A non-ferromagnetic wall thickness quantitative method based on variable pulse width excitation

Abstract

The application discloses a non-ferromagnetic wall thickness quantitative method based on a variable pulse width excitation, and comprises the following steps: a rising edge and a falling edge response of magnetic flux in a pickup coil are picked up through variable pulse width current excitation; with the increase of the pulse width of the excitation current, the signal attenuation of the rising edge speed and the falling edge speed becomes slow; and a feature slope corresponding to different pulse widths is extracted to realize the measurement of the metal wall thickness. The application uses the variable pulse width as the excitation signal, can flexibly control the eddy current penetration depth, realizes the wall thickness quantification from the surface layer to the deep layer, gives the signal dynamic characteristics, and thus higher measurement accuracy and repeatability are realized.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing technology, and in particular to a quantitative method for nonferromagnetic wall thickness based on variable pulse width excitation. Background Technology

[0002] With the development of modern industry, quantitative metal wall thickness measurement technology has become an indispensable part of the industrial field. It not only relates to product quality and performance but is also a key factor in ensuring production safety and efficiency. Accurate measurement of metal wall thickness helps manufacturers ensure their products meet specific specifications, avoiding product defects and usage risks caused by non-compliance with thickness standards. Therefore, the development of quantitative metal wall thickness measurement is of paramount importance.

[0003] In the prior art, a Chinese invention patent entitled "A Pulsed Eddy Current Array Probe and a Method for Detecting the Thickness of Ferromagnetic Metal Parts" (publication number CN119374467A) is disclosed. This patent provides a periodic pulse excitation method. Based on a combination structure of periodic pulses and multiple coils, it analyzes the average amplitude of the induced voltage signal in the attenuation segment, extracts characteristic quantities (such as the absolute value of the slope), and establishes a fitting curve model between the characteristic quantities and the wall thickness to estimate the thickness of the measured part. However, when detecting metals of different thicknesses or materials using periodic (fixed) pulses, its sensitivity range is fixed. If the thickness variation range is too large, the fixed pulse width may lead to signal saturation or insufficient intensity. Summary of the Invention

[0004] In view of the problems existing in the prior art for detecting the wall thickness of non-ferromagnetic metal materials, the present invention is proposed.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a quantitative method for nonferromagnetic wall thickness based on variable pulse width excitation. By using variable pulse width current excitation, the rising and falling edge responses of the magnetic flux in the coil are picked up. As the pulse width of the excitation current increases, the signal attenuation of the rising edge velocity and the falling edge velocity becomes slower. The characteristic slope corresponding to different pulse widths is extracted to realize the measurement of metal wall thickness.

[0006] As a preferred embodiment of the nonferromagnetic wall thickness quantitative method based on variable pulse width excitation in this invention, the nonferromagnetic wall thickness quantitative method based on variable pulse width excitation is a metal wall thickness quantitative method based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain. The dynamic point slope response is a method for quantitatively characterizing metal wall thickness. The dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain is a direct reflection of the metal wall thickness of the measured object. With the increase of pulse width current, the dynamic point of the falling edge magnetic flux in the pickup coil in the semi-logarithmic domain increases linearly. With the increase of metal wall thickness, the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain decreases.

[0007] As a preferred embodiment of the nonferromagnetic wall thickness quantitative method based on variable pulse width excitation in this invention, the method for quantitatively measuring metal wall thickness based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain includes the following steps:

[0008] S101. Use the analytical expression of the pulse-width variable excitation current signal as the excitation source;

[0009] S102. In pulsed eddy current detection technology, by adding an excitation source to the driving coil and acquiring the response signal in the pickup coil, the relationship between magnetic flux and time in pulsed eddy current detection is transformed into a semi-logarithmic domain representation. The core is to transform the exponential decay law of magnetic flux into a linear relationship through coordinate transformation. Before the next pulse width arrives, the magnetic flux with an excitation time of t is recorded as a dynamic point. Each set of increasing dynamic points is linearly fitted and the feature slope is extracted.

[0010] S103. Using an exponential function, the dynamic point slope of the response of different metal wall thicknesses and corresponding falling edge magnetic flux in the semi-logarithmic domain is fitted, and the final relational expression f(x)=Ae is obtained. Bx +Ce Dx In the formula, A, B, C, and D are fitting coefficients, f(x) represents the slope, and x represents the sample wall thickness;

[0011] S104. Substitute the slope of the dynamic point of the falling edge magnetic flux in the semi-logarithmic domain of the response signal of a metal of known thickness under the excitation of a variable pulse width signal into the fitting expression of step S103 to obtain its response value, and then compare it with the actual metal wall thickness to obtain the error.

[0012] As a preferred embodiment of the nonferromagnetic wall thickness quantitative method based on variable pulse width excitation in this invention, the excitation current signal is a piecewise function of time, consisting of a series of square wave pulses with pulse widths increasing according to a set rule, and has the following expression.

[0013] ;

[0014] Among them, FV(t) k () represents the analytical expression for the excitation current signal, where k is a natural number greater than 0, PW0 is the set initial pulse width, and PW Incre For the set pulse width increment, P Spac The fixed interval time base value is set; a (k+1) =a k +(k-1), and a1=0, To incentivize time, This represents the amplitude of the excitation current.

[0015] As a preferred embodiment of the nonferromagnetic wall thickness quantification method based on variable pulse width excitation in this invention, step S102 specifically involves processing the magnetic flux signal Φ(t) in the pickup coil, plotting all variable pulse widths PW(k) and their corresponding magnetic flux response signals Φ(t) in a semi-logarithmic coordinate system, where the vertical axis uses a logarithmic scale with base 10 and the horizontal axis uses a linear scale to form a magnetic flux curve about the pulse width. The magnetic flux is extracted as a dynamic point at time t=0.25s. The excitation source contains 5 different pulse widths, and the dynamic points of each complete excitation are a group. Each group has five dynamic difference curvatures. The five increasing dynamic points in each group are linearly fitted and the characteristic slope is extracted.

[0016] As a preferred embodiment of the nonferromagnetic wall thickness quantitative method based on variable pulse width excitation in this invention, the metal wall thickness is characterized by the dynamic difference curvature slope based on the transition characteristics of the variable pulse width excitation response. The time of the magnetic flux transition characteristics at the falling edge increases with the increase of the pulse width, and the dynamic difference curvature slope increases with the increase of the metal wall thickness. Through this correspondence, different metal wall thicknesses can be effectively measured.

