Data-driven Methods to Predict the Burst Strength of Corroded Line Pipelines Subjected to Internal Pressure

Abstract

A corrosion defect is recognized as one of the most severe phenomena for high-pressure pipelines, especially those served for a long time. Finite-element method and empirical formulas are thereby used for the strength prediction of such pipes with corrosion. However, it is time-consuming for finite-element method and there is a limited application range by using empirical formulas. In order to improve the prediction of strength, this paper investigates the burst pressure of line pipelines with a single corrosion defect subjected to internal pressure based on data-driven methods. Three supervised ML (machine learning) algorithms, including the ANN (artificial neural network), the SVM (support vector machine) and the LR (linear regression), are deployed to train models based on experimental data. Data analysis is first conducted to determine proper pipe features for training. Hyperparameter tuning to control the learning process is then performed to fit the best strength models for corroded pipelines. Among all the proposed data-driven models, the ANN model with three neural layers has the highest training accuracy, but also presents the largest variance. The SVM model provides both high training accuracy and high validation accuracy. The LR model has the best performance in terms of generalization ability. These models can be served as surrogate models by transfer learning with new coming data in future research, facilitating a sustainable and intelligent decision-making of corroded pipelines.

Keywords

Pipelines / Corrosion / Burst strength / Internal pressure / Data-driven method / Machine learning

References

[1] Abbas M, Shafiee M. An overview of maintenance management strategies for corroded steel structures in extreme marine environments. Marine Structures, 2020, 71: 102718

[2] Amaya-Gómez R, Sánchez-Silva M, Bastidas-Arteaga E, Schoefs F, Munoz F. Reliability assessments of corroded pipelines based on internal pressure a review. Engineering Failure Analysis, 2019, 98: 190-214

[3] Amaya-Gómez R, Sánchez-Silva M, Muñoz F. Pattern recognition techniques implementation on data from in-line inspection (ili). Journal of Loss Prevention in the Process Industries, 2016, 44: 735-747

[4] ASME B31G, A (1991) Manual for determining the remaining strength of corroded pipelines. ASME B31G-1991

[5] ASME B31G, A (2012) Manual for Determining the Remaining Strength of Corroded Pipelines: A Supplement to ASME B31 Code for Pressure Piping: an American National Standard. American Society of Mechanical Engineers

[6] Astanin V, Borodachev N, Bogdan S, Kol’tsov V, Savchenko N, Vinogradskii P (2009) Strength of corroded pipelines. Strength of materials 41

[7] Awad M, Khanna R. Support vector regression in efficient learning machines, 2015, Berkeley, CA: Apress, 67-80

[8] Benjamin AC, Vieira RD, Freire JLF, de Castro JT (2000) Burst tests on pipeline with long external corrosion, in: International pipeline conference, American Society of Mechanical Engineers. p. V002T06A013

[9] Bennett KP, Mangasarian OL. Robust linear programming discrimination of two linearly inseparable sets. Optimization methods and software, 1992, 1: 23-34

[10] Bishop CM (2006) Pattern recognition and machine learning. springer

[11] Boser BE, Guyon IM, Vapnik VN (1992) A training algorithm for optimal margin classifiers, in: Proceedings of the fifth annual workshop on Computational learning theory, pp: 144–152

[12] Box GE, Cox DR. An analysis of transformations. Journal of the Royal Statistical Society: Series B (Methodological), 1964, 26: 211-243

[13] Cai J, Chen G, Lützen M, Rytter NGM. A practical ais-based route library for voyage planning at the pre-fixture stage. Ocean Engineering, 2021, 236: 109478

[14] Cai J, Jiang X, Lodewijks G (2017) Residual ultimate strength of offshore metallic pipelines with structural damage-a literature review. Ships and Offshore Structures: 1–19. https://doi.org/10.1080/17445302.2017.1308214

[15] Cai J, Jiang X, Lodewijks G, Pei Z, Wu W. Residual ultimate strength of damaged seamless metallic pipelines with combined dent and metal loss. Marine Structures, 2018, 61: 188-201

[16] Cai J, Jiang X, Lodewijks G, Pei Z, Wu W. Residual ultimate strength of damaged seamless metallic pipelines with metal loss. Marine Structures, 2018, 58: 242-253

[17] Cai J, Jiang X, Lodewijks G, Pei Z, Zhu L. Experimental investigation of residual ultimate strength of damaged metallic pipelines. Journal of Offshore Mechanics and Arctic Engineering, 2019, 141: 1-21

[18] Chauhan V, Crossley J, et al (2009) Corrosion Assessment Guidance for High Strength Steels (Phase 1). Technical Report. GL Industrial Services UK Ltd

[19] Chen Y, Zhang H, Zhang J, Li X, Zhou J. Failure analysis of high strength pipeline with single and multiple corrosions. Materials & Design, 2015, 67: 552-557

[20] Chen Y, Zhang J, Zhang H, Liu X, Li X, Zhou J, Cao J. Ultimate load capacity of offshore pipeline with arbitrary shape corrosion defects. China Ocean Engineering, 2015, 29: 241-252

[21] Chin KT, Arumugam T, Karuppanan S, Ovinis M. Failure pressure prediction of pipeline with single corrosion defect using artificial neural network. Pipeline Science and Technology, 2020, 4: 10-17

[22] Choi J, Goo B, Kim J, Kim Y, Kim W. Development of limit load solutions for corroded gas pipelines. International Journal of Pressure Vessels and Piping, 2003, 80: 121-128

[23] Cortes C, Vapnik V. Support-vector networks. Machine learning, 1995, 20: 273-297

[24] Cronin DS, Pick RJ (2000) Experimental database for corroded pipe: evaluation of rstreng and b31g, in: 2000 3rd International Pipeline Conference, American Society of Mechanical Engineers Digital Collection

[25] Cronin DS, Roberts KA, Pick RJ (1996) Assessment of long corrosion grooves in line pipe, in: International Pipeline Conference, American Society of Mechanical Engineers. pp: 401–408

[26] De Masi G, Gentile M, Vichi R, Bruschi R, Gabetta G (2015) Machine learning approach to corrosion assessment in subsea pipelines, in: OCEANS 2015-Genova, IEEE. pp: 1–6

[27] DNV (2017) Recommended practice dnvgl-rp-f101 corroded pipelines

[28] Downey AB (2011) Think stats. “O’Reilly Media, Inc.”

