Ferromagnetic material defect identification method and system based on magnetic gradient tensor measurement

Abstract

The application discloses a ferromagnetic material defect identification method and system based on magnetic gradient tensor measurement, relates to the technical field of nondestructive testing, and comprises the following steps: collecting and acquiring magnetic gradient tensor data of a measured piece; calculating total horizontal derivatives based on the magnetic gradient tensor data; calculating partial derivatives of each component in the z direction based on the total horizontal derivatives and calculating vertical derivatives based on the magnetic gradient tensor data; identifying defect boundaries by using a magnetic gradient tensor fusion derivative detector and a magnetic gradient tensor noise-resistant boundary detector, and outputting a defect contour map. The application first proposes the MGT_FDED and MGT_NRED detectors specially designed for the industrial defect detection scene, solves the problem that the direct transplantation of the geophysical method is not suitable, and can accurately identify the edges of various defects from surface cracks to deep buried holes, thereby significantly improving the accuracy, noise resistance and intelligent level of detection.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing technology, and in particular to a method and system for identifying defects in ferromagnetic materials based on magnetic gradient tensor measurement. Background Technology

[0002] Ferromagnetic components are widely used in industrial facilities, and their structural integrity is crucial. Currently, magnetic non-destructive testing methods such as Metal Magnetic Memory testing (MMM) and Magnetic Fluxleakage testing (MFL) are widely used for defect detection in these components. However, traditional methods often suffer from limitations in accurately identifying the edges of defects, especially small, deep, or complex-shaped defects, such as blurred boundaries, insensitivity to defect types, and susceptibility to environmental noise interference.

[0003] In the field of geophysical exploration, a series of mature techniques based on magnetic gradient tensors have been developed for the boundary identification of magnetic targets, such as the vertical derivative method, the total horizontal derivative method, and the analytical signal amplitude method. These methods sharpen the boundary by enhancing the gradient changes in the field data.

[0004] Traditional magnetic gradient detection methods have significant limitations: limited detection depth, making it difficult to identify deep defects; diffuse boundaries, resulting in non-convergence; unstable detection results, with slight lift-off variations causing fluctuations, requiring precision mechanical devices to maintain a constant lift-off height; and poor field applicability, making detection difficult on curved or irregular surfaces.

[0005] These problems become particularly prominent when boundary identification technology in the field of geophysical exploration is directly applied to defect detection scenarios.

[0006] Therefore, there is an urgent need for a high-precision magnetic gradient tensor analysis method that is specific to industrial defect detection scenarios and can identify the edges of various defects. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a magnetic gradient tensor defect identification method and system that can adapt to different defect types, achieve accurate edge identification, and has strong noise resistance. This solves the problem of poor performance when traditional geophysical boundary identification methods are directly applied to industrial defect detection. It is suitable for accurate edge identification of various defects such as cracks, holes, corrosion, and stress damage in pipelines, pressure vessels, and steel structures.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] On the one hand, the present invention provides a method and system for defect identification in ferromagnetic materials based on magnetic gradient tensor measurement, comprising the following steps:

[0010] The magnetic gradient tensor data of the test piece is acquired using an AMR magnetic gradient sensor array and then preprocessed.

[0011] Calculate the total horizontal derivative based on the magnetic gradient tensor data;

[0012] Calculate the partial derivatives of each component in the z-direction based on the total horizontal derivative and calculate the vertical derivative based on the magnetic gradient tensor data;

[0013] Defect boundaries are identified using the Magnetic Gradient Tensor FusionDerivative Edge Detector (MGT_FDED) and the Magnetic Gradient Tensor Noise-Resistant Edge Detector (MGT_NRED), and defect contour maps are output.

[0014] In some embodiments, the surface of the workpiece is scanned by a magnetic gradient sensor array to obtain the magnetic induction intensity components Bx, By, and Bz, and the nine components of the magnetic gradient tensor (five of which are independent components) are calculated: Gxx, Gxy, Gxz, Gyx, Gyy, Gyz, Gzx, Gzy, and Gzz.

