Pipeline Integrity Management

Corrosion Growth Model, Reliability and Resiliency Assessment of Infrastructure Network

Resiliency and reliability assessment of natural gas pipeline plays a critical role to maintain a satisfactory level of market demand. Corrosion is one of the critical problem which can lead to the deterioration of pipeline materials, posing significant risks to safety, the environment, and operational efficiency. Therefore, assessing and ensuring the reliability of pipelines with respect to corrosion is of paramount importance. With this in view, we focus on the developing the stochastic growth model for pitting corrosion which involves pit initiation and pit growth. Also, these pipelines may fail due to burst, collapse, leakage, deflection etc. Conditional generative adversarial network is proposed for prediction of horizontal ground displacement considering soil properties, liquefaction, seismic effect etc. Currently we are looking for system reliability assessment of pipeline network using graph neural network which utilizes the effect of physical topology and node properties in pipeline networks.

[1] Woldesellasse, H. and Tesfamariam, S., 2023. Consequence assessment of gas pipeline failure caused by external pitting corrosion using an integrated Bayesian belief network and GIS model: Application with Alberta pipeline. Reliability Engineering & System Safety, 240, p.109573.

[2] Woldesellasse, H. and Tesfamariam, S., 2023. Failure assessment of oil and gas transmission pipelines using an integrated Bayesian belief network and GIS model. International Journal of Pressure Vessels and Piping, p.104984.

[3] Woldesellasse, H. and Tesfamariam, S., 2023. Data augmentation using conditional generative adversarial network (cGAN): Application for prediction of corrosion pit depth and testing using neural network. Journal of Pipeline Science and Engineering, 3(1), p.100091.

[4] Balekelayi, N. and Tesfamariam, S., 2023. Bayesian geoadditive water pipe failure forecasting model by optimizingthe updating period. Journal of Hydroinformatics, 25(1), pp.1-19.