阅读说明：本技术 一种通过磁滞回线进行材料磁学性能的测定方法 (Method for measuring magnetic performance of material through hysteresis loop ) 是由 唐笑年 易维云 苏清萍 李竹 邓子豪 张昊雯 吴豪 王崇柏 王赞昆 朱柏荣 戴奇 于 2021-08-19 设计创作，主要内容包括：本发明涉及一种通过磁滞回线进行材料磁学性能的测定方法,包括以下步骤：测定之前测量并记录一定时间内的空载数据,对测定装置进行标定,开始正式实验,得到该待测试样相应的磁滞回线数据,通过图像处理方法将被测图像与库图像进行识别对比,从而分析出被测材料的磁学性能。本发明提供的检测方法主要分析过程依靠计算机的分析算法,成本较低,所采用的检测装置结构简单,造价低,操作方便；本发明可以对不完整的曲线进行分析,同时本发明针对试样的磁滞回线不仅能分析它的各种性能参数,还会根据这些参数对物质的材料进行判断,判断所测物质中是否含有杂质,测试速度快,效率高,准确率高。(The invention relates to a method for measuring the magnetic performance of a material through a hysteresis loop, which comprises the following steps: measuring and recording no-load data within a certain time before measurement, calibrating a measuring device, starting a formal experiment to obtain the corresponding hysteresis loop data of the sample to be measured, and identifying and comparing the measured image with the library image by an image processing method so as to analyze the magnetic performance of the measured material. The detection method provided by the invention mainly depends on the analysis algorithm of a computer in the analysis process, the cost is lower, and the adopted detection device has the advantages of simple structure, low manufacturing cost and convenient operation; the invention can analyze the incomplete curve, and simultaneously, the invention can not only analyze various performance parameters of the sample aiming at the hysteresis loop of the sample, but also judge the material of the substance according to the parameters to judge whether the substance contains impurities, and has the advantages of high testing speed, high efficiency and high accuracy.)

1. A method for measuring the magnetic performance of a material through a hysteresis loop is characterized in that: the method comprises the following steps:

before a sample to be measured is measured, a measuring device is started, the sample to be measured is not placed, no-load data within a certain time is measured and recorded, and errors generated by experimental data are eliminated in a formal experiment;

calibrating a measuring device, testing by using an iron standard sample, comparing and analyzing data obtained by theory and experiment to obtain a corresponding correction coefficient, correcting the data of the measuring device, and using an electromagnet to perform the following magnetic force equation of the sample:

wherein: f is the magnitude of the magnetic force applied to the sample, μ₀Is a vacuum magnetic conductivity; s₀Is the area of the sample measured in m²(ii) a N is the number of turns of the magnet exciting coil of the electromagnet, and I is exciting current passing through the magnet exciting coil, and the unit is A; l is the air gap length;

after the measuring device is calibrated, the detected surface of the sample to be measured is opposite to the electromagnet of the measuring device and is placed on a sample support of the measuring device, the current passing through the electromagnet is zero, a formal experiment is started, the current input into the electromagnet is controlled to change gradually from small to large, the magnetic field intensity of the sample to be measured and corresponding change data measured by the force sensor are recorded, when the data reach a peak value, the current is controlled to be changed from large to small, and the like, and hysteresis loop data corresponding to the sample to be measured are obtained; the relationship between the magnetic field intensity and the current passing through the electromagnet is as follows:

in the formula: h is the magnetic field intensity with the unit of A/m; n is the number of turns of the magnet exciting coil of the electromagnet; i is electromagnet exciting current, and the unit is A; le is the effective magnetic path length of the sample to be measured, and the unit is m;

the real magnetic force F is the force F' measured by the force sensor to reduce the self gravity G of the object₀And obtaining that as the noise points are more in the measurement process of the force sensor, the sliding weighted average filtering processing is carried out, and the formula is as follows:

the step size of the sliding window is set to 3, w₁、w₂、w₃As a weight, a more real magnetic force value is obtained after denoising;

by the formula:

