Boru hatlarında çizge evrişimsel ağlar yöntemi (GCN) ile arıza tespiti

Yıl 2025, , 673 - 684, 16.08.2024

Ersin Şahin , Hüseyin Yüce

https://doi.org/10.17341/gazimmfd.1306916

Öz

Boru hatları petrol ve doğalgaz gibi enerji kaynaklarının taşınmasından, su kaynaklarının iletilmesi ve dağıtılmasına kadar çok geniş bir alanda kullanım alanına sahiptir. Boru hatlarından sızan petrol ve gaz akışkanları çevreye ciddi zararlar vermektedir. Boru hatlarındaki arızaların doğru bir şekilde tespit edilmesi, ekonomik kayıpların etkilerinden kaçınmak ve çevreyi korumak için önemlidir. Bu çalışmada su akışkanına sahip boru hatları çizgeler (graf) ile temsil edilmiştir. Boru hatlarında sızıntı ve tıkanma durumlarının tespiti için çizge temelli makine öğrenmesi algoritması GCN kullanılmıştır. Deneysel yöntemler kullanılarak GCN algoritması için gerekli olan veriler (basınç verileri), beş farklı senaryo oluşturularak toplanmış ve iki adet veri seti oluşturulmuştur. GCN algoritmasından elde edilen arıza tespit performans değeri diğer çizge makine öğrenmesi algoritmaları; RGCN, HinSAGE ve GraphSAGE ’nin performansları ile karşılaştırılmıştır. Bu çalışmada GCN modelinin performansı diğer algoritmalara göre daha yüksek bulunmuştur. Literatür incelendiğinde makine öğrenmesi algoritmaları kullanılarak boru hatlarında arıza teşhisi için doğruluk oranları %78,51 ile %99 değerleri arasında tespit edilmiştir. Bu çalışmada, GCN, GraphSAGE, HinSAGE ve RGCN algoritmalarının sırasıyla %0,91, %0,90, %0,87, %0,89 doğruluk oranları ile boru hatlarında arıza tespiti yaptıkları bulgusuna varılmıştır. Çizge temelli algoritmaların performanslarının kıyaslanması için klasik makine öğrenmesi SVM modeli kullanılmıştır. Algoritmaların performansları literatür ile karşılaştırıldığında sonuçların literatür ortalamasının üstünde olduğu görülmektedir.

Anahtar Kelimeler

Çizge evrişimli ağlar, Arıza tespiti, Boru hatları

Destekleyen Kurum

Marmara Üniversitesi

Proje Numarası

FDK-2023-10459

Teşekkür

Bu çalışma, Marmara Üniversitesi tarafından FDK-2023-10459 numaralı bilimsel araştırma projesi ile desteklenmiştir. Çalışmanın deneysel aşamalarında Marmara Üniversitesi, Teknoloji Fakültesi Mekatronik Mühendisliği Arıza Teşhis Laboratuvarında bulunan deney seti kullanılmıştır. Bu imkanları sağlayan Marmara Üniversitesine teşekkürü bir borç bilirim.