[0015] In some embodiments, the expression for calculating the total horizontal derivative is:

[0016] ;

[0017] ;

[0018] ;

[0019] In the formula, , , These are the components of the magnetic flux density vector in the x, y, and z directions, respectively. These are the components of the gradient tensor matrix; Let x be the x-component of the total horizontal derivative of the magnetic gradient tensor; Let y be the y-component of the total horizontal derivative of the magnetic gradient tensor; Let z be the z-component of the total horizontal derivative of the magnetic gradient tensor.

[0020] In some embodiments, the expression for calculating the partial derivatives of each component of the total horizontal derivative in the z-direction is as follows:

[0021] ;

[0022] ;

[0023] ;

[0024] In the formula, Let x be the x-component of the total horizontal derivative of the magnetic gradient tensor; Let y be the y-component of the total horizontal derivative of the magnetic gradient tensor; Let z be the z-component of the total horizontal derivative of the magnetic gradient tensor.

[0025] In some embodiments, the expression for calculating the vertical guide is:

[0026] ;

[0027] ;

[0028] ;

[0029] In the formula, These are the components of the gradient tensor matrix.

[0030] In some embodiments, the magnetic gradient tensor fusion derivative detector is:

[0031] ;

[0032] By enhancing the ability to extract deep weak signals through vertical guide, the noise resistance is significantly improved. More importantly, it solves the technical problem that traditional methods are sensitive to changes in lift-off value, and maintains stable detection performance over a wide range of lift-off values.

[0033] The magnetic gradient tensor noise-resistant boundary detector is:

[0034] ;

[0035] Preserve shallow high-frequency details based on the original field components and avoid edge smoothing effects.

[0036] In the formula, Let x be the x-component of the total horizontal derivative of the magnetic gradient tensor; Let y be the y-component of the total horizontal derivative of the magnetic gradient tensor; Let z be the z-component of the total horizontal derivative of the magnetic gradient tensor; These are the components of the gradient tensor matrix; It is a constant.

[0037] In some embodiments, based on the defect type, a lookup table is used to match the correspondence between different types of defects and the magnetic gradient tensor fusion derivative detector or the magnetic gradient tensor noise-resistant boundary detector.

[0038] In some embodiments, surface cracks, dense pores, or pitting defects are identified by the magnetic gradient tensor noise-resistant boundary detector; deep buried holes, stress cracks, or large-area corrosion are identified by the magnetic gradient tensor fusion derivative detector.

[0039] On the other hand, the present invention provides a defect identification system for ferromagnetic materials based on magnetic gradient tensor measurement, for performing the above method, including:

[0040] Data acquisition module: Uses an AMR magnetic gradient sensor array to acquire magnetic gradient tensor data of the test piece;

[0041] Data processing module: Calculates the total horizontal derivative based on the magnetic gradient tensor data;

[0042] Calculate the partial derivatives of each component in the z-direction based on the total horizontal derivative and calculate the vertical derivative based on the magnetic gradient tensor data;

[0043] Defect boundaries are identified using a magnetic gradient tensor fusion derivative detector and a magnetic gradient tensor noise-resistant boundary detector.

[0044] Output module: Displays the processing results of the data processing module.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] This invention introduces for the first time the MGT_FDED and MGT_NRED detectors specifically designed for industrial defect detection scenarios, solving the problem of incompatibility when directly transplanting geophysical methods. The MGT_FDED and MGT_NRED detectors are optimized for deep weak signals and shallow high-frequency signals, respectively, enabling clearer and more accurate identification of the edges of various defects, from surface to depth, and from linear cracks to volumetric pores. The MGT_FDED detector effectively suppresses surface noise through field extension vertical derivative, significantly improving the signal-to-noise ratio and exhibiting stable performance in complex industrial environments.

[0047] The present invention adopts a strategy of adaptively selecting detectors based on defect characteristics (types), which improves versatility and automation, and reduces reliance on operator experience.

[0048] This invention can accurately identify the edges of various defects, from surface cracks to deep buried holes, significantly improving the accuracy, noise resistance, and intelligence of the detection. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the overall process of Embodiment 1 of the present invention;

[0050] Figure 2The three-dimensional defect model used in the simulation experiment of this invention can only be calculated on an accurate three-dimensional model;

[0051] Figure 3 This is a schematic diagram comparing the magnetic gradient tensor noise-resistant boundary detector of Embodiment 1 of the present invention with existing technologies (surface cracks at small lift-off values ​​(MGT_NRED is better in the right figure, less affected by the geomagnetic field)).

