A dynamic-born eddy current sensor and a method for detecting defects of a metal component on line

Abstract

The application discloses a kind of dynamic eddy current sensors and metal component on-line defect detection methods, applied to nondestructive testing field, for the problem that dynamic eddy current has not been used in the detection of the surface defect of moving metal component;The application first proposes a kind of magnetic sensor, the sensing device includes a cuboid permanent magnet;Sensor is arranged on the side of permanent magnet;Clamp for fixing permanent magnet and sensor;The application also proposes a kind of detection method based on the structure sensor, quantitative nondestructive testing of the size of the surface defect of moving metal component can be realized, the magnetic sensor based on dynamic eddy current provided by the application can detect the size of the defect of moving metal component, and the monitoring of the surface defect of moving metal component can be realized.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of non-destructive testing, and particularly relates to a sensor and a non-destructive testing technology based on the same. BACKGROUND

[0002] With the rapid development of science and technology and industrial production, the requirements for material performance are getting higher and higher, but the current metallurgical technology cannot provide perfect materials. At the same time, various equipment may also produce defects in the manufacturing process, such as pores, slag inclusions, incomplete penetration in welding, shrinkage, pores in casting, white spots, folding in forging, etc. When the defective parts, especially high-speed parts, are used as load-bearing parts, it is often very dangerous, and sometimes even causes the entire mechanism to be damaged, resulting in major equipment and personal accidents and causing significant losses.

[0003] Non-destructive testing has formed a relatively complete theoretical system after decades of development. The main methods in the field of non-destructive testing at present include ultrasonic testing, laser testing and electromagnetic testing. The commonly used electromagnetic non-destructive testing technologies at present are eddy current non-destructive testing technology and magnetic flux leakage non-destructive testing technology. Whether it is eddy current non-destructive testing or magnetic flux leakage non-destructive testing, due to the existence of coils and permanent magnets, when the probe moves relative to the metal conductor, eddy current will be generated in the metal conductor. The eddy current generated by the movement interferes with the detection signal in the original electromagnetic non-destructive testing system, affecting the detection accuracy, but the eddy current generated by the movement is also sensitive to the discontinuous features in the metal conductor, and new characteristic signals can be generated to identify the discontinuous features.

[0004] Currently, some scholars have used eddy current generated by movement to conduct non-destructive testing on surface defects of metal components, but there are still many problems worth exploring. For example, the quantitative relationship between the geometric dimensions such as the depth and width of the surface defects and the eddy current detection signal; the relationship between such quantitative relationship and the movement speed; the influence of the movement acceleration at the beginning and end of the detection on the measurement results; when the metal conductor is a ferromagnetic material, the characteristic signals of the eddy current non-destructive testing generated by the movement, etc.

[0005] The related prior art is as follows:

[0006] A wheeled eddy current and magnetic flux leakage detection sensor and a detection method are provided, the sensor adopts a magnetic wheel design, an excitation coil capable of being magnetized in alternating current and direct current and having controllable strength is added, so that the sensor can pick up eddy current induced signals and magnetic flux leakage field signals at the same time, thereby improving the detection depth and surface sensitivity of the near surface of the detected steel plate. By analyzing the changes of the eddy current induced signals and the magnetic flux leakage field signals, the size, position and depth of the surface and near surface crack defects of the detected steel plate can be indirectly obtained. Further, the sensor adopts a multi-wheel triangular structure, and large-area rolling scanning detection is implemented on the detected steel plate, thereby greatly improving the detection efficiency.

[0007] Research on metal surface defect detection methods based on eddy current sensing addresses the problem of microcracks that are easily generated during the manufacturing and service life of engine turbine blades. Using the aviation aluminum alloy 6A02 as the research object, finite element modeling and parameter optimization of traditional eddy current sensing probes were performed. The sensing characteristics of the excitation coil coupled with the specific metal material being tested were analyzed. An eddy current probe for detecting microcracks based on a giant magnetoresistance (GMR) chip was designed, and a corresponding test system was developed. The influence of the probe scanning direction and the GMR chip's sensitive axis on the detection results was studied. A data processing method based on wavelet analysis for noise elimination and feature recognition was proposed, achieving good results.