[0017] As a preferred embodiment of the quantitative method for nonferromagnetic wall thickness based on variable pulse width excitation in this invention, the specific steps for characterizing the metal wall thickness based on the dynamic difference curvature slope of the variable pulse width excitation response transition characteristics are as follows:

[0018] S201. An analytical expression for the curvature K(t) of magnetic flux under excitation current signal FV(t) was proposed and analyzed.

[0019] S202. In pulsed eddy current detection technology, by adding an excitation source to the drive coil, the response signal in the pickup coil is acquired in the Cartesian coordinate system. Before the next pulse width arrives, the curvature change of the transition feature is recorded as dynamic difference curvature. Each set of increasing dynamic difference curvature is linearly fitted and the feature slope is extracted.

[0020] S203. Use a power function to fit the dynamic difference curvature slope of different metal wall thicknesses and corresponding transition features. The fitted relationship expression is f(x)=Ax+B, where f(x) is the dynamic difference curvature slope of the transition feature, and x represents the metal wall thickness. S204. Substitute the dynamic difference slope of the falling edge magnetic flux in the transition feature of the response signal of a metal of known thickness under the excitation of a variable pulse width signal into the fitting expression of step S203 to obtain its response value, and then compare it with the actual metal wall thickness to obtain the error.

[0021] As a preferred embodiment of the nonferromagnetic wall thickness quantitative method based on variable pulse width excitation in this invention, the analytical expression in step S201 is:

[0022] ;

[0023] in, Let be the analytical equation for the curvature of magnetic flux with respect to time, where k is the number of pulse sequences. ω represents the change in magnetic flux, j represents the angular frequency of the excitation, and j represents the phase shift.

[0024] As a preferred embodiment of the nonferromagnetic wall thickness quantitative method based on variable pulse width excitation in this invention, in step S202, the 90%-10% interval of the falling edge of magnetic flux is a transition feature. The curvature at 90% and 10% of the maximum value of the falling edge of magnetic flux is extracted as the dynamic difference curvature. The excitation source contains 5 different pulse widths. The dynamic difference curvature of each complete excitation is a group, and each group has five dynamic difference curvatures. The five increasing dynamic difference curvatures of each group are linearly fitted and the feature slope is extracted.

[0025] The beneficial effects of this invention are as follows: The excitation signal in this application is a variable pulse width. Compared with sinusoidal signals and periodic pulse width signals, the variable pulse width can flexibly control the eddy current penetration depth, realize quantitative wall thickness from the surface to the deep layer, and endow the signal with dynamic characteristics, thereby achieving higher measurement accuracy and repeatability. The feature values ​​provided in this application are based on the dynamic point slope in the semi-logarithmic domain. The semi-logarithmic domain can better show the proportional changes of the data. In the Cartesian coordinate system, when the time is 0.25s, the proportional change of the feature values ​​is not obvious. Therefore, using the semi-logarithmic domain can more clearly present this proportional growth change. This invention starts from the essence of the eddy current change inside the metal, deeply understands the intrinsic relationship between the dynamic feature values ​​of the variable pulse width response and the metal wall thickness, and explores effective feature characterization methods and parameters to describe the geometric dimensional properties of metals. It solves the contradiction between measurement accuracy and detection complexity in the prior art and is suitable for thickness measurement of non-ferromagnetic metal materials with a large thickness (not less than 10mm). Attached Figure Description

[0026] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of the variable pulse width signal provided in the embodiments of this application.

[0028] Figure 2 This is the non-ferromagnetic sample wall thickness model provided in the embodiments of this application.

[0029] Figure 3 These are the response curves of magnetic flux in the semi-logarithmic domain under different pulse widths provided in Embodiment 3 of this application.

[0030] Figure 4 This is a comparison diagram of variable pulse width-dynamic points for different metal wall thicknesses provided in Embodiment 3 of this application.

[0031] Figure 5 This is the non-ferromagnetic sample wall thickness-dynamic point slope fitting curve provided in Example 3 of this application.

[0032] Figure 6 These are the magnetic flux response curves in Cartesian coordinates under different pulse widths provided in Embodiment 4 of this application.

[0033] Figure 7 This is a comparison diagram of the dynamic difference curvature of different metal wall thicknesses provided in Embodiment 4 of this application.

[0034] Figure 8 It is the fitting curve of the non-ferromagnetic sample wall thickness-dynamic difference curvature slope provided in Example 4 of this application.

[0035] Figure 9 These are the response curves of magnetic flux in the semi-logarithmic domain under different pulse widths provided in Embodiment 5 of this application.

[0036] Figure 10 This is a comparison diagram of variable pulse width-dynamic points for different metal wall thicknesses provided in Embodiment 5 of this application.

[0037] Figure 11 This is the non-ferromagnetic sample wall thickness-dynamic point slope fitting curve provided in Example 5 of this application.

[0038] Figure 12 These are the magnetic flux response curves in Cartesian coordinates under different pulse widths provided in Embodiment 6 of this application.

[0039] Figure 13 This is a comparison diagram of the dynamic difference curvature of different metal wall thicknesses provided in Embodiment 6 of this application.

[0040] Figure 14 It is the fitting curve of the non-ferromagnetic sample wall thickness-dynamic difference curvature slope provided in Example 6 of this application.

[0041] The components include: 1. Specimen sample, 2. Insulation layer, 3. Drive coil, and 4. Pickup coil. Detailed Implementation

[0042] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0043] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0045] A quantitative method for nonferromagnetic wall thickness based on variable pulse width excitation includes the following steps: by exciting with a variable pulse width current, the rising and falling edge responses of the magnetic flux in coil 4 are picked up. As the pulse width of the excitation current increases, the signal attenuation of the rising edge velocity and the falling edge velocity becomes slower. The characteristic slope corresponding to different pulse widths is extracted to realize the measurement of the metal wall thickness.

[0046] Example 1: Refer to Figure 1 This is the first embodiment of the present invention. This embodiment provides a quantitative method for nonferromagnetic wall thickness based on variable pulse width excitation. This method is a quantitative method for metal wall thickness based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain. The dynamic point slope response is a quantitative characterization method for metal wall thickness. The dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain is a direct reflection of the metal wall thickness of the measured object. With the increase of pulse width current, the dynamic point of the falling edge magnetic flux in the pickup coil 4 in the semi-logarithmic domain increases linearly. With the increase of metal wall thickness, the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain decreases.

[0047] A quantitative method for metal wall thickness based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain includes the following steps.

