Modeling of pipeline corrosion degradation mechanism with a Lévy Process based on ILI (In-Line) inspections

Abstract

In pipelines, one of the primary testing procedures used to identify the effects and evolution of corrosion over time is through In-Line Inspections (ILI). ILI inspections provide detailed information regarding the inner and outer pipeline condition based on the remaining wall thickness. Based on this information, different approaches have been proposed to predict the degradation extent of the defects detected. However, these predictions are subject of uncertainties due to the inspection tool and the degradation process that poses some challenges for assessing an entire pipeline within the timespan between two inspections. To address this problem, ILI data was used to formulate a degradation model for steel-pipe degradation based on a Mixed Lévy Process. The model combines a Gamma and Compound Poisson Processes aimed for a better description of the degradation reported by the ILI data. The model seeks to estimate corrosion lifetime distribution and the mean time to failure (MTTF) more accurately. The model was tested on an actual segment of an oil pipeline, and the results have been used to support a preventive maintenance program.