[0008] A method and device for measuring the wall thickness of a metal tube based on motional eddy currents is disclosed. The method and device for measuring the wall thickness of a metal tube based on motional eddy currents detect the wall thickness of the metal tube based on the motional eddy current effect, and are particularly suitable for high-speed metal tube detection environments.

[0009] Research on the method of locating defects on the inner and outer walls of pipelines during high-speed magnetic flux leakage testing. A method for locating and distinguishing defects on the inner and outer walls of pipelines during high-speed magnetic flux leakage testing based on motional eddy currents is proposed. The defect positions are distinguished by the difference in the changing characteristics of the magnetic field signals on the inner and outer walls when the eddy current magnetic field is coupled with the external magnetic field.

[0010] Sub-surface defect detection with motion induced eddy currents inaluminiumA method for detecting sub-surface defects in non-ferromagnetic materials based on motion-induced eddy currents is proposed. A probe consisting of a permanent magnet and a sensor is moved near an aluminum plate with sub-surface defects to acquire data.

[0011] Numerical simulation on magnetic flux leakage evaluation at highspeed is proposed to characterize defect information using eddy current signals in a high-speed nondestructive inspection (NDI) system using magnetic flux leakage (MFL).

[0012] At present, motional eddy currents are widely used in auxiliary aspects of electromagnetic non-destructive testing, such as metal pipe wall thickness, inner and outer wall defect positioning, etc., but have not yet been directly used in the detection of surface defects of moving metal components. Summary of the Invention

[0013] To solve the above technical problems, the present invention proposes a sensor based on motional eddy current and a method for detecting defects in moving metal components, which uses a permanent magnet as a source of magnetic signals to detect the size of defects on the surface of moving metal components.

[0014] One of the technical solutions adopted by the present application is: a sensor based on dynamic-born eddy current, comprising: a cuboid permanent magnet, a magnetic sensor and a clamp; the magnetic sensor is arranged on one side of the cuboid permanent magnet, and the clamp is used for fixing the cuboid permanent magnet and the magnetic sensor.

[0015] The second technical solution adopted by the present application is: a method for detecting defects of a moving metal component based on the above-mentioned sensor based on dynamic-born eddy current, comprising:

[0016] S1, placing the sensor based on dynamic-born eddy current parallel to the surface of the moving metal component at a distance of h millimeters;

[0017] S2, acquiring an analog signal: powering the Hall sensor and collecting the magnetic field B of the surface of the metal component in the X direction under the condition of v uniform speed movement x ;

[0018] S3, acquiring a signal curve: filtering the analog signal through a signal conversion circuit and acquiring the magnetic induction intensity curve B x -t through the upper computer;

[0019] S4, extracting a characteristic signal: extracting the base value B x of the curve B x1 when there is no defect, then extracting the corresponding peak value B x2 , obtaining the peak change value ΔB=|B x2 -B x1 |, extracting the time t1 at which the curve B x -t starts to change from the base value B x1 , the arrival time t2 of the peak value, and obtaining the peak arrival time Δt=|t2-t1|;

[0020] S5, determining the defect position: t2 is the peak arrival time, that is, the time when the sensor detects the strongest defect signal, and the position of the defect is obtained according to the distance S=v×t2;

[0021] S6, respectively fitting the peak arrival time and the crack width relationship and the peak change value and the crack depth relationship according to the peak change value ΔB and the peak arrival time Δt; thereby calculating the unknown crack width and crack depth.

[0022] The calculation of the unknown crack width in step S6 comprises the following steps:

[0023] A1, setting multiple experiments with the same crack depth and different crack widths, obtaining the corresponding peak change value and peak arrival time of each experiment, taking the peak arrival time as the abscissa and the crack width as the ordinate, and performing linear fitting to obtain the relationship between the peak arrival time and the crack width;

[0024] A2. Extract the peak arrival time of the unknown defect and substitute the peak arrival time of the unknown defect into the relationship obtained in step A1 to obtain the width of the unknown defect.

