A metal pipeline defect detection method based on time-frequency domain electromagnetic method

Abstract

The application provides a metal pipeline defect detection method based on time-frequency domain electromagnetic method, which comprises the following steps: (1) establishing a time-frequency domain double-mode electromagnetic sensor test system with time-frequency domain coils in parallel, and measuring and acquiring electromagnetic response parameters through two modes; (2) frequency domain measurement, obtaining the depth d of a near-surface small-size defect n The function relationship of the surface size r of the small-size defect and the induced voltage difference U fd ; (3) time domain measurement, obtaining the function relationship of the attenuation rate k and the thickness D; and (4) data fusion of the time domain measurement and the frequency domain measurement for detection optimization of different types of defects.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of nondestructive testing, and relates to a measuring method based on the principle of electromagnetic induction, which is used for non-contact measurement of near-surface and far-surface defects of metal. BACKGROUND

[0002] For marine or buried oil and gas pipelines, the inner and outer surfaces of the pipelines are susceptible to corrosion damage of different degrees under the influence of the transportation medium and the environment, which affects the transportation safety and service life of the pipelines. Due to the complexity of the external environment of the pipeline, the detection personnel hope to carry defect detection sensors on the pipeline robot to measure defects inside the pipeline. The traditional defect detection methods each have their own limitations and application scope, and it is difficult to detect defects with large size span and far-surface defects.

[0003] The eddy current electromagnetic method is based on the principle of electromagnetic induction. When a coil with alternating current is close to a metal test piece to be measured, an alternating electromagnetic field is generated around the coil. This electromagnetic field produces a changing electric eddy current on the surface of the test piece, and the electric eddy current in turn produces a magnetic field that hinders the change of the magnetic field of the coil. This process causes the impedance of the eddy current probe to change, and this change is related to the electrical conductivity, magnetic permeability and surface state of the test piece. When there is a defect on the surface of the metal test piece, the electrical conductivity and magnetic permeability will change, thereby affecting the distribution of the electric eddy current at that place. By detecting the impedance of the eddy current or the amplitude change of the probe, the surface state of the test piece can be detected. Due to the advantages of non-contact, non-invasiveness, low cost and portability, the eddy current detection method has received widespread attention in the field of metal pipeline defect detection

[0004] The eddy current electromagnetic method mainly includes time domain electromagnetic method (i.e. transient electromagnetic method TEM) and frequency domain electromagnetic method (FDEM). The difference lies in that in the frequency domain electromagnetic method, one or more frequency harmonics are used for excitation to obtain a secondary induction magnetic field, and the size, position and shape of the defect are determined by analyzing the impedance change or the amplitude and phase of the induced voltage signal; the transient electromagnetic method obtains a decaying secondary induction magnetic field by pulse or square wave excitation, and detects the defects of the metal pipeline by analyzing the change of the induced voltage signal with time.

[0005] However, the transient electromagnetic method and the frequency domain electromagnetic method each have their own limitations. The transient electromagnetic method can detect large-area corrosion defects, but it is difficult to detect defects smaller than the probe size; the frequency domain electromagnetic method can only characterize the position and size of small-size defects on the metal surface, and cannot measure large-area defects and far-surface defects. Considering the complexity of the characteristics of the metal pipeline defects, dual-mode measurement and data fusion are carried out by combining time domain and frequency domain, which can measure defects of different sizes and can also characterize far-surface defects, so as to obtain more comprehensive defect information. SUMMARY

[0006] The application designs a metal defect detection method based on time-frequency domain electromagnetic method, which measures the metal pipeline or metal plate to be measured by time domain method and frequency domain method respectively, obtains electromagnetic response parameters in time domain and frequency domain, fuses time-frequency domain data, classifies defect signal characteristics of different scales and far and near surfaces, and thus inverses defects, so as to realize a detection method suitable for a wide size range and capable of measuring far surface defects.

[0007] A metal pipeline defect detection method based on time-frequency domain electromagnetic method, comprising the following steps:

[0008] (1) Establishing a time-frequency domain double-mode electromagnetic sensor test system with time-frequency domain coils in parallel, and measuring electromagnetic response parameters by two modes respectively;

[0009] (2) Frequency domain measurement, obtaining the depth d of a near-surface small-size defect n The function relationship of the surface size r of the small-size defect and the induced voltage difference U fd ;

[0010] (3) Time domain measurement, obtaining the function relationship of the attenuation rate k and the thickness D;

[0011] (4) Data fusion of the time domain method and the frequency domain method for detection optimization of different types of defects:

[0012] 1) Calibrating the parameter U fdmin , that is, the minimum voltage difference of the near-surface defect, in the scanning process, when U fd > U fdmin , it is judged that there is a near-surface defect at the corresponding position, otherwise, it is judged that there is no near-surface defect;