B＝μ₀(M+H)

converting the magnetic force F into magnetization M, changing the value H by changing the current value I, and drawing an experimental curve M-H corresponding to different magnetization M values; in the formula: magnetic induction in the air gap between the sample to be tested and the electromagnet, A₀The sectional area of the air gap is obtained when the measuring device is calibrated, and xi is a correction coefficient; k is a constant;

if M-H is a straight line, the non-ferromagnetic object can be judged, and subsequent calculation is not needed; if M-H is a curve, further matching:

adopting opencv in computer vision, taking an experimental curve as an image kernel to be detected, taking known hysteresis loops of various materials with standard sizes as a template, and matching the kernel and windows of the template one by one; the method comprises the following concrete steps:

after an M-H image of a sample to be detected is drawn, graying the image, and setting a threshold value for binarization processing; then drawing an outer contour by using cvFindCountours (), then using cvBoundingget () to circumscribe a rectangular boundary to obtain a shape value, and using the shape value as the size of a window matrix; the template processing is carried out graying and binaryzation in advance, and a window frame is used for extracting an area; the original image starts to slide on the template as a kernel, the similarity degree of the kernel and the position of the template covered by the image is calculated, the similarity detection is carried out, and the similarity detection adopts a normalized correlation coefficient:

in the formula:

r (x, y) is a normalized correlation coefficient of a window with a point (x, y) as a center, a numerator is the inner product sum of a relative value of the template to the mean value and a relative value of the image to the mean value, a denominator is normalization processing, and a correlation value domain is controlled to be [ -1, 1]In addition, the more relevant the correlation coefficient is closer to 1, the higher the similarity degree is; the values of the correlation coefficients of the windows form a new matrix T^*Obtaining T by using minMaxloc () function^*And the respective corresponding window center value positions;

the window corresponding to the central value (x, y) is the area to be found in the template T, and the material category of the sample to be tested is obtained by checking the characteristic curve of which material the whole curve corresponding to the area belongs to; from the start and end-most slopes of the window, according to the formula:

M＝χ_mH

the magnetic susceptibility of the object before detection and the magnetic susceptibility of the object after detection are obtained.

2. A method for determining the magnetic properties of a material through a hysteresis loop as claimed in claim 1, wherein: the procedure for calibrating the assay device is as follows:

the iron standard sample is placed on the sample support, the computer sends a signal, so that the current flowing through the electromagnet is gradually increased from zero until saturation, then is decreased from large to small and reversely reaches saturation, and by analogy, the hysteresis loop curve of the iron standard sample is recorded.

3. A method for determining the magnetic properties of a material through a hysteresis loop as claimed in claim 1, wherein: the measuring device comprises a base, an electromagnet, a support column, a parallel beam type force sensor, a sample support and a control system, wherein the electromagnet is arranged at one end of the base, and the support column is arranged at the other end of the base; one end of the parallel beam type force sensor is connected with the top end of the support column, the other end of the parallel beam type force sensor is connected with the sample support, and the sample support is positioned above the electromagnet; the control system is arranged on the support column and comprises a controller, a switch, a signal transmission module and a power supply module, the controller is arranged in the support column and is respectively connected with the electromagnet and the parallel beam type force sensor, and the switch, the signal transmission module and the power supply module are arranged on the surface of the support column and are respectively connected with the controller; the signal transmission module is connected with the computer, and the power supply module is connected with the power supply.

Technical Field

The invention relates to a method for measuring magnetic performance, in particular to a method for measuring the magnetic performance of a material through a hysteresis loop.

