Analysis of Magnetic-Flux Leakage (MFL) Data for Pipeline Corrosion Assessment
Xiang Peng,Uchenna Anyaoha,Zheng Liu,Kazuhiko Tsukada
DOI: https://doi.org/10.1109/tmag.2020.2981450
IF: 1.848
2020-06-01
IEEE Transactions on Magnetics
Abstract:Oil and gas pipelines transport and distribute large quantities of oil products and natural gas to industrial and residential customers over a long distance. However, pipeline failures could lead to enormous hazards and safety issues. Metal loss, due to corrosion, is a significant cause for pipeline failures. To detect and quantify metal loss, in-line inspection (ILI) is carried out periodically to assess the integrity of the pipeline. Among all the ILI techniques, magnetic-flux leakage (MFL) is the most popular one due to its high efficiency, robustness, and good applicability to both oil and gas pipelines. This article provides a comprehensive review of the pipeline corrosion assessment with the MFL technique from the data analytic perspective. The analyses of the MFL signal and data contribute to both corrosion quantification and prediction. In this article, the state of the art for the MFL measurement technique is briefly reviewed first. For corrosion quantification, the signal processing methods together with the characterization models, which aim to enhance the measurement and characterize the corrosion, are described and discussed. For corrosion prediction, this article investigates multiple MFL data matching methods, which align defects from successive ILI runs. Subsequently, corrosion growth models, which aim to predict the future corrosion status, are presented. Besides, the reliability analysis of the corroded pipeline is reviewed. The potential of fusing MFL with other non-destructive testing (NDT) techniques are explored as well. At the end of this article, we summarize the existing issues and describe the trends for future research on pipeline corrosion assessment.
engineering, electrical & electronic,physics, applied