Kaynakça

• 1. Yang, L., et al., Detection of pipeline blockage using lab experiment and computational fluid dynamic simulation, Journal of Petroleum Science and Engineering 183, 106421, 2019.
• 2. Wang, W., et al., Experimental study on water pipeline leak using In Pipe acoustic signal analysis and artificial neural network prediction, Measurement 186, 110094, 2021.
• 3. Anfinsen, H., & Aamo, O. M., Leak detection, size estimation and localization in branched pipe flows, Automatica 140, 110213, 2022.
• 4. Shivananju, B. N., et al., Real time monitoring of petroleum leakage detection using etched fiber Bragg grating, International Conference on Optics in Precision Engineering and Nanotechnology (icOPEN2013) 8769, SPIE, 2013.
• 5. Meng, S., Li, Y., & Wang, X., A Simulation Study on the Dynamic Stability of a Fluid Conveying Pipe With a Constant Velocity Leakage, Journal of Pressure Vessel Technology 139 (4), 2017.
• 6. Zeng, W., et al., Linear phase detector for detecting multiple leaks in water pipes, Applied Acoustics 202, 109152, 2023.
• 7. Datta, S., & Sarkar, S., A review on different pipeline fault detection methods, Journal of Loss Prevention in the Process Industries 41, 97-106, 2016.
• 8. Liang, L., et al., Pipeline leakage test based on FBG pressure sensor, IOP Conference Series: Earth and Environmental Science 170 (2), 2018.
• 9. Sheltami, T. R., Bala, A., & Shakshuki, E. M., Wireless sensor networks for leak detection in pipelines: a survey, Journal of Ambient Intelligence and Humanized Computing 7 (3), 347-356, 2016.
• 10. Yu, D., et al., Projective Noise Reduction Algorithm for Negative Pressure Wave Signal Processing, International Pipeline Conference 48579, 2008.
• 11. Bai, L., Yue, Q., & Li, H., Sub-sea Pipelines Leak Detection and Location Based on Fluid Transient and FDI, The Fourteenth International Offshore and Polar Engineering Conference One Petro, 2004.
• 12. Ling, K., et al., A new method for leak detection in gas pipelines, Oil and Gas Facilities 4.02, 97-106, 2015. 13. Reddy, R. S., et al., Pressure and flow variation in gas distribution pipeline for leak detection, 2016 IEEE International Conference on Industrial Technology (ICIT), IEEE, 2016.
• 14. Korlapati, N. V. S., et al., Review and analysis of pipeline leak detection methods, Journal of Pipeline Science and Engineering 100074, 2022.
• 15. Goni, M. O. F., et al., Fast and Accurate Fault Detection and Classification in Transmission Lines using Extreme Learning Machine, e-Prime Advances in Electrical Engineering, Electronics and Energy 100107, 2023.
• 16. Mandal, S. K., Chan, F. T., & Tiwari, M. K., Leak detection of pipeline: An integrated approach of rough set theory and artificial bee colony trained SVM, Expert Systems with Applications 39 (3), 3071-3080, 2012.
• 17. Ayati, A. H., Haghighi, A., & Ghafouri, H. R., Machine Learning–Assisted Model for Leak Detection in Water Distribution Networks Using Hydraulic Transient Flows, Journal of Water Resources Planning and Management 148 (2), 04021104, 2022.
• 18. Eastvedt, D., Naterer, G., & Duan, X., Detection of faults in subsea pipelines by flow monitoring with regression supervised machine learning, Process Safety and Environmental Protection 161, 409-420, 2022.
• 19. Ahn, B., Kim, J., & Choi, B., Artificial intelligence-based machine learning considering flow and temperature of the pipeline for leak early detection using acoustic emission, Engineering Fracture Mechanics 210, 381-392, 2019.
• 20. Garðarsson, G. Ö., Boem, F., & Toni, L., Graph-based learning for leak detection and localization in water distribution networks, IFAC Papers Online 55 (6), 661-666, 2022.
• 21. Candelieri, A., Conti, D., & Archetti, F., A graph based analysis of leak localization in urban water networks, Procedia Engineering 70, 228-237, 2014.
• 22. Romero, L., et al., First results in leak localization in water distribution networks using graph-based clustering and deep learning, IFAC Papers Online 53 (2), 16691-16696, 2020.
• 23. Kang, J., et al., Novel leakage detection by ensemble CNN-SVM and graph-based localization in water distribution systems, IEEE Transactions on Industrial Electronics 65 (5), 4279-4289, 2017.
• 24. Zhang, X., et al., Towards deep probabilistic graph neural network for natural gas leak detection and localization without labeled anomaly data, Expert Systems with Applications 231, 120542, 2023.
• 25. Miao, X., Zhao, H., & Xiang, Z., Leakage detection in natural gas pipeline based on unsupervised learning and stress perception, Process Safety and Environmental Protection 170, 76-88, 2023.
• 26. Rashid, S., Akram, U., Khan, S. A., WML: Wireless sensor network based machine learning for leakage detection and size estimation, Procedia Computer Science 63, 171-176, 2015.