[0048] S101. Use the analytical expression of the pulse-width variable excitation current signal as the excitation source;

[0049] S102. In the pulsed eddy current detection technology, by adding an excitation source to the driving coil 3 and acquiring the response signal in the pickup coil 4, the relationship between magnetic flux and time in pulsed eddy current detection is transformed into a semi-logarithmic domain representation. The core is to transform the exponential decay law of magnetic flux into a linear relationship through coordinate transformation. Before the next pulse width arrives, the magnetic flux with an excitation time of t is recorded as a dynamic point. The feature slope is extracted by linearly fitting each set of increasing dynamic points.

[0050] S103. Using an exponential function, the dynamic point slope of the response of different metal wall thicknesses and corresponding falling edge magnetic flux in the semi-logarithmic domain is fitted, and the final relational expression f(x)=Ae is obtained. Bx +Ce Dx In the formula, A, B, C, and D are fitting coefficients, f(x) represents the slope, and x represents the sample wall thickness;

[0051] S104. Substitute the slope of the dynamic point of the falling edge magnetic flux in the semi-logarithmic domain of the response signal of a metal of known thickness under the excitation of a variable pulse width signal into the fitting expression of step S103 to obtain its response value, and then compare it with the actual metal wall thickness to obtain the error.

[0052] The excitation current signal is a piecewise function of time, consisting of a series of square wave pulses with pulse widths increasing according to a set rule, and has the following expression:

[0053] ;

[0054] Among them, FV(t) k () represents the analytical expression for the excitation current signal, where k is a natural number greater than 0, PW0 is the set initial pulse width, and PW Incre For the set pulse width increment, P Spac The fixed interval time base value is set; a (k+1) =a k +(k-1), and a1=0, To incentivize time, This represents the amplitude of the excitation current.

[0055] Step S102 specifically involves processing the magnetic flux signal Φ(t) in the pickup coil 4, plotting all variable pulse widths PW(k) and their corresponding magnetic flux response signals Φ(t) in a semi-logarithmic coordinate system. The vertical axis uses a logarithmic scale with a base of 10, and the horizontal axis uses a linear scale, forming a magnetic flux curve about the pulse width. The magnetic flux is extracted as a dynamic point at time t=0.25s. The excitation source contains 5 different pulse widths. The dynamic points of each complete excitation are a group, and each group has five dynamic difference curvatures. The five dynamic points of each group are linearly fitted and the feature slope is extracted.

[0056] Example 2: Reference Figure 1 This is the second embodiment of the present invention. The difference between this embodiment and the previous embodiment is that the metal wall thickness is characterized by the dynamic difference curvature slope of the variable pulse width excitation response transition characteristics. The time of the falling edge magnetic flux transition characteristics increases with the increase of pulse width, and the dynamic difference curvature slope increases with the increase of metal wall thickness. Through this correspondence, different metal wall thicknesses can be effectively measured.

[0057] The specific steps for characterizing metal wall thickness based on the dynamic difference curvature slope of the variable pulse width excitation response transition characteristics are as follows:

[0058] S201. An analytical expression for the curvature K(t) of magnetic flux under excitation current signal FV(t) was proposed and analyzed.

[0059] S202. In the pulsed eddy current detection technology, by adding an excitation source to the drive coil 3, the response signal in the pickup coil 4 is acquired in the Cartesian coordinate system. Before the next pulse width arrives, the curvature change of the transition feature is recorded as the dynamic difference curvature. The incremental dynamic difference curvature of each group is linearly fitted and the feature slope is extracted.

[0060] S203. Use a power function to fit the dynamic difference curvature slope of different metal wall thicknesses and corresponding transition features. The fitted relationship expression is f(x)=Ax+B, where f(x) is the dynamic difference curvature slope value of the transition feature, and x represents the metal wall thickness.

[0061] S204. Substitute the slope of the dynamic difference in the magnetic flux at the falling edge of the response signal of a metal of known thickness under the excitation of a variable pulse width signal into the fitting expression of step S203 to obtain its response value, and then compare it with the actual metal wall thickness to obtain the error.

[0062] The parsing expression in step S201 is,

[0063] ;

[0064] in, Let be the analytical equation for the curvature of magnetic flux with respect to time, where k is the number of pulse sequences. ω represents the change in magnetic flux, j represents the angular frequency of the excitation, and j represents the phase shift.

[0065] In step S202, the 90%-10% interval of the falling edge of the magnetic flux is a transition feature. The curvature at 90% and 10% of the maximum value of the falling edge of the magnetic flux is extracted as the dynamic difference curvature. The excitation source contains 5 different pulse widths. The dynamic difference curvature of each complete excitation is a group, and each group has five dynamic difference curvatures. The five increasing dynamic difference curvatures of each group are linearly fitted and the feature slope is extracted.

[0066] In step S203, a power function is used to fit the dynamic difference curvature slope of different aluminum wall thicknesses and corresponding transition characteristics. The fitted relationship expression is f(x)=Ax 2 +Bx+C, where f(x) is the dynamic difference curvature slope value of the transition feature, and x represents the metal wall thickness.

[0067] In step S202, under variable pulse width excitation, for short pulses, the eddy currents excited at the rising edge are mainly concentrated on the surface. After the falling edge, the reverse eddy currents quickly cancel each other out, the signal quickly returns to zero, the curvature is large, and the transition time is short. For long pulses, the eddy currents excited at the rising edge have fully diffused into the depth of the material. After the falling edge, the reverse eddy currents are weak, the magnetic field stored inside the material is slowly released, the attenuation is weak, the curvature is small, and the transition time is long.

[0068] Example 3: Reference Figures 1-5 This is the third embodiment of the present invention, which differs from embodiment 1. This embodiment demonstrates the reliability of the method for quantifying metal wall thickness based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain through simulation experiments.

[0069] The dynamic point slope response of variable current excitation in the semi-logarithmic domain is a quantitative characterization method for metal wall thickness. The dynamic point slope of the response of variable pulse width current excitation in the semi-logarithmic domain is a direct reflection of the metal wall thickness of the measured object. With the increase of pulse width current, the dynamic point of the falling edge magnetic flux in the pickup coil 4 in the semi-logarithmic domain increases linearly. With the increase of metal wall thickness, the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain decreases.

[0070] Step S101: Figure 1 This is a schematic diagram of the variable pulse width signal provided in the embodiments of this application, which lays the foundation for a parameterized time-domain excitation function proposed in this invention. As seen in the diagram, the current is only "positive" or "negative," so the variable pulse width excitation current can be decomposed into an integration of "positive" and "negative" currents related to time t.

[0071] When the current performs the "k"th excitation, the duration of the "positive" current excitation is the "k"th time.

