Machine Learning Techniques for the Analysis of Magnetic Flux Leakage Images in Pipeline Inspection

Abstract

The magnetic flux leakage (MFL) technique is commonly used for non-destructive testing of oil and gas pipelines. This testing involves the detection of defects and anomalies in the pipe wall, and the evaluation of the severity of these defects. The difficulty with the MFL method is the extent and complexity of the analysis of the MFL images. In this paper we show how modern machine learning techniques can be used to considerable advantage in this respect. We apply the methods of support vector regression, kernelization techniques, principal component analysis, partial least squares, and methods for reducing the dimensionality of the feature space. We demonstrate the adequacy of the performance of these methods using real MFL data collected from pipelines, with regard to the performance of both the detection of defects, and the accuracy in the estimation of the severity of the defects. We also show how low-dimensional latent variable structures can be effective for visualizing the clustering behaviour of the classifier.

Theory and Application of Magnetic Flux Leakage Pipeline Detection

“…The core idea of segmentation threshold compression algorithms is that in the place where the original amplitude of the signal is relatively large little or no compression can be used, and in the smaller amplitude areas large compression ratios are chosen. Noise reduction is needed to improve the signal quality and increase the accuracy of data analysis [ 73 , 74 ].…”

Multi-Sensor Data Integration Using Deep Learning for Characterization of Defects in Steel Elements

“…The big data sets that must be analyzed by the systems result in a strong need of implementation of new processing algorithms. The procedures should be responsible, at the same time, for data collection, transformation, pattern search, and decision making [ 12 , 16 , 17 , 18 , 19 , 20 , 21 ]. An example of such an approach is the use of machine learning algorithms to automatically search even very complex relations between a large amount of input data [ 17 , 18 , 20 , 21 , 22 , 23 ].…”

Rényi Entropy Filter for Anomaly Detection With Eddy Current Remote Field Sensors

“…A comparative study between magneticand ultrasonic-based inspection techniques is reported in [28]. The authors in [29] proposed and compared three machine learning approaches, namely, support vector machine, kernelized principal component analysis, and kernelized partial least squares, to filter the sensor noise.…”