• 27. Xu, T., et al., Pipeline leak detection based on variational mode decomposition and support vector machine using an interior spherical detector, Process Safety and Environmental Protection 153, 167-177, 2021.
• 28. De Kerf, T., et al., Oil spill detection using machine learning and infrared images, Remote Sensing 12 (24), 4090, 2020.
• 29. Zheng, J., et al., Deeppipe: A deep-learning method for anomaly detection of multi-product pipelines, Energy 259, 125025, 2022.
• 30. Baronti, L., et al., Neural network identification of water pipe blockage from smart embedded passive acoustic measurements, The Canadian Journal of Chemical Engineering 100 (3), 521-539, 2022.
• 31. Bohorquez, J., et al., Leak detection and topology identification in pipelines using fluid transients and artificial neural Networks, Journal of Water Resources Planning and Management 146 (6), 04020040, 2020.
• 32. Wang, B., et al., Prediction model of natural gas pipeline crack evolution based on optimized DCNN-LSTM, Mechanical Systems and Signal Processing 181, 109557, 2022.
• 33. Asghari, V., et al., Machine learning modeling for spectral transient based leak detection, Automation in Construction 146, 104686, 2023.
• 34. Xiao, R., & Li, J., Evaluation of acoustic techniques for leak detection in a complex low pressure gas pipeline network, Engineering Failure Analysis 143, 106897, 2023.
• 35. Banjara, N. K., Sasmal, S., & Voggu, S., Machine learning supported acoustic emission technique for leakage detection in pipelines, International Journal of Pressure Vessels and Piping 188, 104243, 2020.
• 36. Yu, X., & Tian, X., A fault detection algorithm for pipeline insulation layer based on immune neural network, International Journal of Pressure Vessels and Piping 196, 104611, 2022.
• 37. Li, Q., et al., A novel oil pipeline leakage detection method based on the sparrow search algorithm and CNN, Measurement 204, 112122, 2022.
• 38. Liao, Z., et al., A temporal and spatial prediction method for urban pipeline network based on deep learning, Physical A: Statistical Mechanics and its Applications 608, 128299, 2022.
• 39. Edwards, M., & Xie, X., Graph based convolutional neural network, arXiv preprint arXiv: 1609.08965, 2016.
• 40. Shafqat, W., & Byun, Y. C., Incorporating similarity measures to optimize graph Convolutional neural network for product recommendation, Applied Sciences 11 (4), 1366, 2021.
• 41. Zhang, Y. J., & Hu, L. S., Fault Propagation Inference Based on a Graph Neural Network for Steam Turbine Systems, Energies 14, 309, 2021.
• 42. Şahin, E & Yüce, H. "Prediction of Water Leakage in Pipeline Networks Using Graph Convolutional Network Method, Applied Sciences, 13, 1-16, 2023.
• 43. Peng, H., et al., Lingcn: Structural linearized graph convolutional network for homomorphically encrypted inference, Advances in Neural Information Processing Systems 36, 2024.
• 44. Wu, Z., et al., A comprehensive survey on graph neural networks, IEEE Transactions on Neural Networks and Learning Systems 32 (1), 4-24, 2020.
• 45. Kipf, T. N., & Welling, M., Semi-supervised classification with graph convolutional networks, arXiv preprint arXiv: 1609.02907, 2016.
• 46. Tong, H., et al., Detection and classification of transmission line transient faults based on graph convolutional neural Network, CSEE Journal of Power and Energy Systems 7 (3), 456-471, 2021.
• 47. Schlichtkrull, M., et al., Modeling relational data with graph convolutional networks, In the Semantic Web: 15th International Conference, ESWC 2018, Heraklion, Crete, Greece, June 3–7, 2018, Proceedings 15. Springer International Publishing, 2018.
• 48. Hamilton, W., Ying, Z., & Leskovec, J., Inductive representation learning on large graphs. Advances in Neural Information Processing Systems 30, 2017.
• 49. Cho, H. N., et al., Heterogeneous graph construction and HinSAGE learning from electronic medical records, Scientific Reports 12 (2), 1152, 2022.
• 50. Li, Q., et al., Predicting wheat gluten concentrations in potato starch using GPR and SVM models built by terahertz time-domain spectroscopy, Food Chemistry 432, 137235, 2024.
• 51. Aymaz, S., A novel hybrid approach to multi focus image fusion using CNN and SVM methods, Gazi University Journal of the Faculty of Engineering and Architecture, 39 (2), 1123–1136, 2023.
• 52. Hatipoğlu, A., Güneri, Y., Yilmaz, E., A comparative predictive maintenance application based on machine and deep learning, Gazi University Journal of the Faculty of Engineering and Architecture, 39 (2), 1037-1048, 2024.
• 53. Chen, J., et al., Prediction of ovarian cancer related metabolites based on graph neural network, Frontiers in Cell and Developmental Biology 9, 753221, 2021.