Analysis method for internal correlation of magnetic Barkhausen noise and magnetic domain motion characteristics

Abstract

The invention discloses a magnetic Barkhausen noise and magnetic domain motion characteristic internal correlation analysis method, and relates to the technical field of micro-magnetic nondestructive testing. And reconstructing the magnetic domain motion velocity field matrix at each moment through the magnetic domain motion image sequence frame, and obtaining the magnetic domain motion velocity at the moment. A static detection and dynamic scanning dual-mode strategy is provided, and internal correlation of MBN and magnetic domain motion characteristics under two scales of time and space is analyzed respectively. In a static detection mode, a magnetic domain motion image sequence frame and an MBN signal which change along with excitation at the same position are analyzed to obtain a change curve of characteristic parameters of the two signals along with time, correlation of the two curves is compared, and influences of excitation parameters or stress and the like on the correlation of the two curves are quantified; in a dynamic scanning mode, a spatial distribution diagram of MBN characteristic parameters and magnetic domain motion characteristic parameters is generated through step-by-step surface scanning, correlation analysis is established, and influences of microstructures and the like on MBN signals are quantified.

Description

Technical Field

[0001] This application relates to the technical field of micro-magnetic non-destructive testing, and specifically relates to an experimental method for revealing the internal relationship between the two by jointly analyzing the magnetic domain motion velocity characteristics and the Magnetic Barkhausen noise (MBN) signal. Background Art

[0002] Ferromagnetic materials are widely used in industries such as rail transit, aerospace, wind power generation, and construction machinery due to their good thermal stability, strong corrosion resistance, etc. The Magnetic Barkhausen Noise (MBN) technology is a new type of non-destructive testing technology. Compared with conventional non-destructive testing technologies, the MBN technology has the advantages of fast detection speed, non-contact, high sensitivity, etc., and can realize real-time monitoring and quantitative characterization of the microstructure and mechanical properties of ferromagnetic materials.

[0003] Magnetic domains are tiny magnetization regions formed in ferromagnetic materials during the magnetization process to reduce the magnetostatic energy. And MBN is speculated to be the transient magnetic signal generated by the discontinuous movement of magnetic domain walls in ferromagnetic materials, which is closely related to the motion characteristics of magnetic domains. However, at present, there are insufficient correlation analysis methods between the magnetic domain motion characteristics and the macroscopic MBN signal, resulting in unclear MBN detection models and influencing factors, which limit the detection accuracy and application range of MBN. On the other hand, it also leads to the fact that the current engineering applications of MBN can only adopt the method of pre-experimental calibration. For example, to measure the Young's modulus parameter of a material, a series of calibration specimens with different Young's moduli need to be prepared in advance, and the whole process is time-consuming and laborious. Summary of the Invention

[0004] Based on the above, the purpose of the present invention is to provide a correlation analysis method for magnetic domain motion characteristics and MBN signals, laying a foundation for the interpretation of the MBN signal detection mechanism and better engineering applications of MBN.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] An experimental method for the internal relationship between magnetic domain motion characteristics and MBN signals, characterized in that: first, reconstruct the magnetic domain motion velocity field matrix through the magnetic domain motion image sequence frames, and obtain the magnetic domain motion velocity value representing this region from it. Then, under the dual-mode strategies of static detection and dynamic scanning, analyze the internal relationship between MBN and magnetic domain motion characteristics at two scales of time and space. Finally, use these relationships to analyze the influence of excitation parameters, microstructure, and stress on the magnetic domain motion velocity and MBN signal.

[0007] The method for extracting the magnetic domain motion velocity specifically includes the following steps:

[0008] (1) Under certain external magnetic field and stress conditions, use a magneto-optical Kerr microscope to collect a sequence of magnetic domain motion images in the target area. Configure the frame rate of the camera according to the excitation frequency to ensure that sufficient transient behaviors of magnetic domain wall displacements within the excitation period are captured.