[0052] Figure 4 This is a schematic diagram comparing the magnetic gradient tensor fusion derivative detector of Embodiment 1 of the present invention with existing technologies (the buried aperture under a large lift-off value (the left image MGT_FDED is superior, with clear and non-divergent boundaries, and less affected by the lift-off value)).

[0053] Figure 5 This is a schematic diagram of the structure of Embodiment 2 of the present invention. Detailed Implementation

[0054] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0055] Example 1:

[0056] Please see Figures 1-4 This invention presents a defect identification method for ferromagnetic materials based on magnetic gradient tensor measurement, applicable to the accurate edge identification of various defects such as cracks, holes, corrosion, and stress damage in pipes, pressure vessels, and steel structures. This embodiment uses the detection of a welded area on an X80 steel pipe as an example. This area contains a surface crack approximately 1 mm deep and a deep buried hole approximately 6 mm deep.

[0057] Specifically, the following steps are included:

[0058] S1. Acquire the magnetic gradient tensor data of the test piece by collecting data.

[0059] An AMR magnetic gradient sensor array was used to scan the weld area with a lift-off height of 1 mm and a step size of 0.5 mm. Three-component magnetic field data (Bx, By, Bz) of the steel pipe were obtained.

[0060] After data acquisition, the three magnetic field components Bx, By, and Bz are filtered (filter size: To improve the signal-to-noise ratio of BDA2 for shallow defects, preprocessing operations such as background field correction can be used. Background field correction is used to remove slowly varying background field interference from the original magnetic field data.

[0061] In a specific embodiment, the background field correction code, as shown in the Matlab code example, is as follows:

[0062] Use large window mid-range filtering to estimate the background field;

[0063] background_Bz = medfilt2(Bz, [50, 50]);

[0064] Subtract the background field from the original signal;

[0065] Bz_corrected = Bz - background_Bz;

[0066] The magnetic gradient tensor and related characteristics were recalculated using the corrected Bz.

[0067] The nine components (five of which are independent) of the magnetic gradient tensor were calculated using the central difference method: Gxx, Gxy, Gxz, Gyx, Gyy, Gyz, Gzx, Gzy, and Gzz. The grid spacing was dx = dy = 0.5 mm.

[0068] S2. Calculate the total horizontal derivative based on the magnetic gradient tensor data.

[0069] calculate , , Isomorphic characteristics:

[0070] ;

[0071] ;

[0072] ;

[0073] In the formula, , , These are the components of the magnetic flux density vector in the x, y, and z directions, respectively. These are the components of the gradient tensor matrix; Let x be the x-component of the total horizontal derivative of the magnetic gradient tensor; Let y be the y-component of the total horizontal derivative of the magnetic gradient tensor; Let z be the z-component of the total horizontal derivative of the magnetic gradient tensor.

[0074] S3. Based on the total horizontal derivative, calculate the partial derivatives of each component in the z-direction using the frequency domain field extension function, with the extension distance dz set to 0.01m.

[0075] ;

[0076] ;

[0077] ;

[0078] In the formula, Let x be the x-component of the total horizontal derivative of the magnetic gradient tensor; Let y be the y-component of the total horizontal derivative of the magnetic gradient tensor; Let z be the z-component of the total horizontal derivative of the magnetic gradient tensor.

[0079] S4. Calculate the vertical guide number based on the magnetic gradient tensor data.

[0080] ;

[0081] ;

[0082] ;

[0083] In the formula, These are the components of the gradient tensor matrix.

[0084] S5. Calculate two boundary recognition result matrices using a magnetic gradient tensor fusion derivative detector and a magnetic gradient tensor noise-resistant boundary detector to identify defect boundaries. Then, image the final adaptive boundary recognition result to provide inspectors with a clear defect contour diagram.