[0013] 2) Obtaining the surface size r of the near-surface defect in the coordinate range of U fd > U fdmin in the frequency domain measurement, assuming that the diameter of the frequency domain receiving coil is d coil , and there are two cases:

[0014] (a) According to the size of the frequency domain receiving coil, a threshold coefficient h is determined, if r≤hd coil , it is considered that the near-surface defect is a small-size defect, and the depth of the near-surface defect is d n obtained by the frequency domain measurement;

[0015] For the position without near-surface defect or only with small-size defect, the depth of the far-surface defect is:

[0016] d f = D-D t

[0017] Wherein, d fD is the standard thickness of the metal pipeline without defects, D t is the thickness of the metal pipeline at the position detected by the time domain method;

[0018] (b) If r > hd coil , the near-surface defect is considered to be a large-size defect, and the near-surface defect depth is calculated according to the measurement result of the time domain method:

[0019] d n = D - D t - d f

[0020] wherein d n is the near-surface defect depth of the metal pipeline, D is the standard thickness of the metal pipeline without defects, D t is the thickness of the metal pipeline at the position detected by the time domain method; d f is the far-surface defect depth of the metal pipeline, and the d f at the position is considered to be the same as that at the nearby position.

[0021] Further, the method of step (2) is:

[0022] For the metal pipeline to be measured, the change of the induced voltage of the frequency domain receiving coil at different positions is obtained by scanning. If there is a near-surface defect, the induced voltage difference U fd of the frequency domain receiving coil changes with the scanning coordinates, and the extreme point of the voltage change during the scanning process is the center position of the defect. According to the coordinate range of the change of the induced voltage difference U fd , the surface size of the defect is determined.

[0023] Let the depth of the near-surface small-size defect be d n , and the surface size be r. The near-surface small-size defects with different surface sizes and depths are measured, and the corresponding induced voltage difference values U fd collected are calibrated to fit the depth d n of the near-surface small-size defect. The function relationship between the surface size r of the small-size defect and the induced voltage difference value U fd .

[0024] Further, in step (2), the method for obtaining the induced voltage difference U fd of the frequency domain receiving coil is:

[0025] The metal pipeline with a standard thickness without defects is measured, and the induced voltage value of the frequency domain receiving coil is set as U st .

[0026] For the metal pipeline to be measured, the induced voltage value of the frequency domain receiving coil is set as U f .

[0027] The induced voltage difference between the two is U fd =U f -U st .

[0028] Furthermore, the depth d of small-sized defects near the surface n Regarding the surface size r of small defects and the difference in induced voltage U fd The functional relationship is as follows:

[0029]

[0030] In the formula, the constant a ij Let m and n be the fitting coefficients, respectively, and U = 1 / n. fd And the highest power of r.

[0031] Furthermore, in the frequency domain method measurement in step (2), the frequency domain coil consists of an excitation coil and a receiving coil, and is scanned along the extension direction of the pipe while maintaining a certain lift height from the pipe wall; during detection, the signal generator provides the excitation coil with a high-frequency sinusoidal alternating signal to excite the harmonic magnetic field, which forms an induced eddy current in the metal pipe being measured. The secondary induced magnetic field generated by the eddy current forms a secondary induced current in the receiving coil, and the amplitude of the induced voltage of the receiving coil is measured.

[0032] Furthermore, the method for step (3) is as follows:

[0033] Let the attenuation rate of the induced voltage signal in the later stage in the time domain be k, and the thickness of the metal pipe be D. Measure the metal pipes of different thicknesses, obtain the corresponding attenuation rates, calibrate them, and fit the functional relationship between the attenuation rate k and the thickness D.

[0034] The attenuation rate k at different locations is obtained by scanning the metal pipe under test using a time-domain method coil. t Then, the thickness D of the metal pipe at different locations can be calculated. t .

[0035] Furthermore, the functional relationship between the attenuation rate k and the thickness D in step (3) is as follows:

[0036] D = g(k) = ak b

[0037] In the formula, a and b are fitting coefficients.

[0038] Further, in step (3), the time domain coil is composed of an excitation coil and a receiving coil, the excitation coil is inside, and the receiving coil is outside, and the time domain coil is scanned along the extension direction of the pipeline with a certain height from the pipe wall; when detecting, the signal generator gives the excitation coil a low-frequency pulse signal, excites the decaying pulse induced magnetic field, forms an induced eddy current in the measured metal pipeline, and the secondary induced magnetic field generated by the eddy current forms a secondary induced current in the receiving coil, and the curve of the induced voltage signal of the receiving coil changing with time is measured.

[0039] Further, h=2.

[0040] The above technical scheme is adopted in the present application, and the present application has the following advantages:

[0041] (1) The time-frequency domain electromagnetic measurement model of the metal pipeline defect is established by using the information of the time domain electromagnetic and frequency domain electromagnetic dual-mode measurement, and the measurement of the near-surface defect and the far-surface large-size defect is realized, and the detectable range is widened compared with the traditional detection method.