[0025] The calculation of the unknown crack depth in step S6 specifically includes:

[0026] B1. Set up multiple experiments with the same crack width but different depths to obtain the corresponding peak change value and peak arrival time for each experiment. Use the peak change value as the horizontal axis and the different crack depths as the vertical axis to perform linear fitting to obtain the relationship between the peak change value and the crack depth;

[0027] B2. Extract the peak change value of the unknown defect and substitute the peak change value of the unknown defect into the relationship obtained in step B1 to obtain the depth of the unknown defect.

[0028] Beneficial effects of the present invention: The magnetic sensing device of the present invention includes a rectangular permanent magnet; a sensor arranged on one side of the permanent magnet; and a clamp for fixing the permanent magnet and the sensor. The present invention proposes a magnetic sensor and a detection method based on the structural sensor. When the surface of a metal component is smooth and defect-free and the movement speed is stable, the motional eddy current will stabilize at a certain value; however, when cracks appear on the surface of the metal component, the magnetic induction intensity will change significantly, and the difference in defects will also affect the magnitude of the change in magnetic induction intensity. Therefore, the present invention can detect this magnetic induction change by placing a sensor on one side of the permanent magnet, thereby achieving quantitative non-destructive detection of the size of surface defects of moving metal components. The magnetic sensor based on motional eddy currents provided by the present invention can detect the size of defects in moving metal components and can monitor surface defects of moving metal components. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Schematic diagram of the magnetic sensor structure of the present invention;

[0030] Figure 2 Schematic diagram of the relationship between the S′ reference frame and the S reference frame;

[0031] Where (a) is the S reference system, (b) is the S′ reference system;

[0032] Figure 3 A schematic diagram of motional vortex generation provided by an embodiment of the present invention;

[0033] Figure 4 A flow chart of sensor signal conversion provided by an embodiment of the present invention;

[0034] Figure 5 The t2 and t1 extraction method provided in the embodiment of the present invention;

[0035] Figure 6The comprehensive magnetic induction intensity change generated by the permanent magnet and the pipeline movement provided by the embodiment of the application

[0036] Figure 7 The magnetic induction intensity contour map of the crack provided by the embodiment of the application Figure 6

[0037] Figure 8 The 4mm depth crack group partial curve data provided by the embodiment of the application

[0038] Figure 9 The 4mm depth crack group partial curve data provided by the embodiment of the application

[0039] Figure 10 The 0.5mm width depth crack group partial curve data provided by the embodiment of the application

[0040] Figure 11 The maxwell finite element geometric model schematic diagram provided by the embodiment of the application DETAILED DESCRIPTION

[0041] In order to facilitate those skilled in the art to understand the technical content of the application, the content of the application is further explained below in combination with the drawings.

[0042] With the development of detection requirements and process technology, simple magnetic flux leakage nondestructive testing and eddy current nondestructive testing cannot meet the needs of increasingly complex detection environments, for example, the use of magnetic flux leakage or eddy current detection on metal components in a high-speed motion state will be affected by the dynamic eddy current. The application directly utilizes the dynamic eddy current to detect defects of the metal component in high-speed motion, which is more accurate.

[0043] The magnetic sensor based on the dynamic eddy current provided by the application has the structure as shown in Figure 1

[0044] The magnetic sensor is composed of a permanent magnet and a Hall sensor, wherein the permanent magnet is in the shape of a cuboid, and the size of the permanent magnet is a*b*c; the Hall sensor is located on one side of the permanent magnet, and the placement position of the Hall sensor is specifically the side of the permanent magnet relative to the metal component in motion, for example, if the permanent magnet moves to the right, the sensor is placed on the right side of the permanent magnet; the size of a single permanent magnet is d1*d2*d3, that is, the length is d1, the width is d2, and the height is d3, and the distance from the permanent magnet is l. Of course, other magnetic sensors can also be used in actual application.