Background

At present, the models of the hysteresis loop common measuring instruments comprise an AMH-4020 hysteresis loop measuring instrument, a DH4516 hysteresis loop tester and the like, the instruments are generally complex, the operation is not visual and simple enough, the number of steps needing manual operation is large, the interference is eliminated by using a complex experimental device, the braking degree is low, the testing speed is low, the accuracy is low, and the popularization degree is low. At present, the method makes a major breakthrough in the fields of computer technology and experimental data processing, can optimize the determination method of the magnetic performance of the material by further processing data in combination with the related technology instead of eliminating interference by using a complex experimental device, and thus, a more portable device is used for detection.

Disclosure of Invention

In order to solve the above technical problems, the present invention provides a method for measuring the magnetic properties of a material through a hysteresis loop, comprising the steps of:

before a sample to be measured is measured, a measuring device is started, the sample to be measured is not placed, no-load data within a certain time is measured and recorded, and errors generated by experimental data are eliminated in a formal experiment;

calibrating a measuring device, testing by using an iron standard sample, comparing and analyzing data obtained by theory and experiment to obtain a corresponding correction coefficient, correcting the data of the measuring device, and using an electromagnet to perform the following magnetic force equation of the sample:

wherein: f is the magnitude of the magnetic force applied to the sample, μ₀Is a vacuum magnetic conductivity; s₀Is the area of the sample measured in m²(ii) a N is the number of turns of the magnet exciting coil of the electromagnet, and I is exciting current passing through the magnet exciting coil, and the unit is A; l is the air gap length;

after the measuring device is calibrated, the detected surface of the sample to be measured is opposite to the electromagnet of the measuring device and is placed on a sample support of the measuring device, the current passing through the electromagnet is zero, a formal experiment is started, the current input into the electromagnet is controlled to change gradually from small to large, the magnetic field intensity of the sample to be measured and corresponding change data measured by the force sensor are recorded, when the data reach a peak value, the current is controlled to be changed from large to small, and the like, and hysteresis loop data corresponding to the sample to be measured are obtained; the relationship between the magnetic field intensity and the current passing through the electromagnet is as follows:

in the formula: h is the magnetic field intensity with the unit of A/m; n is the number of turns of the magnet exciting coil of the electromagnet; i is electromagnet exciting current, and the unit is A; le is the effective magnetic path length of the sample to be measured, and the unit is m;

the force F' measured by the force sensor is firstly subtracted from the gravity G of the object₀And obtaining a real magnetic force F, wherein the sliding weighted average filtering processing is carried out because more noise points exist in the measuring process of the force sensor, and the formula is as follows:

the step length of the sliding window is set to be 3, and the weight is set to be w₁＝0.25,w₂＝0.5，w₃0.25. After denoising, obtaining a more real magnetic force value;

by the formula:

B＝μ₀(M+H)

converting F into magnetization M, changing H value by changing current value I, and drawing an experimental curve M-H corresponding to different magnetization M values; in the formula: b is the magnetic induction in the air gap between the sample to be tested and the electromagnet, A₀Is the sectional area of the air gap, xi is the correction coefficient, obtained by calibrating the measuring device; k is a constant;

if M-H is a straight line, the non-ferromagnetic object can be judged, and subsequent calculation is not needed; if M-H is a curve, further matching:

adopting opencv in computer vision, taking an experimental curve as an image kernel to be detected, taking known hysteresis loops of various materials with standard sizes as a template, and matching the kernel and windows of the template one by one; the method comprises the following concrete steps:

after an M-H image of a sample to be detected is drawn, graying the image, and setting a threshold value for binarization processing to ensure that the image is not black or white and has clear outline;

then drawing an outer contour by using cvFindCountours (), then using cvBoundingget () to circumscribe a rectangular boundary to obtain a shape value, and using the shape value as the size of a window matrix; the template processing is carried out graying and binaryzation in advance, and a window frame is used for extracting an area; the original image starts to slide on the template as a kernel, the similarity degree of the kernel and the position of the template covered by the image is calculated, and the similarity detection is carried out by using a cv2.TM _ CCOEFF _ NORMED function, wherein the similarity detection adopts a normalized correlation coefficient:

in the formula:

