Mojtaba Rostami Kandroodi - Control and Intelligent Processing Center of Excellence, School of ECE, College of Engineering,
Babak Nadjar Araabi
Majid Nili Ahmadabadi
Farshad Shirani

Magnetic Flux Leakage (MFL) testing is the most widely used non-destructive techniques for the in-service inspection of oil and gas pipelines. This study presents novelapproaches to detect defects by employing MFL signals. Estimating the number of defects, locations, and orientationsfrom measurements is a typical inverse problem inelectromagnetic non-destructive evaluations (NDE). A detection algorithm on axial flux is proposed for defect detection basedon image processing approaches and morphological methods. Finally, the efficacy and accuracy of the proposed algorithm isvalidated through examinations on simulated defects and real experimental MFL data. Simulated defects are generated inpresence of multiple uncertainties and noises; including thevariation of the pipeline thickness, sensors liftoff, magnetization level, and shape irregularity.

MFL, defect detection, image processing, axial flux, radial flux, non-destructive testing