[0072] The next pulse is the time taken from the start of the rise time to the point where the current drops to 0 amperes. The rise time is t. r (k), with a falling edge time of t f (k), so the duration of the "k"th excitation, i.e., the "positive" current, is...

[0073] PW(k)=t f (k)-t r (k)=PW0+(k-1)·PW Incre .

[0074] After the positive current excitation ends, there is a waiting time for the negative current before the (k+1)th excitation can begin. Therefore, the time occupied by the negative current in the (k)th pulse excitation can be calculated by subtracting the falling edge time of the (k+1)th pulse from the rising edge time of the (k)th pulse. Specifically, t interval (k)=PSpac =t r (k+1)-t f (k), P Spac The pulse interval.

[0075] Set tr(1)=0s, current is I0, and current expression is FV(t), which is the current of the "k"th excitation.

[0076] A complete pulse width is equal to the sum of a positive current and a negative current, specifically satisfying the following:

[0077] PW(k)+P Spac =PW0+K·PW Incre +P Spac ;

[0078] Specifically, during the first excitation pulse, the current is caused to flow from t... r (1) To begin, the expression is:

[0079] I0·(t≥0&&t <PW0)+0·(t≥PW0&&t<PW0+P Spac ) .

[0080] Because when the second excitation pulse begins, it is based on the first excitation pulse. Specifically, with time as the coordinate axis, the first excitation pulse is plotted first, followed immediately by the second excitation pulse. The current as a function of time can be expressed as:

[0081] I O ·(t≥0&&t <PW0)+0·(t≥PW0&&t<PW0+P Spac )+I O ·(t≥PW0+P Spac &&t<2·PW0+P Spac +PW Incre )+0·(t≥2·PW0+P Spac +PW Incre &&t<2·PW0+2·P Spac +PW Incre ) ;

[0082] Similarly, when the "k"th excitation pulse begins, it can be further expressed as:

[0083] ;

[0084] In the formula a k+1 =a k +(k-1), and a1=1; n is the number of stimuli.

[0085] This application provides a mathematical expression for a time-varying pulse width excitation current that can be applied to the driving coil 3. Specifically, this mathematical expression is not an isolated function definition, but a parameterized time-domain excitation model that is deeply embedded in the physical detection model and tightly coupled with the electromagnetic response characteristics of the material.

[0086] The expression explicitly characterizes the rising time t of the k-th pulse in the pulse sequence. r (k) Falling edge time t f (k) and initial pulse width PW0, pulse width increment PW Incre and pulse interval P Spac The analytical relationship between them.

[0087] Figure 2 This is the non-ferromagnetic sample wall thickness detection model provided in the embodiments of this application. In the non-ferromagnetic sample detection model provided in this application, no covering layer structure is set. This design is based on the following mechanism:

[0088] The capping layer is typically made of high-conductivity materials such as aluminum and copper. In the detection of non-ferromagnetic samples, eddy current loss is the primary heat source, while hysteresis loss is negligible. Therefore, the "magnetic field concentration" effect that the capping layer may produce is unlikely to bring benefits such as the significant reduction in hysteresis loss seen in ferromagnetic materials. On the contrary, due to its extremely high conductivity, the eddy current skin effect is very pronounced, and eddy currents will be highly concentrated in the very shallow area of ​​the capping layer. If such a good conductor capping layer is applied, it will induce strong eddy currents. The reverse magnetic field generated by these eddy currents will have a significant shielding effect on the original excitation magnetic field, thus hindering the magnetic field from further penetrating into the internal non-ferromagnetic sample, leading to signal attenuation and reducing the quantitative sensitivity and accuracy of the internal sample wall thickness.

[0089] The non-ferromagnetic sample testing model in this application mainly includes a sample 1, an insulating layer 2 placed on the upper side of the sample 1, and a coaxially arranged drive coil 3 and pickup coil 4 above the insulating layer 2. The device for installing the drive coil 3 and pickup coil 4 is preferably made of plastic and has grooves for accommodating the drive coil 3 and pickup coil 4. The bottoms of the drive coil 3 and pickup coil 4 should be kept horizontal. The device can be a horizontal receiving platform with handles on both sides. Hold the handles on both sides of the device and place the bottom detection surface (parallel to the coil) above the metal to be tested. Alternatively, guide holes can be made on the horizontal receiving platform on both sides of the coil. A plastic guide rod is inserted into the horizontal receiving platform through the guide holes. The two sides of the guide rod along its length can be connected to the frame, which is fixed to a fixed platform. The axis of the guide rod and the bottom detection surface of the horizontal receiving platform need to be horizontal. The height of the coil can be adjusted simply by changing the height of the guide rod. The above structure for coil movement testing is a conventional structure and not an improvement in this application. The experimental technique can be understood even without accompanying drawings. The drive coil 3 has 800 turns, and its inner diameter, outer diameter, and height are r... 1d =32mm, r 2d =80mm and l 2d -l 1d =34mm; the number of turns of pickup coil 4 is 1200, and the inner diameter, outer diameter and height are r respectively. 1p =114 mm, r 2p =152mm and l 2p -l 1p =6mm, its lift-off height is set to l 1d =l 1p =1mm.

[0090] In this embodiment, a COMSOL Multiphysics magnetic field (mf) physical field is used to establish a system as follows: Figure 2 A two-dimensional axisymmetric model was constructed. Materials were selected from COMSOL's built-in material library and configured accordingly. The mesh in the model used a physics-controlled conventional mesh, and magnetic flux data was obtained through transient solution.

[0091] The excitation parameters are configured as follows: excitation current amplitude I0, set to 4A; initial pulse width PW0, 25 ms; pulse width increment PW... Incre The pulse interval time is 25ms. Spac The pulse interval is set to 1000ms (the interval between non-ferromagnetic pulses is set to 500ms). The first 5 pulses are taken, and the pulse widths are 25mm, 50mm, 75mm, 100mm, and 125mm.

[0092] In this embodiment, the standard sample is an aluminum plate. An excitation signal FV(t) is applied to the drive coil 3, generating a primary magnetic field. This primary magnetic field passes through the insulating layer 2 and then through the sample 1 in the air domain, inducing eddy currents in the sample 1. These eddy currents generate a secondary magnetic field, which also passes through the insulating layer 2 and then through the air domain. The pickup coil 4 senses the superposition of the primary and secondary magnetic fields, causing a change in the magnetic flux within the pickup coil 4.