r (x, y) is a normalized correlation coefficient of a window with a point (x, y) as a center, a numerator is the inner product sum of a relative value of the template to the mean value and a relative value of the image to the mean value, a denominator is normalization processing, and the more relevant the correlation coefficient is, the higher the similarity degree is; the values of the correlation coefficients of the windows form a new matrix T^*Obtaining T by using minMaxloc () function^*And the respective corresponding window center value positions;

the window corresponding to the central value (x, y) is the area to be found in the template T, and the material category can be known by checking the characteristic curve of which material the whole curve corresponding to the area belongs to; from the slope of the starting point and the end of the window, according to the formula

M＝x_mH

The magnetic susceptibility of the object before detection and the magnetic susceptibility of the object after detection are obtained.

The procedure for calibrating the assay device is as follows:

the iron standard sample is placed on the sample support, the computer sends a signal, so that the current flowing through the electromagnet is gradually increased from zero until saturation, then is decreased from large to small and reversely reaches saturation, and by analogy, the hysteresis loop curve of the iron standard sample is recorded.

The measuring device comprises a base, an electromagnet, a support column, a parallel beam type force sensor, a sample support and a control system, wherein the electromagnet is arranged at one end of the base, and the support column is arranged at the other end of the base; one end of the parallel beam type force sensor is connected with the top end of the support column, the other end of the parallel beam type force sensor is connected with the sample support, and the sample support is positioned above the electromagnet; the control system is arranged on the support column and comprises a controller, a switch, a signal transmission module and a power supply module, the controller is arranged in the support column and is respectively connected with the electromagnet and the parallel beam type force sensor, and the switch, the signal transmission module and the power supply module are arranged on the surface of the support column and are respectively connected with the controller; the signal transmission module is connected with the computer, and the power supply module is connected with the power supply. The switch control device is turned on or turned off, the controller adjusts the current flowing through the electromagnet according to the computer instruction, meanwhile, data signals of the parallel beam type force sensor are collected, and the signal transmission module connects the controller with the computer for data interactive transmission.

The invention has the beneficial effects that:

the detection method provided by the invention mainly depends on the analysis algorithm of a computer in the analysis process, the cost is lower, and the adopted detection device has the advantages of simple structure, low manufacturing cost and convenient operation; different from the conventional method for measuring the hysteresis loop, the method can also analyze the incomplete curve if the electromagnet cannot reach the external magnetic field required by the saturation magnetization of the material. Meanwhile, the invention can analyze various performance parameters of the hysteresis loop of the sample, and can judge the material of the substance according to the parameters to judge whether the measured substance contains impurities, and the test speed is high, the efficiency is high, and the accuracy is high.

Drawings

FIG. 1 is a schematic view of the structure of an assay device according to the present invention;

FIG. 2 is a diagram illustrating the effect of three noise filtering modes according to an embodiment of the present invention;

1. base 2, electro-magnet 3, support column 4, parallel beam formula force transducer

5. The device comprises a sample support 6, a switch 7, a signal transmission module 8 and a power supply module.

Detailed Description

Please refer to fig. 1:

the invention provides a method for measuring the magnetic performance of a material through a hysteresis loop, which comprises the following steps:

before a sample to be measured is measured, a measuring device is started, the sample to be measured is not placed, no-load data within a certain time is measured and recorded, and errors generated by experimental data are eliminated in a formal experiment;

calibrating a measuring device, testing by using an iron standard sample, placing the iron standard sample on a sample support, sending a signal by a computer, so that the current flowing through an electromagnet is gradually increased from zero until saturation, then is decreased from large to small, and reversely reaches saturation, and by analogy, recording a hysteresis loop curve of the iron standard sample; after comparing and analyzing data obtained by theory and experiment, obtaining corresponding correction coefficient, correcting data of the measuring device, and making the magnetic force equation of the electromagnet on the sample as follows:

