDepartment of Intelligent Mechatronics Engineering, Sejong University, Seoul 05006, Republic of KoreaEnsuring operational reliability in machinery requires accurate fault detection. While time-domain vibration pulsation signals are intuitive for pattern recognition and feature extraction, downsampling can reduce analytical complexity, but may result in low-precision data, affecting fault detection performance. To address this, we propose time–frequency feature fusion, combining information from both the time and frequency domains for fault detection. Our approach transforms vibrational pulse data into instantaneous revolutions per minute (RPM) and employs statistical analysis for the time-domain features. For the frequency-domain features, we use the combined method of empirical mode decomposition and independent component analysis (EMD-ICA), along with the Wigner bispectrum method to capture the nonlinear characteristics and phase conjugation. Using a deep neural network (DNN), we classify the anomaly states, demonstrating the effectiveness and versatility of our approach in detecting anomalies and improving diagnostic precision. Compared to using time or frequency features alone, our time–frequency feature fusion model achieves higher accuracy, with 100% accuracy at lower downsampling rates and 96.3% accuracy at a downsampling rate of 100×.https://www.mdpi.com/2077-1312/11/8/1577fault detectionmarine main enginedeep neural networkpredictive maintenancetime–frequency feature fusion |
| spellingShingle | Hong Je-Gal Seung-Jin Lee Jeong-Hyun Yoon Hyun-Suk Lee Jung-Hee Yang Sewon Kim A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main Engine Journal of Marine Science and Engineering fault detection marine main engine deep neural network predictive maintenance time–frequency feature fusion |
| title | A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main Engine |
| title_full | A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main Engine |
| title_fullStr | A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main Engine |
| title_full_unstemmed | A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main Engine |
| title_short | A Novel Time–Frequency Feature Fusion Approach for Robust Fault Detection in a Marine Main Engine |
| title_sort | novel time frequency feature fusion approach for robust fault detection in a marine main engine |
| topic | fault detection marine main engine deep neural network predictive maintenance time–frequency feature fusion |
| url | https://www.mdpi.com/2077-1312/11/8/1577 |
| work_keys_str_mv | AT hongjegal anoveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT seungjinlee anoveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT jeonghyunyoon anoveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT hyunsuklee anoveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT jungheeyang anoveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT sewonkim anoveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT hongjegal noveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT seungjinlee noveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT jeonghyunyoon noveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT hyunsuklee noveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT jungheeyang noveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine AT sewonkim noveltimefrequencyfeaturefusionapproachforrobustfaultdetectioninamarinemainengine |