[0093] Step S102: Figure 3 This embodiment of the application provides a magnetic flux versus time curve for a sample (aluminum metal) with a wall thickness of 25 mm. Specifically, an excitation signal of time-varying pulse width current expression FV(t) is applied to the driving coil 3 (specifically, a variable pulse width excitation signal FV(t) is generated using a signal generator (DDS function arbitrary waveform signal generator), the output of the signal generator is connected to the input of a power amplifier (OPA541 module), and the output of the power amplifier (OPA541 module) is connected to the two terminals of the driving coil 3), and the magnetic flux signal induced in the pickup coil 4 is acquired (the signal can be acquired by a host computer; specifically, step one: the pickup coil 4 is connected to an integrating circuit (because the induced voltage is the derivative of the magnetic flux with respect to time, i.e., the slope). The integrating circuit adopts an active integrator structure based on an operational amplifier. Specifically, the integrating circuit consists of an operational amplifier (such as LM358, OP07 or equivalent), an input resistor R1 and a feedback capacitor C1, forming an inverting input type integrator topology. The connection method is: the input resistor R1... One end of R1 is connected to the output of pickup coil 4, and the other end of R1 is connected to the inverting input of the operational amplifier; the non-inverting input of the operational amplifier is grounded; the feedback capacitor C1 is connected between the output and inverting input of the operational amplifier; the integrating circuit is existing technology and is not an improvement of this application, so no additional drawings are required; Step 2: The voltage signal can be restored to a signal proportional to the real-time magnetic flux through the integrating circuit. The magnitude of the ratio is related to the number of turns N. The integrated signal is sent to the differential amplifier for amplitude amplification and high-frequency noise is filtered out by the low-pass filter; Step 3: The adjusted analog signal is input to the ADC (analog-to-digital converter) for quantization and acquisition, and the output digital signal is used for analysis by the microprocessor (MCU); Step 4: The output digital signal is corrected using the calibration coefficients (such as the capacitance difference coefficient) pre-stored in the EEPROM memory to eliminate hardware differences; Step 5: The microprocessor (MCU) uploads the corrected digital signal to the host computer via USB or Wi-Fi module and uses software to draw the "magnetic flux-time" waveform.

[0094] The horizontal axis (time) is scaled linearly, and the vertical axis (magnetic flux) is scaled logarithmically to base 10. Exponential decay data appears as a straight line in the semi-logarithmic plot, facilitating feature extraction. Specifically, each curve (corresponding to a fixed pulse width) exhibits a linear decreasing trend, with the absolute value of the slope being the magnetic flux decay coefficient, reflecting the decay rate at that pulse width. Furthermore, the curve clusters do not intersect or overlap, and the decay characteristics corresponding to each pulse width are clearly distinguishable. As the pulse width increases, the magnetic flux at the same time point gradually increases.

[0095] For pulse excitation with a maximum pulse width PW0 equal to 125ms, the fall time is between 0.1s and 0.2s. To ensure that the selected eigenvalues ​​are obtained after all pulse width excitations have stabilized, t is preferably chosen. d It takes 0.25 seconds.

[0096] Figure 4 This embodiment of the application presents a comparison of variable pulse width-dynamic point curves for aluminum samples with different wall thicknesses. Under the same variable pulse width excitation, non-ferromagnetic metal samples (aluminum) with different wall thicknesses exhibit distinct magnetic flux variation characteristics. These differences provide a solid foundation for quantitative wall thickness measurement using dynamic point slope characteristic parameters. A detailed comparison of the pulse width-magnetic flux curves can further reveal the response characteristics of non-ferromagnetic metal materials under excitation, thereby improving the accuracy and stability of the quantitative wall thickness method.

[0097] With the continuous application of variable pulse width excitation, the magnetic flux generated by specimens with different wall thicknesses exhibits different trends. For specimens with larger wall thickness, the dynamic point change is relatively gradual, while for specimens with smaller wall thickness, the dynamic point change is more drastic.

[0098] The slope characteristic parameters were obtained using the method described above. These slope characteristic parameters exhibit high accuracy and stability, and can effectively characterize the magnetic flux variation characteristics of samples with different wall thicknesses under variable pulse width excitation. Using these slope characteristic parameters, quantitative models for different wall thicknesses can be constructed.

[0099] Step S103: Figure 5 The non-ferromagnetic sample wall thickness-dynamic point slope fitting curve provided in the embodiments of this application is shown in Table 1. The sample is a non-ferromagnetic metallic material, specifically aluminum. The characteristic signal is preferably obtained based on the magnetic flux method, that is, the key characteristic parameters (slope) corresponding to samples with different wall thicknesses are extracted.

[0100] Table 1. Dynamic point slopes corresponding to aluminum wall thicknesses

[0101]

[0102] By establishing a functional relationship between the wall thickness and the slope through fitting, and using fitting tools for fitting, the wall thickness-slope fitting formula for nonferromagnetic specimens (magnetic flux method) is as follows: , where A=0.001427; B= -1.758; C=0.01364; D=-0.07575.

[0103] The goodness-of-fit evaluation results are shown in Table 2.

[0104] Table 2 Evaluation of wall thickness fitting for nonferromagnetic specimens

[0105]

[0106] As can be clearly seen from the goodness-of-fit evaluation data in Table 2, the magnetic flux (dynamic point slope) method has very small fitting accuracy error in terms of residual sum of squares (SSE), coefficient of determination (R²), adjusted coefficient of determination (Adjusted R²), and root mean square error (RMSE).

[0107] The above results demonstrate that the feature quantities extracted based on the semi-logarithmic domain magnetic flux signal exhibit higher stability and a stronger correlation with sample thickness, making it a superior and more reliable feature signal. Compared to other traditional methods, this approach significantly improves measurement accuracy and can more accurately reflect the wall thickness variations of non-ferromagnetic metallic materials.

[0108] Step S104: Select a non-ferromagnetic sample (aluminum) with a wall thickness d = 22 mm. Measure the magnetic flux using the semi-logarithmic domain method. Given that the characteristic value is obtained under variable pulse width excitation at a time t = 0.25 s with a wall thickness of d = 22 mm, calculate the slope of the linear equation fitted by the magnetic flux (semi-logarithmic domain method) to be 0.01385. Substitute this into the fitting formula from step S103 to determine the non-ferromagnetic sample wall thickness as 22.38 mm. Therefore, the error between the actual and measured values ​​is 1.73%.

[0109] Example 4: Reference Figure 1 , Figure 2 and Figures 6-8 This is the fourth embodiment of the present invention. The similarity between this embodiment and embodiment 3 is that the standard sample is a metal aluminum plate. The difference between this embodiment and embodiment 3 is that this embodiment demonstrates the reliability of the dynamic difference curvature slope characterizing the metal wall thickness based on the transition characteristics of the variable pulse width excitation response through simulation experiments.