[0045] Detection principle

[0046] The Maxwell equation set is the basis of all electromagnetic theories, and formula (1) is the differential form of the Maxwell equation set.

[0047] ​​

[0048] where, is the gradient, t is the time in the reference frame S, H is the magnetic field strength (A / m), B is the magnetic induction (T), E is the electric field strength (V / m), J is the current density (A / m2), D is the electric displacement vector (C / m2), and p is the charge density (C / m3).

[0049] The Lorentz transformation is a method of coordinate transformation between two inertial reference frames moving at a constant velocity relative to each other. The formula is as follows:

[0050]

[0051] where x, y, z are the coordinate axes in S; t is the time in S; x', y', z' are the coordinate axes in S'; t' is the time in S'; and c is the speed of light in a vacuum.

[0052] As Figure 2 shown in the figure, the relationship between the reference frames S and S' is shown, Figure 2 (a) Taking a plane wave as an example, the diffusion pattern of the plane wave in the S reference frame is shown, which is stationary relative to S. When the S' reference frame moves to the left relative to the S reference frame, the diffusion pattern of the plane wave in the S' reference frame is shown in Figure 2 (b) Due to the movement of the S' reference frame relative to the S reference frame, the velocity vectors in all directions in the S' reference frame are affected. Similarly, the electromagnetic field will also be affected due to the relative motion.

[0053] In an electromagnetic non-destructive testing system, the detection speed v is much smaller than c, so formula (2) will be simplified, and the Galilean transformation formula is as follows:

[0054]

[0055] In differential operation,

[0056]

[0057] where, is the differential operator in S; is the corresponding differential operator in S' corresponding to S. According to formula (3) and formula (4), the transformation equation of the field is:

[0058]

[0059] where H, E, B, D are the fields in S; H', E', B', D' are the corresponding fields in S' corresponding to S. Bringing formula (5) into Maxwell's equation, we get:

[0060]

[0061]

[0062] In formula (7), μ is a parameter, which is used to represent the linear relationship between H and B;

[0063]

[0064]

[0065] In formula (9), μ is a parameter, which is used to represent the linear relationship between H and B; is a scalar potential, the displacement current is ignored, and the Coulomb gauge is introduced and the magnetic potential A is a vector potential, which is introduced by the Coulomb gauge. Formula (6) and (7) can be simplified as:

[0066]

[0067]

[0068] In formula (10), the current density J mainly includes the excitation current J0 (when the magnetic field source is a coil) and the eddy current J m , then J can be written as:

[0069] J = J0 + J m (12)

[0070]

[0071] By bringing formula (12) and (13) into formula (10), we have:

[0072]

[0073] In formula (13) and (14), when the magnetic field source is a DC excitation coil or a permanent magnet, and the metal conductor is infinite and has no cracks, is zero; when the excitation coil is connected with alternating current, or the metal conductor contains cracks and other conditions that can change the magnetic field, will not be zero.

[0074] When the magnetic field source is taken as the reference system, the process that the metal test piece moves below the magnetic field source with a speed v can be equivalent to that the metal test piece experiences a magnetic field that changes with time. The changing magnetic field will induce an electric field, and therefore the eddy current will be generated on the metal surface, i.e. the eddy current generated by motion. According to the Faraday's law of electromagnetic induction, when the relative motion occurs between the metal member and the magnetic field source, the magnetic induction intensity passing through the loop will change, thereby forming the eddy current generated by motion in the metal member.

[0075] The magnetic field B generated by the permanent magnet is constant, and the positive and negative poles are as followsFigure 3 The relative movement of the metal member and the magnetic field source is regarded as the movement of the metal member cutting the magnetic induction lines along the negative direction of the Z axis at a speed v, and the direction of the magnetic induction intensity of the permanent magnet is as shown in Fig. 2. Figure 3 The magnetic field direction of the metal surface directly below the permanent magnet is perpendicular to the metal surface and downward or upward, the free electrons in the metal member are subjected to the Lorentz force F and move directionally to form the eddy current. According to the right-hand rule, it can be judged that the N-pole current direction of the metal surface directly below the magnetic field source is along the negative direction of the Y axis, and the S-pole current direction is along the positive direction of the Y axis. Since the current needs to form a closed loop, the current flows counterclockwise to the left along the magnetic induction lines of the metal passing through the permanent magnet after flowing along the positive / negative direction of the Y axis, and finally forms two eddy currents in the same direction at the two poles of the permanent magnet. The two eddy currents in different directions are respectively denoted as J m1 and J m2 .

