TY - JOUR
PY - 2021//
TI - Deep learning-based reliability model for oil and gas pipeline subjected to stress corrosion cracking: a review and concept
JO - Hunan Daxue Xuebao
A1 - Soomro, Afzal Ahmed
A1 - Mokhtar, Ainul Akmar
SP -
EP -
VL - 48
IS - 5
N2 - Stress corrosion cracking is considered one of the major causes of failures in oil and gas pipelines. This is why modeling the reliability of oil and gas pipelines subjected to stress corrosion cracking is very important; at the same time it is very complex due to various parameters affecting the stress corrosion cracking. Modern modeling approaches include physical-based and data-driven models that are still not competitive to cope with the complex nature of the stress corrosion cracking mechanism. In today's research, researchers prefer machine learning oriented algorithms and models to address such complex mechanisms due to their increasing popularity. These algorithms and models have the capability of tackling multiple factors and their impact on output response, allowing a prediction of the probability of failure. This research proposes some extensive simulations that lead eventually to a rich dataset that will define some significant factors on which stress corrosion cracking depends. In addition to this, the proposed research not only involves the correlation of derived dataset with the already published dataset but will also provide a comprehensive validation in between the proposed experimental work and machine learning based simulations. This research aims to propose a model that considers the most frequent parameters so that the performance of the proposed technique can be evaluated robustly and may provide a better understanding to upcoming researchers, including oil and gas personals. Keywords: corrosion, finite element, reliability, machine learning, artificial intelligence. Keywords: Stationary Transportation; Pipeline Transportation Language: en
LA - en
SN - 1674-2974
UR - http://dx.doi.org/
ID - ref1
ER -