Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking prediction
Pipelines are crucial for transporting energy sources, yet corrosion especially stress corrosion cracking (SCC) poses a complex and potentially catastrophic form of material degradation. Traditional techniques like finite element analysis (FEA) have been utilized for SCC prediction, but it suffers f...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-12-01
|
| Series: | Engineering Applications of Computational Fluid Mechanics |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/19942060.2024.2302906 |
| _version_ | 1797348255798919168 |
|---|---|
| author | Umair Sarwar Ainul Akmar Mokhtar Masdi Muhammad Rano Khan Wassan Afzal Ahmed Soomro Majid Ali Wassan Shuaib Kaka |
| author_facet | Umair Sarwar Ainul Akmar Mokhtar Masdi Muhammad Rano Khan Wassan Afzal Ahmed Soomro Majid Ali Wassan Shuaib Kaka |
| author_sort | Umair Sarwar |
| collection | DOAJ |
| description | Pipelines are crucial for transporting energy sources, yet corrosion especially stress corrosion cracking (SCC) poses a complex and potentially catastrophic form of material degradation. Traditional techniques like finite element analysis (FEA) have been utilized for SCC prediction, but it suffers from high computational cost and limited scalability. Deep learning (DL) with integration of FEA leverages large-scale data and learn complex nonlinear patterns for SCC prediction. Currently, literature on deep learning-enabled finite element analysis for pipelines SCC prediction is scarce, offering limited insights into this emerging approach and lack of comprehensive review. This paper reviews and investigates the current research directions and applications of DL-enabled FEA methodologies for simulation of SCC prediction. The importance of DL, technique type and network are also outlined in this review paper. This paper delves into integration of DL algorithms with FEA and their ability to grab complex interactions between mechanical stress, material properties, and environmental factors. Based on this comprehensive review, it was found that DL and FEA have proven to be strong prediction tools with high accuracy and lower training cost. DL-enabled FEA techniques are also being utilized to replace time-consuming methods and conventional codes. Furthermore, article discusses potential of this integrated approach for enhancing accuracy and efficiency of SCC prediction, leading to improved pipeline integrity management practices.Abbreviations: SCC: Stress corrosion cracking; FE: Finite element; FEA: Finite element analysis; FEM: Finite element method; ML: Machine learning; DL: Deep learning; EGIG: European gas pipeline incident data group; PHMSA: Pipeline and hazardous materials safety administration; DNNs: Deep neural networks; HpHSCC: High pH stress corrosion cracking; SMYS: Specified minimum yield strength; HE: Hydrogen embrittlement; NNpHSCC: Near-neutral pH stress corrosion cracking; ASME: American society of mechanical engineers; DNV: Det norske veritas; AI-FEM: Advanced iterative finite element method; CGR: Crack growth rate; ME: Mechanoelectrochemical; XGB: XGBoost; CAT: Catboost; CP: Cathodic Protection; AI: Artificial intelligence; ANNs: Artificial neural networks; CNNs: Convolutional neural networks; RNNs: Recurrent neural networks; ReLU: Rectified linear unit |
| first_indexed | 2024-03-08T12:03:09Z |
| format | Article |
| id | doaj.art-ede409e31d00446ab146116c030d87b1 |
| institution | Directory Open Access Journal |
| issn | 1994-2060 1997-003X |
| language | English |
| last_indexed | 2024-03-08T12:03:09Z |
| publishDate | 2024-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Engineering Applications of Computational Fluid Mechanics |
| spelling | doaj.art-ede409e31d00446ab146116c030d87b12024-01-23T17:17:46ZengTaylor & Francis GroupEngineering Applications of Computational Fluid Mechanics1994-20601997-003X2024-12-0118110.1080/19942060.2024.2302906Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking predictionUmair Sarwar0Ainul Akmar Mokhtar1Masdi Muhammad2Rano Khan Wassan3Afzal Ahmed Soomro4Majid Ali Wassan5Shuaib Kaka6Department of Mechanical Engineering, Universiti Teknologi Petronas, Seri Iskandar, MalaysiaDepartment of Mechanical Engineering, Universiti Teknologi Petronas, Seri Iskandar, MalaysiaDepartment of Mechanical Engineering, Universiti Teknologi Petronas, Seri Iskandar, MalaysiaDepartment of Industrial Engineering and Management, Dawood University of Engineering and Technology, Karachi, PakistanDepartment of Mechanical Engineering, Universiti Teknologi Petronas, Seri Iskandar, MalaysiaDepartment of Mechanical Engineering, Massachusetts Lowell, Lowell, USADepartment of Industrial Engineering and Management, Dawood University of Engineering and