[0076] When the surface of the metal member is smooth without defects and the movement speed is stable, the eddy current will also be stable at a certain value. However, when the surface of the metal member has a crack, the magnetic induction intensity will change more obviously, and the different defects will also affect the size of the change in the magnetic induction intensity. Therefore, by placing a sensor on one side of the permanent magnet, the change in the magnetic induction can be detected, and then by converting the detected magnetic induction signal into a digital signal through a circuit, the defect condition of the metal surface can be known, as shown in Fig. 4. Figure 4

[0077] After obtaining the signal curve, the curve is analyzed to obtain specific defect information.

[0078] The movement metal member defect detection method based on the magnetic sensor of the eddy current includes the following steps:

[0079] S1, the magnetic sensor is placed parallel to the surface of the moving metal member at a distance of e millimeters; the value of e should be as small as possible, and the surface roughness of the detection object should be considered in the value, without damaging the sensor under the premise of being as small as possible, and a margin of at least 1 mm should be provided in actual application. Therefore, when the surface roughness is Ra(max / μm) (max is the thickness of the corresponding surface), the thickness of the measured object is r, and the speed is v, the expression of e is:

[0080]

[0081] In the embodiment, Ra / r is 0.06 mm, v is 5 m / s, and e is actually 1.45. To facilitate control, it is set to 1.5.

[0082] S2, obtain the analog signal: power the Hall sensor, and collect the magnetic field B x of the metal member surface in the X direction under the condition of uniform speed v movement;​

[0083] S3, Obtain signal curve: through signal conversion circuit, filter the analog signal, and obtain the magnetic induction intensity curve B through the upper computer x -t, B x B is the magnetic induction intensity, and t is time;

[0084] S4, Extract characteristic signal: extract the base value B of the curve when there is no defect x1 Then, the corresponding peak value B is extracted x2 , and the peak change value ΔB = |B x2 -B x1 | is obtained. The time t1 at which the curve starts to change from the base value and the arrival time t2 of the peak value are extracted, and the peak arrival time Δt = |t2-t1| is obtained. The extraction method of t2 and t1 is shown in Figure 5 .

[0085] The specific extraction method is as follows: first, the curve is differentiated to obtain the differentiated data and its length, and each data point is respectively provided with a time value and a differential value. Then, two conditions are judged through a loop. The first condition is whether the differential value of each data point is greater than the differential values of the previous and subsequent 500 data points (500 data points are the best value after the simulation data is differentiated, and the actual number of previous and subsequent data points can be modified according to the actual situation). If it is greater than the differential values of the previous and subsequent 500 data points, it is judged to be an extreme point, and is stored in the last position of matrix A; the second condition is to calculate the average value a of the differential values of the previous 20 data points and the average value b of the differential values of the subsequent 40 data points, and if 0.2<b-a<0.3, it is stored in matrix B.

[0086] Then, it is judged whether the time value difference between each data point in matrix B and the previous data point is less than or equal to x (which can be modified according to the actual data). If it is less than, it is not stored in matrix C; if it is greater than, it is judged to be a change starting point, and is stored in the last position of matrix C, so as to eliminate adjacent starting points.

[0087] Loop until the last data point.

[0088] Since each data is looped from front to back, matrix A and C must have a one-to-one correspondence. The time value of the data point in A is subtracted from the time value of the data point in C to obtain the peak arrival time.