[0110] Implementation Method: Under different pulse width excitation conditions, the transition characteristic of the falling edge of the magnetic flux becomes more pronounced with increasing pulse width. Specifically, the transition time required for the magnetic flux increases with increasing pulse width. The transition time of the falling edge is shortest when PW0 = 25 ms, and the magnetic flux value is maximum at time t = 0.025 s. The transition region is the area where the magnetic flux changes significantly, specifically from time t = 0.0276 s to time t = 0.15 s, which accounts for 90% and 16.7% of the maximum magnetic flux time, respectively. Figure 6 As shown. The dynamic curvature k is taken at two time points, 90% and 16.7% of the time of maximum magnetic flux. i Where i = 1, 2. Calculate the dynamic difference curvature Δk by subtracting the values. As the pulse width increases, the points that account for 90% and 16.7% of the maximum flux time are closer to the transition region.

[0111] Therefore, the specific steps of the method for characterizing metal wall thickness based on the dynamic difference curvature slope of the variable pulse width excitation response transition characteristics are as follows: ① An analytical expression for the curvature of magnetic flux under the excitation current signal FV(t) is proposed and analyzed; ② The difference between the curvature at 90% and 16.7% of the maximum value of the falling edge of the magnetic flux is extracted as the dynamic difference curvature, and the five increasing dynamic difference curvatures in each group are linearly fitted and the characteristic slope is extracted; ③ The dynamic difference curvature slope of different metal wall thicknesses and corresponding transition characteristics is fitted using a power function; ④ According to the input-output fitting function, the obtained dynamic difference curvature slope value is input into the fitting function to solve for the measured metal wall thickness.

[0112] The transition feature is the transition segment, with Figure 6 Taking this as an example, compared with a pulse width of 25ms, the decay rate of magnetic flux on the falling edge is more gradual and the duration of the magnetic flux transition segment on the falling edge is significantly longer.

[0113] Implementation process:

[0114] Step S201: The relationship between magnetic flux and induced voltage in the frequency domain is as follows: ,

[0115] ;in, Let be the permeability in free space. For the amplitude of the harmonic component, l 1p To determine the lift-off height of pickup coil 4, l 2p To obtain the height from the top of coil 4 to insulation layer 2, l 1d For the lift-off height of drive coil 3, l 2d r is the height from the top of the drive coil 3 to the insulation layer 2. 2p To pick up the outer radius of coil 4, r 1p To pick up the inner radius of coil 4, r2d r is the outer radius of the driving coil 3. 1d Let n be the inner radius of the driving coil 3. p To pick up the number of 4 turns of the coil, n d Let h be the number of turns of the driving coil (3), h be the distance between the magnetic insulation boundaries, and J0 be the zeroth-order Bessel function of the first kind. = , = α i =(α 2 +jωμ0σ i ) 1 / 2 , σ i Let be the conductivity of the i-th layer in the layered structure. For discrete eigenvalues, Let J1 be the axial propagation coefficient of the i-th layer in the structure, where J1 is the first-order Bessel function of the first kind; The reflection coefficient of the layered structure;

[0116] The magnetic flux ΔΦ(ω) can be expressed as: ;

[0117] The time-domain response ΔΦ(t) can be obtained by inverting ΔΦ(ω) through the inverse Fourier transform: According to the general formula for curvature: , For function The absolute value of the second derivative, For function The first derivative yields the general formula for the curvature of the magnetic flux ΔΦ(ω): .

[0118] Step S202: According to the above implementation method, calculate the difference between the curvature at 90% and 16.7% of the maximum value of the falling edge of the magnetic flux under different pulse width excitations, and use this difference as the dynamic difference curvature. Eight numerical points were selected within the range of aluminum wall thickness d = 10~40 mm for the study. The dynamic difference curvature and slope of the dynamic difference curvature under different pulse width excitations for the eight aluminum wall thicknesses are shown in Table 3.

[0119] Table 3 Dynamic Difference Curvature and Slope Corresponding to Aluminum Wall Thickness

[0120]

[0121] Figure 7 This is a comparison diagram of the dynamic difference curvature for different metal wall thicknesses provided in the embodiments of this application. As the metal wall thickness increases, and under the same pulse width excitation, the transition characteristics of the magnetic flux become more obvious, and the points accounting for 90% and 16.7% of the maximum magnetic flux time are closer to the transition region. Therefore, the greater the metal wall thickness, the greater the value of the dynamic difference curvature, showing an increasing relationship.

[0122] Step S203: Use a power function to fit the eight metal wall thicknesses to their corresponding dynamic difference curvature slopes, such as... Figure 8 The curve obtained by fitting is shown below.

[0123] The fitted relationship between the metal wall thickness and the slope of the dynamic difference curvature is expressed as f(x) = Ax 2 +Bx+C, f(x) is the dynamic difference curvature slope value of the transition feature, and x represents the metal wall thickness. In the expression, A=8.189e-09, B=1.531e-07, C=-2.712e-06.

[0124] The data for evaluating the fitting effect are shown in Table 4.

[0125] Table 4. Wall thickness fitting evaluation for nonferromagnetic specimens

[0126]

[0127] The goodness-of-fit evaluation data in Table 4 clearly show that the magnetic flux (semi-logarithmic domain) method performs better in terms of residual sum of squares (SSE) and coefficient of determination (R²). 2 The fitting accuracy error is very small across the four evaluation metrics: adjusted coefficient of determination (R²), root mean square error (RMSE).

[0128] Step S204: Select 22mm thick aluminum metal and evaluate it using the above method. The resulting slope is 4.55644E-6. Substitute this into the fitting formula in step S203 to calculate the aluminum metal wall thickness as 21.88mm. Therefore, the error between the actual value and the measured value is 0.55%.

[0129] Based on the methods described in the above embodiments, this invention provides a mathematical expression for a time-varying pulse width excitation current that can be applied to the driving coil 3. It describes a specific, non-periodic excitation process. Specifically, this mathematical expression is not an isolated function definition, but a parameterized time-domain excitation model deeply embedded in the physical detection model and tightly coupled with the electromagnetic response characteristics of the material.

[0130] Based on the methods described in the above embodiments, this invention provides a signal processing method for quantitatively determining the wall thickness of non-ferromagnetic materials. This method is based on variable pulse width square wave excitation, but by processing different dynamic response signals and extracting the corresponding slopes, it fits the slopes to the thickness, achieving high-precision thickness inversion.

