آزمایش و مطالعه شبیه سازی در مورد سنجش میدان مغناطیسی سه بعدی برای خصوصیات نقص نشت شار مغناطیسی

Magnetic flux leakage (MFL) testing is widely used to detect and characterise defects in pipelines, rail tracks and other structures. The measurement of the two field components perpendicular to the test surface and parallel to the applied field in MFL systems is well established. However, it is rarely effective when the shapes of the specimens and defects with respect to the applied field are arbitrary. In order to overcome the pitfalls of traditional MFL measurement, measurement of the three-dimensional (3D) magnetic field is proposed. The study is undertaken using extensive finite element analysis (FEA) focussing on the 3D distribution of magnetic fields for defect characterisation and employing a high sensitivity 3-axis magnetic field sensor in experimental study. Several MFL tests were undertaken on steel samples, including a section of rail track. The experimental and FEA test results show that data from not only the x- and z-axes but also y-axis can give comprehensive positional information about defects in terms of shape and orientation, being especially advantageous where the defect is aligned close to parallel to the applied field. The work concludes that 3D magnetic field sensing could be used to improve the defect characterisation capabilities of existing MFL systems, especially where defects have irregular geometries.

Magnetic flux leakage (MFL) is one of the most widely used electromagnetic NDT techniques and has been used in pipeline inspection gauges (PIGs) for gas pipeline inspection since the 1960s [1], [2], [3], [4] and [5]. MFL testing relies on the fact that when a magnetic field is applied to a ferromagnetic material, any geometrical discontinuity and local gradients in magnetic permeability in the test material will cause the field to leak out of the material, into the air. This flux leakage is measured by a magnetic field sensor and used to estimate the dimensions of the defect. Recent advances in the field include the development of residual field measurement techniques [6], [7] and [8]; where the magnetic field remaining in the sample from magnetisation applied during previous testing or generated by stresses in the sample is measured, reflecting both geometrical discontinuities and variations in material properties of the sample due to applied stresses, etc. The concept of 2-axis measurement of the field components perpendicular to the surface of the material under test (Bz) and parallel to the applied field (Bx) has been investigated via experiment and finite element analysis (FEA) in previous MFL research, concentrating on the MFL signal response to regularly shaped defects [9] and [10]. The two-dimensional (2D) magnetic field over defects has given sufficient information for defect characterisation, as a result of which the third field component (By) and its utilisation have not caught much attention in industry [11]. However, in in-situ conductive specimens, defects with irregular shapes are more frequently found than regularly shaped defects. Therefore, the traditional measurement of 2D magnetic field cannot accurately characterise natural defects, e.g. stress corrosion cracks [12], [13] and [14]. In this work, following the FEA investigation, measurement of the 3D magnetic field was conducted and its contribution to defect characterisation, focussing on location and orientation was investigated. The residual magnetisation technique has been used along with a low field range, high-resolution magneto-resistive sensor designed for compassing applications to investigate the possible advantages of using 3-axis field measurement to provide extra information for defect characterisation in terms of shape and orientation. The paper is organised as follows: Section 2 presents FEA of MFL with an irregular-shaped defect and discusses the advantage of 3D magnetic field sensing in the characterisation of irregular defects; Section 3 elaborates on the experimental study for MFL inspection of a slotted plate and a section of rail track with a naturally occurring, irregularly shaped crack. The experimental results show good correlation with the simulation and indicate that 3D magnetic field measurement has a lot of potential for the characterisation of irregularly shaped defects using MFL.

The FEA of MFL unveils the magnetic field over the defect region, and particularly the distribution of By in the immediate area of the irregularly shaped defect, which shows that measurement of the y-axis field component not only complements the x- and z-axes signals but also enhances the detection and characterisation of defects in specimens. The simulation study also indicates the necessity of 3D magnetic field measurement for characterisation of irregular-shaped defects especially in terms of shape and orientation during MFL inspection. Following the FEA analysis, two tests were undertaken for MFL inspection of specimens with a rectangular slot and a natural irregularly shaped crack. The magnetic field was quantified using a 3-axis AMR field sensor. The first test shows that with a clearly defined, machined slot, the y-axis sensor gives a predictable output containing signal features that clearly correspond to the defect position. Although the rail track test results are not as significant as the results from the machined slot, they exhibit a good correlation between defect position and sensor output. By is particularly useful in detecting the diagonally orientated section of the crack, whereas Bx and Bz give very little indication of crack position. Although there are localised discrepancies in the magnetic field distribution between experimental and simulated results, due to the inhomogeneous field and nonlinear material of the arbitrary shaped test samples, the overall distribution of the 3D magnetic field from experimental study has good agreement with FEA simulation results. These initial tests indicate that the use of a 3-axis system would be advantageous in certain situations to give orientation information, especially where irregularly shaped defects or defects orientated close to parallel to the applied field are expected. Future work will involve an investigation into the implementation of the 3-axis system to detect defects in components with irregular surface geometries, e.g. free curvature surface, and evaluation of feature extraction techniques for three axis signals via experimental effort and FEA. Feature extraction for the signals is a particularly important task for defect and material characterisation. So the contribution of variations in material properties to the results will be investigated and techniques developed to decouple them from signal components influenced by geometrical discontinuities alone.