[0085] The magnetic gradient tensor fusion derivative detector is calculated based on the vertical derivative, and the calculation expression is:

[0086] ;

[0087] The magnetic gradient tensor fusion derivative detector utilizes vertical derivatives to enhance the extraction capability of deep, weak, and slowly varying signals, exhibiting good noise resistance. Through vertical derivative normalization, it effectively compensates for signal attenuation caused by increased lift-off, fundamentally solving the lift-off value sensitivity problem.

[0088] The vertical guide normalization adopts the tan-1(dTHDx_dz2 | dTHDy_dz2) structure.

[0089] The following is a Matlab code example of a magnetic gradient tensor fusion derivative detector:

[0090] numerator_FDED = dTHDx_dz.^2 + dTHDy_dz.^2;

[0091] denominator_FDED = dGxz_dz.^3 + dGyz_dz.^3 + eps;

[0092] FDED = numerator_FDED . / denominator_FDED.

[0093] The magnetic gradient tensor noise-resistant boundary detector is calculated based on the original field components, which better preserves high-frequency details and spatial distribution characteristics. The calculation expression is as follows:

[0094] ;

[0095] In the formula, Let x be the x-component of the total horizontal derivative of the magnetic gradient tensor; Let y be the y-component of the total horizontal derivative of the magnetic gradient tensor; These are the components of the gradient tensor matrix; It is a constant, a minimum value, to prevent division by zero errors.

[0096] The following is a Matlab example of a magnetic gradient tensor-based noise-resistant boundary detector:

[0097] numerator_NRED = THDx.^2 + THDy.^2;

[0098] denominator_NRED= Gxz.^3 + Gyz.^3+ eps;

[0099] NRED = numerator_NRED . / denominator_NRED;

[0100] Normalize FDED and NRED to zero mean and unit variance for fusion.

[0101] FDED_norm = (FDED - mean(FDED(:))) / (std(FDED(:)) + eps);

[0102] NRED_norm = (NRED - mean(NRED(:))) / (std(NRED(:)) + eps).

[0103] This embodiment uses a defect type-based strategy: a lookup table is established, and the optimal method is selected based on prior knowledge or preliminary identification results.

[0104] Table 1. Defect Type and Detector Correspondence Lookup Table

[0105]

[0106] The fundamental differences between the magnetic gradient tensor fusion derivative detector and the magnetic gradient tensor noise-resistant boundary detector in terms of basic principles and processing targets are as follows:

[0107] 1. Differences in signal characteristics: MGT_FDED is based on the vertical guide, which is sensitive to weak signals in the deep area but smooths out shallow details; MGT_NRED is based on the original field components, which preserves high-frequency details but has a weak response to deep signals.

[0108] 2. Differences in noise processing: MGT_FDED suppresses surface noise through vertical guide, but amplifies deep noise; MGT_NRED processes the raw data directly, resulting in different noise sensitivities.

[0109] 3. Inconsistent feature space: The boundary responses output by the two detectors are different in terms of dimensions, dynamic range, and spatial distribution.

[0110] 4. Optimization target conflict: MGT_FDED pursues deep signal enhancement, while MGT_NRED pursues surface detail preservation, resulting in a fundamental optimization conflict.

[0111] Therefore, simple weighted fusion can lead to blurred boundaries and a decrease in signal-to-noise ratio. The most suitable detector must be selected based on the physical characteristics of the defect.

[0112] like Figures 3-4 As shown, to verify the results, the identified defect areas can be confirmed using ultrasonic testing (for re-inspection of internal buried holes) or magnetic particle testing (for re-inspection of surface cracks). Experiments show that, after adopting this invention, the boundary positioning error of deep defects can be less than 1 mm, and the boundary positioning error of shallow defects can be less than 0.5 mm.

[0113] In this embodiment, for a 1mm surface crack region, the alpha_map value is close to 0, and the output result is mainly contributed by NRED, thus preserving clear crack details. For a 6mm deep buried hole region, the alpha_map value is approximately (6-3) / 5=0.6, and the output result is dominated by FDED (accounting for 60%) and assisted by NRED (accounting for 40%), effectively enhancing the deep signal while preserving some structural features.