[0089] S5, Determine the defect position: t2 is the peak arrival time, that is, the time when the sensor detects the strongest defect signal, and the position of the defect is obtained according to the distance S = v × t2;

[0090] As shown in Figure 6 , the magnetic field intensity distribution when the permanent magnet moves relative to the surface of the test piece is shown, which reflects the influence of the crack on the detection signal under high-speed motion,Figure 7 for Figure 6 The enlarged image shows the extent of the crack's influence on the magnetic field.

[0091] S6. Calculate crack width:

[0092] S61. Establish the relationship between △t, △B and the width and depth of metal surface cracks: set m groups of n experiments according to actual needs, where the i-th experiment of the j-th group (i=1,2,…,n,j=1,2,…,m) is to set the crack width p on the metal component. i (p i Each group corresponds to the same and different within the group), the depth is h j (h j The defects of the same group and different groups are measured, △t, △B. First collect the defects of the jth group with depth h j The defect data is linearly fitted with △t as the horizontal coordinate and p as the vertical coordinate to obtain the width-△t equation p=α1*△t+β1, where α1 and β1 are constant coefficients determined by the sensor structure parameters;

[0093] S62, extracting the characteristic signal Δt of the unknown defect, substituting Δt into the obtained width-Δt equation to obtain the width p of the defect;

[0094] S7. Calculate the crack depth:

[0095] S71, collect the same width p of each experimental group in S61 i The defect data is linearly fitted with △B as the horizontal coordinate and h as the vertical coordinate to obtain the depth-△B equation h=α2*△B+β2, where α2 and β2 are constant coefficients determined by the sensor structure parameters;

[0096] S72, extracting the characteristic signal △B of the unknown defect, substituting △B into the obtained depth-△B equation to obtain the depth h of the defect;

[0097] Example:

[0098] The sensor chip uses the Infineon TLE4998S4 chip, which is 5.34mm*3.71mm*1.00mm in size and has a working power supply of 5V.

[0099] The specific geometric parameters of the magnetic sensor are shown in Table 1. The liftoff distance d4 is its distance from the metal component surface. The permanent magnet is made of NdFe30, the metal component is made of X80 steel, and the relative motion velocity v is set to 5 m / s. The specific parameters are shown in Tables 2 and 3, respectively.

[0100] Table 1 Magnetic sensor parameters

[0101]

[0102] Table 2 Physical properties and geometric parameters

[0103]

[0104] In the ansys maxwell software, a model is established for experiments, and three groups of experiments are set, each group having 9 experiments. The depths of the three groups of experiments are set to 4mm, 6mm and 10mm respectively, and the crack widths of each group of experiments are 0.1mm, 0.2mm, 0.3mm, 0.4mm, 0.5mm, 0.6mm, 0.7mm, 0.9mm and 1mm respectively, and experimental data is collected. Taking the 4mm depth experiment group as an example, part of the curve data at the defect is as shown in Figure 8 .

[0105] The peak arrival time △t and crack width of the 4mm depth experiment group are fitted, and the fitting curve is as shown in Figure 9 .

[0106] As the peak arrival time △t increases, the corresponding p will increase. The fitted curve is p = 0.08006 * △t - 13.14, R 2 2 is 0.979, which proves that △t has a linear response to p. When △t changes by 1ms, the corresponding crack width p will change by 0.08mm.

[0107] Similarly, the same operation fitting is performed on the 6mm depth and 10mm depth experiment groups, and the 6mm depth curve is p = 0.07894 * △t - 13.21, R 2 2 is 0.985, and the 10mm depth curve is p = 0.08050 * △t - 13.02, R2 is 0.981. It can be obtained that the change of depth has little effect on the relationship curve between peak arrival time △t and p.

[0108] The data of fitting the same width and different depth curves △B and depth is obtained, and taking 0.5mm width as an example, the fitting result is as shown in Figure 10 .

[0109] The final fitting results of all groups are as shown in Table 3.

[0110] Table 3 Fitting results of △B and depth

[0111]

[0112] As the width p decreases, the fitting curve α2 and β2 as a whole show a downward trend, and R 2 2 is relatively high as a whole, indicating that △B has a linear response to h.

