Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibration
The health monitoring and fault diagnosis of heavy-duty engines are increasingly important for energy storage ecosystem. During operation, vibration characters corresponding to the specific fault need to be extracted from the overall system vibration. Faulty characteristics emanating from one single...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-05-01
|
| Series: | Energy and AI |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666546824000223 |
| _version_ | 1797263596722323456 |
|---|---|
| author | Xiaolong Zhu Junhong Zhang Xinwei Wang Hui Wang Yedong Song Guobin Pei Xin Gou Linlong Deng Jiewei Lin |
| author_facet | Xiaolong Zhu Junhong Zhang Xinwei Wang Hui Wang Yedong Song Guobin Pei Xin Gou Linlong Deng Jiewei Lin |
| author_sort | Xiaolong Zhu |
| collection | DOAJ |
| description | The health monitoring and fault diagnosis of heavy-duty engines are increasingly important for energy storage ecosystem. During operation, vibration characters corresponding to the specific fault need to be extracted from the overall system vibration. Faulty characteristics emanating from one single cylinder are also mixed with those from other cylinders. Besides, the change of working condition brings strong nonlinearities in surface vibration. To solve these problems, an improved deep residual shrinkage network (IDRSN) is developed for detecting diverse engine faults at various degrees using single channel surface vibration signal. Within IDRSN, a wide convolution kernel is utilized in first convolution layer to capture the long-term fault-related impacts and eliminate the short-time random impact. The residual network module is adopted to enhance the focus the relevant components of vibration signals. Mini-batch training strategy is used to improve the model stability. Meanwhile, Gradient-weighted class activation map is adopted to assess the consistency between the learned knowledge and the fault-related information. The IDRSN is implemented to diagnosing a diesel engine under various faults, faulty degrees and operating speeds. Comparisons with existing models are analyzed in terms of hyper-parameters, training samples, noise resistance, and visualization. Results demonstrate the proposed IDRSN's superior performance on fault diagnosis accuracy, stability, anti-noise performance, and anti-interference performance. An average accuracy rate of 98.38 % was achieved by the proposed IDRSN, in comparison to 96.64 % and 93.56 % achieved by the DRSN and the wide-kernel deep convolutional neural network respectively. These results highlight the proposed IDRSN's superiority in diagnosing multiple faults under various working conditions, offering a low-cost, highly effective, and applicable approach for complex fault diagnosis tasks. |
| first_indexed | 2024-04-25T00:15:31Z |
| format | Article |
| id | doaj.art-20a2c88a8e814e9cb8a8e0c76160e6bb |
| institution | Directory Open Access Journal |
| issn | 2666-5468 |
| language | English |
| last_indexed | 2024-04-25T00:15:31Z |
| publishDate | 2024-05-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Energy and AI |
| spelling | doaj.art-20a2c88a8e814e9cb8a8e0c76160e6bb2024-03-13T04:46:26ZengElsevierEnergy and AI2666-54682024-05-0116100356Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibrationXiaolong Zhu0Junhong Zhang1Xinwei Wang2Hui Wang3Yedong Song4Guobin Pei5Xin Gou6Linlong Deng7Jiewei Lin8State Key Laboratory of Engines, Tianjin University, Tianjin 300350, ChinaState Key Laboratory of Engines, Tianjin University, Tianjin 300350, China; Tianjin Renai College, Tianjin 301636, ChinaState Key Laboratory of Engines, Tianjin University, Tianjin 300350, China; Weichai Power Co., Ltd., Weifang 261061, ChinaState Key Laboratory of Engines, Tianjin University, Tianjin 300350, China; Weichai Power Co., Ltd., Weifang 261061, ChinaWeichai Power Co., Ltd., Weifang 261061, ChinaState Key Laboratory of Engines, Tianjin University, Tianjin 300350, ChinaState Key Laboratory of Engines, Tianjin University, Tianjin 300350, ChinaState Key Laboratory of Engines, Tianjin University, Tianjin 300350, ChinaState Key Laboratory of Engines, Tianjin University, Tianjin 300350, China; Corresponding author.The health monitoring and fault diagnosis of heavy-duty engines are increasingly important for energy storage ecosystem. During operation, vibration characters corresponding to the specific fault need to be extracted from the overall system vibration. Faulty characteristics emanating from one single cylinder are also mixed with those from other cylinders. Besides, the change of working condition brings strong nonlinearities in surface vibration. To solve these problems, an improved deep residual shrinkage network (IDRSN) is developed for detecting diverse engine faults at various degrees using single channel surface vibration signal. Within IDRSN, a wide convolution kernel is utilized in first convolution layer to capture the long-term fault-related impacts and eliminate the short-time random impact. The residual network module is adopted to enhance the focus the relevant components of vibration signals. Mini-batch training strategy is used to improve the model stability. Meanwhile, Gradient-weighted class activation map is adopted to assess the consistency between the learned knowledge and the fault-related information. The IDRSN is implemented to diagnosing a diesel engine under various faults, faulty degrees and operating speeds. Comparisons with existing models are analyzed in terms of hyper-parameters, training samples, noise resistance, and visualization. Results demonstrate the proposed IDRSN's superior performance on fault diagnosis accuracy, stability, anti-noise performance, and anti-interference performance. An average accuracy rate of 98.38 % was achieved by the proposed IDRSN, in comparison to 96.64 % and 93.56 % achieved by the DRSN and the wide-kernel deep convolutional neural network respectively. These results highlight the proposed IDRSN's superiority in diagnosing multiple faults under various working conditions, offering a low-cost, highly effective, and applicable approach for complex fault diagnosis tasks.http://www.sciencedirect.com/science/article/pii/S2666546824000223Improved deep residual shrinkage networkFault diagnosisEngineVibration signalMultiple working conditionsDeep learning |
| spellingShingle | Xiaolong Zhu Junhong Zhang Xinwei Wang Hui Wang Yedong Song Guobin Pei Xin Gou Linlong Deng Jiewei Lin Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibration Energy and AI Improved deep residual shrinkage network Fault diagnosis Engine Vibration signal Multiple working conditions Deep learning |
| title | Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibration |
| title_full | Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibration |
| title_fullStr | Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibration |
| title_full_unstemmed | Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibration |
| title_short | Improved deep residual shrinkage network for a multi-cylinder heavy-duty engine fault detection with single channel surface vibration |
| title_sort | improved deep residual shrinkage network for a multi cylinder heavy duty engine fault detection with single channel surface vibration |
| topic | Improved deep residual shrinkage network Fault diagnosis Engine Vibration signal Multiple working conditions Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S2666546824000223 |
| work_keys_str_mv | AT xiaolongzhu improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT junhongzhang improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT xinweiwang improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT huiwang improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT yedongsong improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT guobinpei improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT xingou improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT linlongdeng improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration AT jieweilin improveddeepresidualshrinkagenetworkforamulticylinderheavydutyenginefaultdetectionwithsinglechannelsurfacevibration |