wherein: f is the magnitude of the magnetic force applied to the sample, μ₀Is a vacuum magnetic conductivity; s₀Is the area of the sample measured in m²(ii) a N is the number of turns of the magnet exciting coil of the electromagnet, I is the exciting current passed by the magnet coil, and the unit is A; l is the air gap length;

after the measuring device is calibrated, the detected surface of the sample to be measured is opposite to the electromagnet of the measuring device and is placed on a sample support of the measuring device, the current passing through the electromagnet is zero, a formal experiment is started, the current input into the electromagnet is controlled by a control system to gradually change from small to large, the magnetic field intensity of the sample to be measured and corresponding change data measured by a force sensor are recorded, when the data reach a peak value, the current is controlled to be reduced from large to small, and the like, and hysteresis loop data corresponding to the sample to be measured are obtained; the relationship between the magnetic field intensity and the current passing through the electromagnet is as follows:

in the formula: h is the magnetic field intensity with the unit of A/m; n is the number of turns of the magnet exciting coil of the electromagnet; i is electromagnet exciting current, and the unit is A; le is the effective magnetic path length of the sample to be measured, and the unit is m;

the core idea of the invention is to convert the measured F-H curve into an M-H curve, namely a hysteresis loop measured by experiments, to use the known hysteresis loops of various material standard sizes as templates, to perform similarity detection by using correlation coefficient normalization, to perform template matching, and to match a window capable of minimizing the cost function in the templates. Therefore, whether the object to be detected is a ferromagnetic object or not is detected, and if the object to be detected is a ferromagnetic object, which material is further determined, the magnetization intensity of the object is determined.

Since the most essential feature of a ferromagnetic object is that the permeability μ varies with the applied field strength H, the permeability μ of a non-ferromagnetic object is constant. By the following formula:

B＝μ₀(M+H) (1)

M＝χ_mH (2)

μ_r＝1+χ_m (3)

B＝μ₀(1+χ_m)H＝μ₀μ_rH＝μH (4)

where B is the magnetic induction, M is the magnetization, H is the magnetic field strength, χ_mIs magnetic susceptibility, mu₀Is a vacuum permeability, mu_rAs the relative permeability, μ is the permeability. It can be concluded that the hysteresis loop of a ferromagnetic object essentially has a different tendency to increase or decrease the susceptibility at different magnetic field strengths H. And objects made of different materials have different increasing or decreasing tendencies of magnetic susceptibility under the same magnetic field intensity H. I.e. the variation trend of the slope of the M-H image, and the ability to distinguish different materials.

The parallel beam type force sensor has the advantages that the force sensing is sensitive, and the measured force F' should be firstly reduced to reduce the self gravity G of an object₀And obtaining the real magnetic force F, wherein the filtering processing is needed because the noise points are more in the measuring process of the sensor. The effect of the three noise filtering modes of the moving median, the moving average and the moving weighted average is shown in fig. 2, and it can be seen that the filtering effect of the moving weighted average is the best.

Here, a moving weighted average filtering process is performed as in equation (5)

The step length of the sliding window is set to be 3, and the weight is set to be w₁＝0.25,w₂＝0.5，w₃0.25. And after denoising, obtaining a more real magnetic force value. Is given by the formula (6)

B is the magnetic induction in the air gap between the sample to be tested and the electromagnet, A₀The cross-sectional area of the air gap is the size of the sample to be tested, and the device has limited the size of the sample to be tested, and in the embodiment, the cross-sectional area is 1cm × 1mm, S ═ 1c square meter, and ξ is a correction coefficient. k is a constant value and k is a constant value,that is to saySince k is greatly influenced by the environment, when the iron standard sample is used for calibration, the experimental value and the theoretical value are compared to obtain a correction coefficient xi. F is converted into magnetization M by the formulas (1) and (6), while the value of H can be changed by changing the current value I in the experiment, and an experiment curve M-H can be drawn corresponding to different values of magnetization M.