[0042] (2) The defects existing near the surface and far from the surface can be measured at the same time by the organic fusion of the time-frequency domain electromagnetic measurement data, and a more complete and comprehensive metal pipeline defect thickness distribution map is obtained.

[0043] (3) The parameter information obtained by the time-frequency domain electromagnetic method can be compensated for each other, so that the reliability and accuracy of the detection data are greatly improved. The method is simple to operate, has good portability, and can be measured online. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 The structure block diagram of the metal pipeline defect detection system of the time-frequency dual-mode electromagnetic method provided by the present application is shown.

[0045] Figure 2 The sensor coil structure diagram of the time domain and frequency domain coil for measuring the metal test piece in the measurement method of the present application is shown.

[0046] Figure 3 The influence curve diagram of the defect depth on the detection voltage difference in the frequency domain electromagnetic method for the embodiment of the present application is shown.

[0047] Figure 4 The detection voltage curve diagram of the test piece with different thicknesses in the time domain electromagnetic method for the embodiment of the present application is shown.

[0048] Figure 5 The schematic diagram when the large-area near-surface defect and the far-surface defect exist at the same time for the embodiment of the present application is shown.

[0049] Figure 6 The flow chart of detecting the defect of the metal test piece by the time-frequency dual-mode electromagnetic method for the embodiment of the present application is shown. DETAILED DESCRIPTION

[0050] The present application provides a kind of metal pipeline defect detection method based on time-frequency domain electromagnetic method, it is applicable to the detection of large diameter ferromagnetic metal pipeline or flat plate.Based on the theoretical model of metal pipeline defect detection of frequency domain electromagnetic method and time domain electromagnetic method, the metal pipeline defect detection model of time-frequency domain electromagnetic method is obtained by experiment calibration and data fusion.The present application is described in detail in conjunction with the drawings and examples as follows:

[0051] Figure 1 It is the structure diagram of metal specimen defect detection system based on time-frequency domain electromagnetic method, including signal generating device, power amplifier, coil, signal processing module, data acquisition module and parameter calculation module.

[0052] Figure 2 (a) and (b) are respectively the spatial distribution schematic diagram of the measured metal specimen measured by frequency domain and time domain coils.The excitation coil and the receiving coil used by frequency domain method are the same structure, and the number of turns is about 100 turns;The excitation coil is inside, and the receiving coil is outside in time domain method, and the number of turns is 800 and 1000 respectively.

[0053] Figure 6 It is the flow chart of metal specimen defect detection of time-frequency dual mode electromagnetic method, and the metal pipeline defect detection method based on time-frequency domain electromagnetic method is described in detail as follows.

[0054] The detection method includes the following steps:

[0055] Step one: the theoretical model of frequency domain electromagnetic method for detecting near-surface small size defect is

[0056]

[0057] Wherein, d n The depth of near-surface defect of metal pipeline, U fd The difference between the frequency domain detection voltage at this position and the detection voltage of standard specimen without defect, r is the size of near-surface defect, a ij Is the constant coefficient calibrated by least square method and other optimization algorithms.

[0058] The theoretical model of time domain electromagnetic method for detecting ferromagnetic metal specimen thickness can be referred to the paper: Chen Xinglei, Lei Yinzhao.Induction voltage time domain approximation formula of ferromagnetic flat plate pulse eddy current field[J].Transaction of Electrical Engineering Technology, 2015, 30 (10): 14-19.DOI:10.19595 / j.cnki.1000-6753.tces.2015.10.003.

[0059] The theoretical model is:

[0060]

[0061] where u e (t) is the time-domain solution of eddy current induced voltage in the receiving coil under the falling edge pulse excitation, i′ f (t) is the derivative of the falling edge of the excitation current with respect to time, "*" represents the convolution operation with respect to time, λ i is the i-th positive root of the transcendental equation J1(ρx) = 0, ξ tk and ξ ck are the k-th positive roots of the equation tanξ + 2ξ / μ r λ i D = 0 and cosξ - 2ξ / μ r λ i D = 0. C d (λ i ) and C p (λ i ) are the coil coefficients of the excitation coil and the detection coil, respectively. D, ρ, σ, μ are the thickness, density, conductivity and permeability of the ferromagnetic plate specimen, respectively, where μ = μ0μ r , μ0 and μ r are the vacuum permeability and the relative permeability, respectively.

