For decades, magnetic flux leakage (MFL) technology has been recognized as one of the most reliable techniques for detecting metal loss in pipelines.

However, traditional MFL approaches – whether axial (MFL-A) or circumferential (MFL-C) – rely on a single unidirectional magnetic field, resulting in limitations in both detection and sizing capabilities following the morphology of the metal loss anomalies (i.e., complex defects). To address these limitations, many pipelines are inspected using both techniques. The two collected datasets are analyzed separately, with the results combined in one report. Unfortunately, this approach introduces additional complications, including additional testing to handle subjective and conflicting results. Moreover, metal loss reporting does not provide sufficient information on the depth profiles of anomalies for the increasingly needed performance of realistic and accurate failure pressure calculations.

The fusion of MFL-A and MFL-C data leverages the complementary aspects of the perpendicular and axial magnetic fields, enhancing feature characterization across all defect morphologies. This fused data enables the creation of a comprehensive pipe surface map, including a high-resolution 3D depth profile of all detected anomalies. The resulting river bottom profile (RBP) serves as an enhanced input for analysis activities such as the classification and sizing of anomalies, which significantly increases the certainty of the findings and improves burst pressure calculations. The depth prediction and the burst pressure calculation are validated using API STD 1163 Level 3 methods.

The MFL Data Fusion process is powered by recent advancements in algorithms, big-data management and neural networks (AI), leveraging MFL inspection technology to achieve a new level of consistency and accuracy. This breakthrough significantly improves the certainty and reliability of MFL inspection results.