[0131] Example 5: Refer to Figure 1 , Figure 2 and Figures 9-11This is the fifth embodiment of the present invention. This embodiment differs from embodiment 3. The material selected in the simulation test is metallic copper, which proves the reliability of measuring the wall thickness of metallic copper.

[0132] Eight numerical points were selected within the range of copper wall thickness d = 10~40 mm for the study. The dynamic point curvature and slope of the dynamic point curvature under different pulse width excitations corresponding to the eight copper wall thicknesses are shown in Table 5.

[0133] Table 5. Dynamic point slopes corresponding to copper wall thickness.

[0134]

[0135] The dynamic point slopes of the response of different metal wall thicknesses and corresponding falling-edge magnetic flux in the semi-logarithmic domain are fitted using an exponential function, and the final relationship expression is f(x)=Ae Bx +Ce Dx In the formula, A=0.0005539; B= -1.957; C=0.01087; D=-0.05518, f(x) represents the slope, and x represents the sample wall thickness.

[0136] The goodness-of-fit evaluation results are shown in Table 6.

[0137] Table 6. Wall thickness fitting evaluation for nonferromagnetic specimens

[0138]

[0139] As can be clearly seen from the goodness-of-fit evaluation data in Table 6, the magnetic flux (dynamic point slope) method has very small fitting accuracy error in terms of residual sum of squares (SSE), coefficient of determination (R²), adjusted coefficient of determination (Adjusted R²), and root mean square error (RMSE).

[0140] In step S104, the slope of the dynamic point of the magnetic flux at the falling edge of the response signal of the 22mm thick copper metal under the excitation of the variable pulse width signal in the semi-logarithmic domain is substituted into the fitting expression to obtain its value of 22.16mm. Then, compared with the actual metal wall thickness, the error is found to be 0.727%.

[0141] The above results show that the feature quantity extracted based on the semi-logarithmic domain magnetic flux signal in this embodiment has higher stability and a stronger correlation with the sample thickness, making it a superior and more reliable feature signal.

[0142] Figure 9This represents the change in magnetic flux of the tested metallic copper in the semi-logarithmic domain under variable pulse width excitation. In existing techniques, the change in magnetic flux is typically extracted in a Cartesian coordinate system under periodic pulse width excitation. Therefore, under these techniques, the change in magnetic flux does not change with the arrival of the next pulse width, and in the Cartesian coordinate system, the flux tends to zero within a time frame of 0.2s-0.5s, failing to exhibit linearity; that is, a linearly increasing magnetic flux cannot be extracted.

[0143] from Figure 10 It can be seen that under variable pulse width excitation, the dynamic point increases linearly, and the slope of the dynamic point gradually decreases as the metal thickness increases. In contrast, in the prior art, i.e., under periodic pulse width excitation, the dynamic point changes horizontally, and the slope does not decrease with the increase of thickness.

[0144] from Figure 11 It can be seen that the fitting curve has a very small accuracy error and the detection range is 10mm-40mm, making it a reliable method.

[0145] Example 6: Refer to Figure 1 , Figure 2 and Figures 12-14 This is the sixth embodiment of the present invention. This embodiment differs from embodiment 4. The material selected in the simulation test is metallic copper, which proves the reliability of the dynamic difference curvature slope characterizing the metal wall thickness based on the transient characteristics of the variable pulse width excitation response.

[0146] The study selected eight numerical points within the range of copper wall thickness d = 10~40 mm. The dynamic difference curvature and slope of the dynamic difference curvature under different pulse width excitations corresponding to the eight copper wall thicknesses are shown in Table 7.

[0147] Table 7. Dynamic Difference Curvature and Slope Corresponding to Copper Wall Thickness

[0148]

[0149] In step S203, a power function is used to fit the dynamic difference curvature slope of different metal wall thicknesses and corresponding transition characteristics. The fitted relationship expression is f(x)=Ax 2 +Bx+C, where f(x) is the dynamic difference curvature slope value of the transition feature, x represents the metal wall thickness, A=3.308e-09, B=1.756e-07, C=-2.142e-06.

[0150] The goodness-of-fit evaluation results are shown in Table 8.

[0151] Table 8. Wall thickness fitting evaluation for non-ferromagnetic specimens

[0152]

[0153] As can be clearly seen from the goodness-of-fit evaluation data in Table 8, the magnetic flux (dynamic difference curvature slope) method has very small fitting accuracy error in terms of residual sum of squares (SSE), coefficient of determination (R²), adjusted coefficient of determination (Adjusted R²), and root mean square error (RMSE).

[0154] Using the above method to evaluate a 22mm thick copper sample, the slope obtained is 3.2432e-6. Substituting this into the fitting formula in step S203, the copper wall thickness is calculated to be 21.75mm. Therefore, the error between the actual and measured values ​​is 1.136%.

[0155] Figure 12 This represents the change in magnetic flux of the tested metallic copper in a Cartesian coordinate system under variable pulse width excitation. As the pulse width increases, the duration of the magnetic flux transition region gradually increases, meaning the transition characteristics become more pronounced. However, existing techniques typically apply periodic pulse width excitation, so the duration of the magnetic flux transition region does not increase with the arrival of the next pulse width, meaning the transition characteristics remain unchanged.

[0156] from Figure 13 As can be seen, the slope of the dynamic interpolation curvature gradually increases with the increase of the metal wall thickness, and the dynamic difference curvature increases linearly, which is a reference-free method. In contrast, in existing technologies, under periodic pulse width excitation, the slope of the dynamic difference curvature remains unchanged and is horizontal.

[0157] from Figure 14 As can be seen, the fitting curve has a very small accuracy error and a detection range of 10mm-40mm, making it a reliable method.

[0158] It should be noted that in this application, the signal attenuation at the falling edge speed is inseparable from the signal attenuation at the rising edge speed, and the signal attenuation at the rising edge speed directly affects the signal attenuation at the falling edge speed. Figure 6 and Figure 12 It can be seen that the rising edge signals are superimposed and do not have the clear layering of the falling edge signals. Therefore, the falling edge is selected to extract feature values ​​for subsequent wall thickness quantification.

[0159] As can be seen from Examples 3 and 6, when using the present invention to measure the wall thickness of non-ferromagnetic metal materials with a large thickness, the measurement error is within 2%, which is highly accurate.

[0160] Examples 3 and 5 are quantitative methods for metal wall thickness based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain. The higher the conductivity of the measured metal material, the smaller the value of the dynamic point slope. However, the final dynamic point slope of Examples 3 and 5 is consistent with the fitting effect of the measured metal wall thickness. This indicates that the method is not affected by the metal material.

