A review on oil and gas pipelines corrosion growth rate modelling incorporating artificial intelligence approach

Abstract

One of the necessities of an effective oil and gas pipeline safety Management Plan (SMP) is the establishment of safe and efficient risk assessment strategy for pipelines where the significant danger is corrosion. Corrosion growth is related to several factors involving pipe material, pipe condition, and defect geometrical imperfection. Thus, the assurance of a proper corrosion assessment requires the prediction and evaluation of corrosion growth rates. The prediction of corrosion growth rate precisely, would minimize the cost of pipelines maintenance through the determination of the deteriorated pipeline segments. In line inspection (ILI) has been used to detect the pipelines corrosion, also the corrosion can be detected by other inspection tools such as Magnetic flux leakage (MFL) and Ultrasonic tool (UT). However, there are numerous models have been utilized to anticipate the corrosion growth rate such as deterministic and probabilistic models. Recently, there are conducted researches on the application of artificial intelligence in predicting corrosion growth rate for oil and gas pipelines such as artificial neural network (ANN) and fuzzy logic (FL). This paper aims to provide a comprehensive comparison between the conventional methods, i.e. deterministic and probabilistic and artificial intelligence methods, i.e. Artificial neural network (ANN) and fuzzy logic (FL) in the prediction of corrosion growth rate of oil and gas pipelines. This review would be helpful to pipelines operators to understand the effectiveness of artificial intelligence approach compared to conventional methods in corrosion growth rate modelling.

Pipeline In-Line Inspection Method, Instrumentation and Data Management

“…These models are based on the physical properties and mechanics of a certain phenomenon, which are known as deterministic models as well. Meanwhile, linear, non-linear fitting equations and single value degradation models related to the failure of the pipelines are available [ 173 ]. By assuming the linear process in defects growth, the linear growth rate models are applied to forecast the depth of defects in linear modelling.…”

“…This information can be obtained by appropriate mathematical physics models or artificial intelligence technologies with intelligent data management. According to the datasets, pipeline safety planning and management should take place without delay if the defect growth rates are high or the predicted lifetime is coming to an end [ 173 ].…”

Burst pressure strength of corroded subsea pipelines repaired with composite fiber-reinforced polymer patches

“…1. Rehabilitation of corroded steel pipelines specimens [10] Extensive research in offshore engineering has led to the implementation of the Artificial Neural Networks (ANN) in corroded subsea pipelines, which have been carried out by many researchers [11]- [13].…”