[0113] After all the fitting curves are obtained, maxwell software is used to set multiple groups of cracks to explore the defect detection capability of the sensor, and the model is as shown in Figure 11

[0114] After the experimental curves are obtained, the △t and △B are counted, and are brought into the fitting curves calculated above, and the results are as shown in Table 4.

[0115] Table 4 Experimental results

[0116]

[0117] The width error of the magnetic sensor is within ±0.05mm, and the depth error is within ±0.6mm, and the dynamic eddy current magnetic sensor based on the present application has good detection capability for defects of moving metal components.

[0118] Those skilled in the art will appreciate that the embodiments described herein are presented for the purpose of aiding the reader in understanding the principles of the present application, and should be understood as not limiting the scope of protection of the present application to such specific recitations and embodiments. The present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of protection of the claims of the present application.​

Claims

1. A method of detecting defects in a moving metal member, characterized by, The method comprises the following steps: S1, placing a sensor based on eddy current generated by motion in parallel at a distance of e millimeters from the surface of a moving metal component; The sensor based on eddy current generated by motion comprises a cuboid permanent magnet, a magnetic sensor and a clamp; the magnetic sensor is arranged on one side of the cuboid permanent magnet, and the clamp is used for fixing the cuboid permanent magnet and the magnetic sensor; the magnetic field direction of the metal surface directly below the permanent magnet is perpendicular to the metal surface and downward or upward, the free electrons in the metal component are subjected to the action of the Lorentz force F and are oriented to move, thereby forming the eddy current generated by motion; The value of e should be set to have a margin of at least 1 mm; S2, acquire analog signal: power supply for magnetic sensor, collect the surface of the metal components in the state of uniform motion X direction magnetic field B x ; S3, acquisition signal curve: through signal conversion circuit, analog signal is filtered, and the magnetic induction intensity curve B is acquired through the upper computer x -t, t is time; S4, extracting characteristic signal: extracting curve B when there is no defect x - the base value B of t x1 , then extracting the corresponding peak value B x2 , obtaining the peak change value ΔB = |B x2 - B x1 ; extracting curve B x - t from the base value B x1 , the time t1 when the peak value starts to change, the arrival time t2 of the peak value, obtaining the peak arrival time Δt = |t2-t1|; S5, determining the defect position: t 2 is the peak arrival time, that is, the time when the sensor detects the strongest defect signal, and the position of the defect is obtained according to the mileage. S6, fitting the peak value change value ΔB and the peak value arrival time Δt to obtain a relationship between the peak value arrival time and the crack width and a relationship between the peak value change value and the crack depth, respectively; thereby calculating the unknown crack width and the crack depth; The calculation of the unknown crack width in step S6 comprises the following steps: A1, setting multiple experiments with the same crack depth and different crack widths, obtaining the peak value change value and the peak value arrival time corresponding to each experiment, taking the peak value arrival time as the horizontal coordinate and the crack width as the vertical coordinate, and performing linear fitting to obtain a relationship between the peak value arrival time and the crack width; A2, extracting the peak value arrival time of the unknown defect, and bringing the peak value arrival time of the unknown defect into the relationship obtained in step A1 to obtain the width of the unknown defect; The calculation of the unknown crack depth in step S6 comprises the following steps: B1, setting multiple experiments with the same crack width and different crack depths, obtaining the peak value change value and the peak value arrival time corresponding to each experiment, taking the peak value change value as the horizontal coordinate and the crack depth as the vertical coordinate, and performing linear fitting to obtain a relationship between the peak value change value and the crack depth; B2, extracting the peak value change value of the unknown defect, and bringing the peak value change value of the unknown defect into the relationship obtained in step B1 to obtain the depth of the unknown defect.

2. A method of detecting defects in a moving metal member according to claim 1, wherein The magnetic sensor is arranged on the side of the cuboid permanent magnet opposite to the side of the metal component.

3. A method of detecting defects in a moving metal member according to claim 2, wherein The magnetic sensor is a Hall sensor.

Images