Technology, Karachi, PakistanPipelines are crucial for transporting energy sources, yet corrosion especially stress corrosion cracking (SCC) poses a complex and potentially catastrophic form of material degradation. Traditional techniques like finite element analysis (FEA) have been utilized for SCC prediction, but it suffers from high computational cost and limited scalability. Deep learning (DL) with integration of FEA leverages large-scale data and learn complex nonlinear patterns for SCC prediction. Currently, literature on deep learning-enabled finite element analysis for pipelines SCC prediction is scarce, offering limited insights into this emerging approach and lack of comprehensive review. This paper reviews and investigates the current research directions and applications of DL-enabled FEA methodologies for simulation of SCC prediction. The importance of DL, technique type and network are also outlined in this review paper. This paper delves into integration of DL algorithms with FEA and their ability to grab complex interactions between mechanical stress, material properties, and environmental factors. Based on this comprehensive review, it was found that DL and FEA have proven to be strong prediction tools with high accuracy and lower training cost. DL-enabled FEA techniques are also being utilized to replace time-consuming methods and conventional codes. Furthermore, article discusses potential of this integrated approach for enhancing accuracy and efficiency of SCC prediction, leading to improved pipeline integrity management practices.Abbreviations: SCC: Stress corrosion cracking; FE: Finite element; FEA: Finite element analysis; FEM: Finite element method; ML: Machine learning; DL: Deep learning; EGIG: European gas pipeline incident data group; PHMSA: Pipeline and hazardous materials safety administration; DNNs: Deep neural networks; HpHSCC: High pH stress corrosion cracking; SMYS: Specified minimum yield strength; HE: Hydrogen embrittlement; NNpHSCC: Near-neutral pH stress corrosion cracking; ASME: American society of mechanical engineers; DNV: Det norske veritas; AI-FEM: Advanced iterative finite element method; CGR: Crack growth rate; ME: Mechanoelectrochemical; XGB: XGBoost; CAT: Catboost; CP: Cathodic Protection; AI: Artificial intelligence; ANNs: Artificial neural networks; CNNs: Convolutional neural networks; RNNs: Recurrent neural networks; ReLU: Rectified linear unithttps://www.tandfonline.com/doi/10.1080/19942060.2024.2302906Oil and gassteel pipelinefinite element analysisstress corrosion cracking predictiondeep learning |
| spellingShingle | Umair Sarwar Ainul Akmar Mokhtar Masdi Muhammad Rano Khan Wassan Afzal Ahmed Soomro Majid Ali Wassan Shuaib Kaka Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking prediction Engineering Applications of Computational Fluid Mechanics Oil and gas steel pipeline finite element analysis stress corrosion cracking prediction deep learning |
| title | Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking prediction |
| title_full | Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking prediction |
| title_fullStr | Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking prediction |
| title_full_unstemmed | Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking prediction |
| title_short | Enhancing pipeline integrity: a comprehensive review of deep learning-enabled finite element analysis for stress corrosion cracking prediction |
| title_sort | enhancing pipeline integrity a comprehensive review of deep learning enabled finite element analysis for stress corrosion cracking prediction |
| topic | Oil and gas steel pipeline finite element analysis stress corrosion cracking prediction deep learning |
| url | https://www.tandfonline.com/doi/10.1080/19942060.2024.2302906 |
| work_keys_str_mv | AT umairsarwar enhancingpipelineintegrityacomprehensivereviewofdeeplearningenabledfiniteelementanalysisforstresscorrosioncrackingprediction AT ainulakmarmokhtar enhancingpipelineintegrityacomprehensivereviewofdeeplearningenabledfiniteelementanalysisforstresscorrosioncrackingprediction AT masdimuhammad enhancingpipelineintegrityacomprehensivereviewofdeeplearningenabledfiniteelementanalysisforstresscorrosioncrackingprediction AT ranokhanwassan enhancingpipelineintegrityacomprehensivereviewofdeeplearningenabledfiniteelementanalysisforstresscorrosioncrackingprediction AT afzalahmedsoomro enhancingpipelineintegrityacomprehensivereviewofdeeplearningenabledfiniteelementanalysisforstresscorrosioncrackingprediction AT majidaliwassan enhancingpipelineintegrityacomprehensivereviewofdeeplearningenabledfiniteelementanalysisforstresscorrosioncrackingprediction AT shuaibkaka enhancingpipelineintegrityacomprehensivereviewofdeeplearningenabledfiniteelementanalysisforstresscorrosioncrackingprediction |