[0009] (2) Arrange the sequence of magnetic domain images under the current excitation conditions at equal time intervals along the time axis. Starting from the first frame, input two adjacent frames into the optical flow algorithm for calculation to obtain the magnetic domain wall velocity field matrix between every two frames of images.

[0010] (3) Then, by analyzing the velocity field matrix, obtain the magnetic domain motion velocity value of the current image, which can be obtained by summing the velocity norms of all pixel points in the image and taking the average, or by first statistically analyzing the probability distribution curve of the velocity amplitudes of all pixel points in the region and then extracting the velocity corresponding to the peak of the curve.

[0011] Furthermore, the present invention proposes a method for time-correlation analysis of magnetic domain motion and MBN signals in a static detection mode. The specific scheme is as follows:

[0012] (1) Select a detection area. Under fixed different excitation parameters and stress conditions, obtain a sequence of magnetic domain motion images and MBN signals in the same area within multiple excitation periods.

[0013] (2) Plot the curves of magnetic domain motion velocity and MBN envelope changing with time, from which the characteristic parameters of both (including curve peak value, curve mean value, curve full width at half maximum, etc.) can be obtained and extracted.

[0014] (3) Analyze the correlation between the two curves through Pearson correlation coefficient, Spearman rank correlation coefficient, etc.

[0015] (4) Change the excitation parameters and stress conditions, and analyze the variation rules of the characteristic parameters of both with the change of excitation parameters or stress.

[0016] Furthermore, the present invention proposes a method for spatial-correlation analysis of magnetic domain motion and MBN signals in a dynamic scanning mode. The specific scheme is as follows:

[0017] (1) Select a certain area of the material. Under constant excitation magnetic field and stress parameters, perform a step-by-step surface scan, and synchronously obtain MBN signals and magnetic domain motion image signals within each sub-region.

[0018] (2) Extract the characteristic parameters of the two signals and generate spatial distribution maps of the characteristic parameters respectively.

[0019] (3) Perform correlation analysis on the two spatial distribution maps through hash algorithms, structural similarity, etc.

[0020] (4) Analyze the influence of changes in microstructure or mechanical properties on the MBN signal.

[0021] The beneficial effects of the present invention are as follows:

[0022] The present invention proposes a spatio-temporal correlation analysis method for MBN and magnetic domain motion characteristics, which can quantitatively study the influence of microstructure, excitation parameters, and stress on the MBN signal, contribute to the explanation of the MBN signal detection mechanism, and broader engineering applications of MBN. Description of the Drawings

[0023] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0024] Figure 1 Schematic diagram for extracting the magnetic domain motion speed and its variation curve with time;

[0025] Figure 2 Schematic diagram of the time correlation analysis method for MBN signal and magnetic domain motion speed in the static detection mode;

[0026] Figure 3 Schematic diagram of the spatial correlation analysis method for MBN signal and magnetic domain motion speed in the dynamic scanning mode. Specific implementation method

[0028] The following further describes the present invention in combination with the drawings and the above embodiments. The specific implementation examples provided below are only descriptive and not restrictive, and cannot be used to limit the protection scope of the present invention.

[0029] The specific implementation of the steps for extracting the magnetic domain motion speed and its variation curve with time can be as Figure 1 shown:

[0030] (1) Under certain external magnetic field and stress conditions, use a magneto-optical Kerr microscope to collect a sequence of magnetic domain motion images of the target area;

[0031] (2) Use the optical flow algorithm to reconstruct the velocity field of magnetic domain motion. Arrange the sequence of magnetic domain images under the current excitation conditions at equal time intervals along the time axis. Starting from the first frame, input two adjacent frames into the optical flow algorithm for calculation to obtain the magnetic domain velocity field matrix between every two frames of images;

[0032] (3) By analyzing the velocity field matrix, obtain the magnetic domain motion speed value of the current image. It can be obtained by summing the velocity norms of all pixel points in the image and taking the average, or by first statistically analyzing the probability distribution curve of the velocity amplitudes of all pixel points in the region and then extracting the velocity corresponding to the peak of the curve;

[0033] (4) Plot the curve of the image motion speed changing with time.