If M-H is a straight line, the non-ferromagnetic object can be judged, and subsequent calculation is not needed; if M-H is a curve, further matching;

since the permeability μ before detection, that is, the magnetization M after conversion cannot be determined by the sample to be measured, the point where the values of the experimental graph and the template graph H are the same cannot be simply regarded as the same point. And the magnetizing saturation point of the object can not be reached because the magnetic energy can not be ensured to be added to the object. It is decided that the invention does not follow a complete hysteresis loop but a certain section of a hysteresis loop wire.

Dynamic linearization is not accurate enough, and a large number of learning samples are needed for support vector machines and deep learning. Therefore, the opencv in computer vision is adopted, the experimental curve is used as the image kernel to be tested, the known hysteresis loops of various materials and standard sizes are used as the template, and the kernel is matched with the windows of the template one by one; the method comprises the following concrete steps:

after an M-H image of a sample to be detected is drawn, graying the image to ensure that each pixel point of the image outputs not three rgb channel values but a gray value, and then setting a threshold value to carry out binarization processing to ensure that the image is 'non-black, namely white' and has clear outline;

then drawing an outer contour by using cvFindCountours (), then using cvBoundingget () to circumscribe a rectangular boundary to obtain a shape value, and using the shape value as the size of a window matrix; the template processing is carried out graying and binaryzation in advance, and a window frame is used for extracting an area; the template matching and convolution principles are similar, an original image starts to slide on a template as a kernel, the similarity degree of the kernel and the position of the template covered by the kernel is calculated, and the similarity degree is detected by using a cv2.TM _ CCOEFF _ NORMED function, wherein the similarity degree detection adopts a normalized correlation coefficient:

in the formula:

r (x, y) is a normalized correlation coefficient of a window with a point (x, y) as a center, a numerator is the inner product sum of a relative value of the template to the mean value and a relative value of the image to the mean value, a denominator is normalization processing, and the more relevant the correlation coefficient is, the higher the similarity degree is; the values of the correlation coefficients of the windows form a new matrix T^*Obtaining T by using minMaxloc () function^*And the respective corresponding window center value positions (x, y);

the window corresponding to the central value (x, y) is the area to be found in the template T, and the material category can be known by checking the characteristic curve of which material the whole curve corresponding to the area belongs to; and (3) solving the magnetic susceptibility of the object before detection and the magnetic susceptibility of the object after detection according to a formula (2) through the slope of the starting point and the tail end of the window, so as to perform proper demagnetization or magnetic compensation according to the existing magnetic susceptibility and the actual requirement of the object.

The magnetic field change design can be carried out on a computer according to the magnetic field environment of the sample in practical application, the magnetic field change along with time can be met by adjusting the current, the magnetic force applied to the sample in practical application, the magnetic performance of each time period in practical application and the like can be simulated.

The measuring device comprises a base 1, an electromagnet 2, a support column 3, a parallel beam type force sensor 4, a sample support 5 and a control system, wherein the electromagnet 2 is arranged at one end of the base 1, and the support column 3 is arranged at the other end of the base 1; one end of the parallel beam type force sensor 4 is connected with the top end of the support column 3, the other end of the parallel beam type force sensor is connected with the sample bracket 5, and the sample bracket 5 is positioned above the electromagnet 2; the control system is arranged on the support column 3 and comprises a controller, a switch 6, a signal transmission module 7 and a power supply module 8, the controller is arranged in the support column 3 and is respectively connected with the electromagnet 2 and the parallel beam type force sensor 4, and the switch 6, the signal transmission module 7 and the power supply module 8 are arranged on the surface of the support column 3 and are respectively connected with the controller; the signal transmission module 7 is connected with a computer, and the power supply module 8 is connected with a power supply. The switch 6 controls the device to be turned on or off, the controller adjusts the current flowing through the electromagnet 2 according to the computer instruction, meanwhile, the data signal of the parallel beam type force sensor 4 is collected, and the signal transmission module 7 connects the controller with the computer for data interactive transmission.