[29] Freire J, Vieira R, Castro J, Benjamin A. Part 3: Burst tests of pipeline with extensive longitudinal metal loss. Experimental Techniques, 2006, 30: 60-65

[30] Fukami K, Fukagata K, Taira K (2020) Assessment of supervised machine learning methods for fluid flows. Theoretical and Computational Fluid Dynamics: 1–23

[31] Ghaboussi J, Sidarta D. New nested adaptive neural networks (nann) for constitutive modeling. Computers and Geotechnics, 1998, 22: 29-52

[32] Gholami H, Shahrooi S, Shishesaz M. Predicting the burst pressure of high-strength carbon steel pipe with gouge flaws using artificial neural network. Journal of Pipeline Systems Engineering and Practice, 2020, 11: 04020034

[33] Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks, in: Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp: 249–256

[34] Haghighat E, Raissi M, Moure A, Gomez H, Juanes R (2020) A deep learning framework for solution and discovery in solid mechanics: linear elasticity. arXiv preprint arXiv: 2003.02751

[35] Haykin S (2007) Neural networks: a comprehensive foundation. Prentice-Hall, Inc

[36] Hoerl AE, Kennard RW. Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 1970, 12: 55-67

[37] Kiefiier J, Vieth P (1990) New method corrects criterion for evaluating corroded pipe, oil and gas journal, aug. 6

[38] Kiefner J, Vieth P (1989) Project pr-3-805: A modified criterion for evaluating the remaining strength of corroded pipe. Pipeline Corrosion Supervisory Committee of the Pipeline Research Committee of the American Gas Association

[39] LeCun Y, Bengio Y, Hinton G. Deep learning. Nature, 2015, 521: 436-444

[40] Ling J, Templeton J. Evaluation of machine learning algorithms for prediction of regions of high reynolds averaged navier stokes uncertainty. Physics of Fluids, 2015, 27: 085103

[41] Liu H, Liu Z, Taylor B, Dong H. Matching pipeline in-line inspection data for corrosion characterization. NDT & E International, 2019, 101: 44-52

[42] Macdonald K, Cosham A. Best practice for the assessment of defects in pipelines-gouges and dents. Engineering Failure Analysis, 2005, 12: 720-745

[43] Mattioli M, Cherubini P, Baldoni A (2019) New frontiers for pipeline integrity management

[44] Mohd MH, Lee BJ, Cui Y, Paik JK. Residual strength of corroded subsea pipelines subject to combined internal pressure and bending moment. Ships and offshore Structures, 2015, 10: 554-564

[45] Mok D, Pick R, Glover A, Hoff R. Bursting of line pipe with long external corrosion. International journal of pressure vessels and piping, 1991, 46: 195-216

[46] Naftaly U, Intrator N, Horn D. Optimal ensemble averaging of neural networks. Network: Computation in Neural Systems, 1997, 8: 283-296

[47] Nair V, Hinton GE (2010) Rectified linear units improve restricted boltzmann machines, in: ICML

[48] Ossai CI. Corrosion defect modelling of aged pipelines with a feed-forward multi-layer neural network for leak and burst failure estimation. Engineering Failure Analysis, 2020, 110: 104397

[49] Rafiei MH, Adeli H. A novel machine learningbased algorithm to detect damage in high-rise building structures. The Structural Design of Tall and Special Buildings, 2017, 26: e1400

[50] Rosen J, Potts A, Sincock P, Carra C, Kilner A, Kriznic P, Gumley J, (2016) Novel methods for asset integrity management in a low oil-price environment, in: Offshore Technology Conference, Offshore Technology Conference. https://doi.org/10.4043/27076-MS

[51] Rumelhart DE, Hinton GE, Williams RJ. Learning representations by back-propagating errors. nature, 1986, 323: 533-536

[52] Sen D, Aghazadeh A, Mousavi A, Nagarajaiah S, Baraniuk R, Dabak A. Data-driven semi-supervised and supervised learning algorithms for health monitoring of pipes. Mechanical Systems and Signal Processing, 2019, 131: 524-537

[53] Smola AJ, Schölkopf B. A tutorial on support vector regression. Statistics and computing, 2004, 14: 199-222

[54] Specification, A (2004) 51, specification for line pipe. Edition March. Tabachnick BG, Fidell LS, Ullman JB (2007) Using multivariate statistics. volume 5. Pearson Boston, MA

[55] Taylor ME, Stone P (2009) Transfer learning for reinforcement learning domains: A survey. Journal of Machine Learning Research 10

[56] Tibshirani R. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 1996, 58: 267-288

[57] Weiss K, Khoshgoftaar TM, Wang D. A survey of transfer learning. Journal of Big data, 2016, 3: 1-40

[58] Xie M, Tian Z. A review on pipeline integrity management utilizing in-line inspection data. Engineering Failure Analysis, 2018, 92: 222-239

[59] Zou H, Hastie T. Regularization and variable selection via the elastic net. Journal of the royal statistical society: series B (statistical methodology), 2005, 67: 301-320