[0114] Example 2

[0115] like Figure 5 As shown, a defect identification system for ferromagnetic materials based on magnetic gradient tensor measurement is used to perform the above method, including:

[0116] Data acquisition module: Uses an AMR magnetic gradient sensor array to acquire magnetic gradient tensor data of the device under test.

[0117] Data processing module: Calculates the total horizontal derivative based on the magnetic gradient tensor data;

[0118] Calculate the partial derivatives of each component in the z-direction based on the total horizontal derivative and calculate the vertical derivative based on the magnetic gradient tensor data;

[0119] Defect boundaries are identified using a magnetic gradient tensor fusion derivative detector and a magnetic gradient tensor noise-resistant boundary detector.

[0120] Output module: Image the final adaptive boundary recognition results to provide inspectors with a clear diagram of the defect outline.

[0121] This invention discloses a ferromagnetic material defect identification system based on magnetic gradient tensor measurement, which can be installed in a computer device. The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a ferromagnetic material defect identification program based on magnetic gradient tensor measurement. The memory includes at least one type of readable storage medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. The processor is the control core of the electronic device, connecting various components of the computer device via various interfaces and lines. It executes programs or modules stored in the memory and calls data stored in the memory to perform various functions and process data.

[0122] The module described in this invention refers to a series of computer program segments that can be executed by the processor of a computer device and can perform a fixed function, and which are stored in the memory of the computer device.

[0123] The system provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0124] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0125] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall fall within the scope of the present invention.

Claims

1. A method for defect identification in ferromagnetic materials based on magnetic gradient tensor measurement, characterized in that, Includes the following steps: Acquire magnetic gradient tensor data of the test piece: Scan the surface of the test piece with a magnetic gradient sensor array to obtain the magnetic induction intensity components Bx, By, and Bz, and calculate the nine components of the magnetic gradient tensor: Gxx, Gxy, Gxz, Gyx, Gyy, Gyz, Gzx, Gzy, and Gzz. Calculate the total horizontal derivative based on the magnetic gradient tensor data: ； ； ； In the formula, , , These are the components of the magnetic flux density vector in the x, y, and z directions, respectively. These are the components of the gradient tensor matrix; Let x be the x-component of the total horizontal derivative of the magnetic gradient tensor; Let y be the y-component of the total horizontal derivative of the magnetic gradient tensor; Let z be the z-component of the total horizontal derivative of the magnetic gradient tensor; The partial derivatives of each component in the z-direction of the total horizontal derivative are calculated based on the total horizontal derivative, and the vertical derivative is calculated based on the magnetic gradient tensor data; the expression for calculating the partial derivatives of each component in the z-direction of the total horizontal derivative is as follows: ； ； ； The expression for calculating the vertical derivative is: ； ； ； Defect boundaries are identified using a magnetic gradient tensor fusion derivative detector and a magnetic gradient tensor noise-resistant boundary detector, and a defect contour map is output. The magnetic gradient tensor fusion derivative detector is: ； The magnetic gradient tensor noise-resistant boundary detector is: ； In the formula, It is a constant; Based on the defect type, a lookup table is used to match the correspondence between different types of defects and the magnetic gradient tensor fusion derivative detector or the magnetic gradient tensor noise-resistant boundary detector.

2. The method for defect identification of ferromagnetic materials based on magnetic gradient tensor measurement according to claim 1, characterized in that, Surface cracks, dense pores, or pitting defects are identified using the magnetic gradient tensor noise-resistant boundary detector; deep buried holes, stress cracks, or large-area corrosion are identified using the magnetic gradient tensor fusion derivative detector.

3. A defect identification system for ferromagnetic materials based on magnetic gradient tensor measurement, used to perform the method as described in any one of claims 1-2, characterized in that, include: Data acquisition module: used to acquire magnetic gradient tensor data of the test piece; Data processing module: Calculates the total horizontal derivative based on the magnetic gradient tensor data; Calculate the partial derivatives of each component in the z-direction based on the total horizontal derivative and calculate the vertical derivative based on the magnetic gradient tensor data; Defect boundaries are identified using a magnetic gradient tensor fusion derivative detector and a magnetic gradient tensor noise-resistant boundary detector. Output module: Displays the processing results of the data processing module.

Images