2024 – 65

Anthony Tindall1, Steve Farnie1, Jeff Sutherland2, Stuart Clouston2

1Baker Hughes, Cramlington, UK., 2Baker Hughes, Calgary, Canada

Abstract

It is well known that when corrosion pits interact, the interpretation of inspection data gathered by Magnetic Flux leakage tools (an indirect measurement technique) becomes challenging due to the nature of the overlapping and superimposing signal responses across the multi-defect profiles. Such “complex” corrosion is of particular concern to pipeline operators as it is these areas, particularly axially oriented, that are often the most critical to pipeline integrity. And then, being the most difficult to both interpret and size accurately, these can lead to “outliers” relative to conventional ILI specifications and industry guidelines (e.g. API 1163). Such outliers are increasingly grouped into two types, namely Safety Outliers (where defect severity is under called leading to underestimated remaining strength) or Resource Outliers (when defect severity is overcalled potentially leading to unnecessary digs).

The goal of any in-line inspection method is to represent corrosion fully and accurately on the pipe. To achieve best performance, we must be able to first detect when these complex or interacting signals occur and then, more importantly, interpret them correctly.  In physics terms, the nature of MFL field leakage is a 3-dimensional vector and predictable, yet different signal responses to metal loss (and other features) are observed in each of these 3 vectors which can be quantified and characterized as an improved means to translate leakage signals to the corrosion they more accurately represent.

Single axis MFL, whether conducted with axially or circumferentially oriented magnetizers, will reach limits in its capabilities regardless of sensing resolution. MFL is a mature technology, and an operator today should expect most tools to perform to specification on isolated corrosion. However, the adoption of long proven Triaxial sensors, at optimal resolution, on Magnetic Flux Leakage inspection vehicles provides an integrity engineer significant advantages to minimize potential outliers when things are not so straight forward.

This paper will outline how the three independent components of the triaxial flux leakage response provide unique identifiers for cases of axial corrosion, highly asymmetrical defects, corrosion in corrosion, wide area erosion/corrosion with pitting, to name but a few. At the same time, the additional signal components provide independent measures that enable better performance in latest generation algorithm techniques and development than those trained on fewer inputs.

The authors will present real world examples of these unique identifiers being used in practice to interpret highly complex corrosion and the latest work Baker Hughes has been conducting in close partnership with our customers with the common goal to best manage these “complex morphologies” successfully and efficiently with one inspection.

Read/Download PaperGrade