[0034] In this embodiment, in the static detection mode, a certain area is selected to explore the time-correlation analysis method of magnetic domain motion and MBN signal. The operation steps are as Figure 2 shown as follows:

[0035] (1) Select a certain area inside the grain of grain-oriented silicon steel as the signal acquisition area. The excitation signal is a sine wave, the frequency is selected from 20 Hz to 50 Hz, the amplitude range is 0.5 V to 2 V, and the camera frame rate is set to 3500 fps / s. Measure with a magneto-optical Kerr microscope and MBN respectively, and collect the magnetic domain motion image signals and MBN signals for 5 excitation cycles;

[0036] (2) Process the MBN signal and the magnetic domain image respectively, and plot the curves of the MBN envelope and the magnetic domain motion speed changing with time. Among them, the original MBN signal can be filtered and smoothed by moving average to obtain the MBN signal envelope line within a single cycle. The curve of the magnetic domain motion changing with time can be extracted from the magnetic domain motion image sequence according to Figure 1 the method described;

[0037] (3) Analyze the correlation between the two curves through the Pearson correlation coefficient;

[0038] (4) Change the excitation parameters and analyze the variation law of their characteristic parameters with the change of the excitation parameters.

[0039] In this embodiment, in the dynamic scanning mode, the spatial correlation analysis method of magnetic domain motion and MBN signal is explored. The operation steps are as Figure 3 shown as follows:

[0040] (1) The excitation signal is a sine wave, the frequency is selected as 50 Hz, the amplitude is 2 V, the frame rate is set to 3500 fps / s. Select the area near the grain boundary of grain-oriented silicon steel as the signal measurement area. This area takes the grain boundary as the central boundary line. Under the constant magnetic field and stress parameters, a step-by-step surface scan is carried out with a step distance of 0.2 mm. Synchronously obtain the MBN signal and the magnetic domain motion image signal in each sub-area.

[0041] (2) Extract the characteristic parameters of the two signals and generate the spatial distribution maps of the characteristic parameters respectively;

[0042] (3) Carry out correlation analysis on the two spatial distribution maps through the hash algorithm, etc., and calculate the similarity degree. The larger the similarity degree value, the higher the dependence of the two signals;

[0043] (4) Analyze the influence of the grain boundary, a microscopic structure, on MBN.

Claims

1. A method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics, characterized in that: First, the magnetic domain motion velocity field matrix is ​​reconstructed through the magnetic domain motion image sequence frames, from which the magnetic domain motion velocity value representing the area is obtained. Under the dual-mode strategy of static detection and dynamic scanning, the intrinsic correlation between MBN and magnetic domain motion characteristics in time and space is analyzed. Finally, these correlations are used to analyze the influence of excitation parameters, microstructure and stress on magnetic domain motion velocity and MBN signal.

2. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 1, characterized in that: The method for extracting the magnetic domain movement speed specifically comprises the following steps: (1) Under certain external magnetic field and stress conditions, a magneto-optical Kerr microscope is used to collect a sequence of magnetic domain motion images in the target area. The frame rate of the camera is configured appropriately according to the excitation frequency to ensure that enough transient behaviors of magnetic domain wall displacement are captured during the excitation cycle. (2) Arranging the magnetic domain image sequence under the current excitation condition according to the time axis with equal time difference, starting from the first frame, inputting two adjacent frames into the optical flow algorithm for calculation, and obtaining the magnetic domain wall velocity field matrix between every two frames of images; (3) The velocity field matrix is ​​then analyzed to obtain the magnetic domain motion velocity value of the current image. This value is obtained by summing up the velocity modulus of all pixels in the image and taking the average, or by first calculating the velocity amplitude probability distribution curve of all pixels in the region and then extracting the velocity corresponding to the peak value of the curve.

3. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 1, characterized in that: In the static detection mode, the time correlation analysis method of magnetic domain movement and MBN signal specifically includes the following steps: (1) Select the detection area, fix different excitation parameters and stress conditions, and obtain the magnetic domain motion image sequence and MBN signal of the same area in multiple excitation cycles; (2) plotting the curves of the magnetic domain movement velocity and the MBN envelope changing with time, and obtaining and extracting the characteristic parameters of the two, including the curve peak value, curve mean value, and curve half-peak width; (3) Analyze the correlation between the two curves using the Pearson correlation coefficient and the Spearman rank correlation coefficient; (4) Change the excitation parameters and stress conditions, and analyze the change patterns of their characteristic parameters as the excitation parameters or stress conditions change.

4. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 1, characterized in that: In the dynamic scanning mode, the spatial correlation analysis method of magnetic domain motion and MBN signal specifically includes the following steps: (1) Select a certain area of ​​the material and perform step-by-step surface scanning under constant excitation magnetic field and stress parameters to synchronously obtain MBN signals and magnetic domain motion image signals in each sub-area; (2) extracting characteristic parameters of the two signals and generating spatial distribution diagrams of the characteristic parameters respectively; (3) Perform correlation analysis on the spatial distribution maps of the two through hash algorithm and structural similarity; (4) Analyze the impact of changes in microstructure or mechanical properties on MBN signals.

5. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 1, characterized in that: The method for extracting the magnetic domain movement velocity comprises the following steps: (1) Using a magneto-optical Kerr microscope to collect a sequence of images of the motion of magnetic domains of a material under the action of an excitation magnetic field; (2) The velocity field of magnetic domain motion between adjacent images is calculated frame by frame through the optical flow algorithm to generate a velocity field matrix that corresponds one-to-one to the pixels of the original image; (3) Extract the value of the magnetic domain movement velocity at that moment from the velocity field matrix.

6. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 5, characterized in that: The magnetic domain movement velocity value is obtained by calculating the sum of the velocity amplitudes of all pixels in the area.

7. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 5, characterized in that: The magnetic domain movement speed value is obtained by statistically analyzing the speed amplitude probability distribution curve of all pixel points in the area and then extracting the speed corresponding to the peak value of the curve.

8. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 5, characterized in that: The temporal correlation process between the magnetic domain motion characteristics and the MBN signal includes the following steps: (1) Select a measurement area of ​​the specimen under constant excitation field parameters, including frequency, amplitude, and direction; (2) respectively collecting magnetic domain motion image sequences and MBN signals of the selected area; (3) Extracting the MBN signal envelope of the grain and grain boundary regions and the curve of the change of the magnetic domain movement speed over time, and extracting the characteristic parameters of the two from them, the characteristic parameters include the curve peak value, curve mean value, and curve half-peak width; (4) Multiple methods are used to analyze the temporal correlation between the two. The correlation between the two curves is analyzed by Pearson correlation coefficient and Spearman rank correlation coefficient, and the correlation between the characteristic parameters of the two is analyzed.

9. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 5, characterized in that: By changing different excitation field parameters and stress, the influence of excitation field parameters or stress on the correlation between magnetic domain motion characteristics and MBN signals is analyzed.

10. The method for analyzing the intrinsic correlation between magnetic Barkhausen noise and magnetic domain motion characteristics according to claim 8, characterized in that: The spatial correlation analysis method of magnetic domain motion and MBN signal is specifically implemented as follows: (1) Divide the test area of ​​the specimen into blocks; (2) acquiring MBN and magnetic domain motion image sequences in each sub-region; (3) Obtain the characteristic parameters of the MBN envelope and magnetic domain motion in each sub-region; (3) Draw the spatial distribution diagram of MBN signals and magnetic domain motion characteristic parameters; (4) The spatial distribution maps of the two are correlated and analyzed through hash algorithm and structural similarity. By correlating the signals of the two at different spatial locations, the influence of the microstructure or mechanical property mutation on MBN is analyzed.

Images