[0161] Examples 4 and 6 characterize metal wall thickness using the dynamic difference curvature slope based on the transition characteristics of the variable pulse width excitation response. The higher the conductivity of the measured metal material, the more obvious the pulse width transition characteristics. The better the fit between the dynamic difference curvature slope and the measured metal wall thickness.

[0162] It is understood that the variable pulse width excitation parameters involved in the embodiments of the present invention are not limited to the specific values ​​given above and non-ferromagnetic metals. Those skilled in the art can make adaptive adjustments based on the electromagnetic properties of the material being tested (such as ferromagnetism and non-ferromagnetism, conductivity, and permeability) and the quantitative range of wall thickness.

[0163] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0164] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0165] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0166] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0167] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A quantitative method for nonferromagnetic wall thickness based on variable pulse width excitation, characterized in that: The method includes the following steps: using variable pulse width current excitation, the rising and falling edge responses of the magnetic flux in the coil are picked up. As the pulse width of the excitation current increases, the signal attenuation of the rising and falling edge velocities becomes slower. The characteristic slopes corresponding to different pulse widths are extracted to achieve the measurement of the metal wall thickness.

2. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 1, characterized in that... It lies in: The quantitative method for nonferromagnetic wall thickness based on variable pulse width excitation is a method for quantitatively measuring metal wall thickness based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain. The dynamic point slope response is a quantitative characterization method for metal wall thickness. The dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain is a direct reflection of the metal wall thickness of the measured object. With the increase of pulse width current, the dynamic point slope of the falling edge magnetic flux in the pickup coil in the semi-logarithmic domain increases linearly, while with the increase of metal wall thickness, the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain decreases.

3. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 2, characterized in that: A quantitative method for metal wall thickness based on the dynamic point slope of the variable pulse width excitation response in the semi-logarithmic domain includes the following steps. S101. Use the analytical expression of the pulse-width variable excitation current signal as the excitation source; S102. In pulsed eddy current detection technology, by adding an excitation source to the driving coil and acquiring the response signal in the pickup coil, the relationship between magnetic flux and time in pulsed eddy current detection is transformed into a semi-logarithmic domain representation. The core is to transform the exponential decay law of magnetic flux into a linear relationship through coordinate transformation. Before the next pulse width arrives, the magnetic flux with an excitation time of t is recorded as a dynamic point. Each set of increasing dynamic points is linearly fitted and the feature slope is extracted. S103. Using an exponential function, the dynamic point slope of the response of different metal wall thicknesses and corresponding falling edge magnetic flux in the semi-logarithmic domain is fitted, and the final relational expression f(x)=Ae is obtained. Bx +Ce Dx In the formula, A, B, C, and D are fitting coefficients, f(x) represents the slope, and x represents the sample wall thickness; S104. Substitute the slope of the dynamic point of the falling edge magnetic flux in the semi-logarithmic domain of the response signal of a metal of known thickness under the excitation of a variable pulse width signal into the fitting expression of step S103 to obtain its response value, and then compare it with the actual metal wall thickness to obtain the error.

4. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 3, characterized in that: The excitation current signal is a piecewise function of time, consisting of a series of square wave pulses with pulse widths increasing according to a set rule, and has the following expression: ； Among them, FV(t) k () represents the analytical expression for the excitation current signal, where k is a natural number greater than 0, PW0 is the set initial pulse width, and PW Incre For the set pulse width increment, P Spac The fixed interval time base value is set; a (k+1) =a k +(k-1), and a1=0, To incentivize time, This represents the amplitude of the excitation current.

5. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 4, characterized in that: Step S102 specifically involves processing the magnetic flux signal Φ(t) in the pickup coil, plotting all variable pulse widths PW(k) and their corresponding magnetic flux response signals Φ(t) in a semi-logarithmic coordinate system. The vertical axis uses a logarithmic scale with a base of 10, and the horizontal axis uses a linear scale, forming a magnetic flux curve about the pulse width. The magnetic flux is extracted as dynamic points at time t=0.25s. The excitation source contains 5 different pulse widths. The dynamic points of each complete excitation are grouped together, and each group has five dynamic difference curvatures. The five increasing dynamic points in each group are linearly fitted and the feature slope is extracted.

6. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 2, characterized in that: The dynamic difference curvature slope based on the transition characteristics of variable pulse width excitation response characterizes the metal wall thickness. The time of the magnetic flux transition characteristic at the falling edge increases with the increase of pulse width, while the dynamic difference curvature slope increases with the increase of metal wall thickness. This correspondence can be used to effectively measure different metal wall thicknesses.

7. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 6, characterized in that: The specific steps for characterizing metal wall thickness based on the dynamic difference curvature slope of the variable pulse width excitation response transition characteristics are as follows: S201. An analytical expression for the curvature K(t) of magnetic flux under excitation current signal FV(t) was proposed and analyzed. S202. In pulsed eddy current detection technology, by adding an excitation source to the drive coil, the response signal in the pickup coil is acquired in the Cartesian coordinate system. Before the next pulse width arrives, the curvature change of the transition feature is recorded as dynamic difference curvature. Each set of increasing dynamic difference curvature is linearly fitted and the feature slope is extracted. S203. Use a power function to fit the dynamic difference curvature slope of different metal wall thicknesses and corresponding transition features. The fitted relationship expression is f(x)=Ax+B, where f(x) is the dynamic difference curvature slope value of the transition feature, and x represents the metal wall thickness. S204. Substitute the slope of the dynamic difference in the magnetic flux at the falling edge of the response signal of a metal of known thickness under the excitation of a variable pulse width signal into the fitting expression of step S203 to obtain its response value, and then compare it with the actual metal wall thickness to obtain the error.

8. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 7, characterized in that: The parsing expression in step S201 is, ； in, Let be the analytical equation for the curvature of magnetic flux with respect to time, where k is the number of pulse sequences. ω represents the change in magnetic flux, j represents the angular frequency of the excitation, and j represents the phase shift.

9. The method for quantitatively measuring nonferromagnetic wall thickness based on variable pulse width excitation as described in claim 8, characterized in that: In step S202, the 90%-10% interval of the falling edge of the magnetic flux is a transition feature. The curvature at 90% and 10% of the maximum value of the falling edge of the magnetic flux is extracted as the dynamic difference curvature. The excitation source contains 5 different pulse widths. The dynamic difference curvature of each complete excitation is a group, and each group has five dynamic difference curvatures. The five increasing dynamic difference curvatures of each group are linearly fitted and the feature slope is extracted.

Images