[0062] When the relative permeability μ r >> 1, and the eddy current diffuses to the far surface of the specimen (late, t >> 0), the approximate formula of the late induced voltage signal can be obtained by simplifying equations (2) and (3)

[0063]

[0064] where,

[0065]

[0066] τ d = μσD 2 / π 2 (6)

[0067] From equations (5) and (6), U e is a function related to the material and the lift-off height, and τ d is the eddy current diffusion time constant defined for the ferromagnetic plate. Taking the derivative of the logarithm of both sides of equation (6) with respect to time, taking the absolute value, and substituting equation (6) into it, we get:

[0068]

[0069] where k is the decay rate of the induced voltage signal in the late stage, represented by the absolute value of the signal decay slope in the single logarithmic coordinate system, π 2 / μσ is a fixed value, so the theoretical model of the thickness of the metal specimen and the decay rate can be obtained

[0070] D = g(k) = ak b (8)

[0071] Where D is the thickness of the ferromagnetic metal specimen, k is the attenuation rate of the induced voltage signal in the later stage, and a and b are constants measured through experimental calibration.

[0072] Step 2: Measure the standard metal specimen using frequency domain electromagnetic method and time domain electromagnetic method to obtain the experimental calibration parameters.

[0073] A schematic diagram of frequency domain electromagnetic detection is shown below. Figure 2 As shown in (a), the coil is kept at a certain lifting height from the metal specimen under test. A sinusoidal signal with an amplitude of 10V and a frequency of 100kHz is passed into the excitation coil. The induced magnetic field generated by the signal induces eddy currents in the metal specimen under test. The secondary induced magnetic field generated by the eddy currents induces a voltage signal in the receiving coil. The differential signal at both ends of the receiving coil is input into the signal processing module, and after acquisition and analysis, the frequency domain detection voltage at that position is obtained.

[0074] The standard frequency domain test voltage value U was obtained by measuring a defect-free standard thickness metal specimen using the frequency domain electromagnetic method. st A series of metal specimens with near-surface defects of different sizes and depths were fabricated, and the corresponding detection voltage values ​​U were obtained using the frequency domain electromagnetic method. f Thus, the induced voltage difference is obtained.

[0075] U fd =U f -U st (9)

[0076] U was obtained through experiments. fd With different dimensions r and depth d n The defect relationship correspondence table is used to fit a function to calculate the equation (1)d. n =f(U fd The specific expression for r). Figure 3 The figure shows the effect curve of defect depth on detection voltage difference in the frequency domain electromagnetic method obtained by COMSOL finite element simulation. The horizontal axis represents the near-surface defect depth d. n The vertical axis represents the induced voltage difference U. fd The defects used in the simulation are circular defects with the same surface size r.

[0077] A schematic diagram of time-domain electromagnetic detection is shown below. Figure 2As shown in (b), the coil is kept at a certain lifting height from the metal specimen under test. A pulse signal with an amplitude of 50V, a frequency of 1Hz, and a duty cycle of 10% is passed into the excitation coil, which generates a decaying pulsed induced magnetic field. Decaying eddy currents are generated in the metal specimen under test. The secondary induced magnetic field generated by the eddy currents generates a rapidly decaying induced voltage signal in the receiving coil. The differential signal at both ends of the receiving coil is input to the signal processing module, and after acquisition and analysis, the time-domain detection voltage change curve at this position is obtained.

[0078] Defect-free metal specimens of different standard thicknesses were measured using the time-domain electromagnetic method. Figure 4 The figure shows the curves of induced voltage versus time during time-domain electromagnetic method testing, obtained through COMSOL finite element simulation. The four curves represent the detection voltage curves for metal specimens with thicknesses of 10, 12, 14, and 16 mm, respectively. Figure 4 It can be seen that when the induced voltage decays to the late stage, it decays approximately linearly in a single logarithmic coordinate system. The absolute value of the slope of the signal in the late stage in the single logarithmic coordinate system is used as the attenuation rate k. The correspondence between the attenuation rate k and the thickness D is obtained. The parameters a and b in equation (8) are obtained by fitting.

[0079] Step 3: Measure the metal specimen using the time-domain electromagnetic method. A time-domain electromagnetic coil is used to scan the specimen to obtain the attenuation rate k at different locations. t Then, the thickness D of the metal specimen at different locations can be calculated. t =g(k t Under the time-domain electromagnetic method, the effect of small near-surface defects on the attenuation rate is minimal and can be ignored. That is, what is measured is the thickness change caused by large-area corrosion thinning.

[0080] Step 4: Measure the metal specimen using the frequency domain electromagnetic method. A frequency domain electromagnetic coil is used to scan the specimen, obtaining the voltage difference U at different locations. fd Calibrate a parameter U fdmin U is the minimum voltage difference at which a near-surface defect is considered to exist. fdmin Determined by the equipment's detection sensitivity. fd >U fdmin When the surface is at a certain location, a near-surface defect exists; otherwise, there is no near-surface defect. (U) fd >U fdmin The coordinate range is used to obtain the surface size r of the near-surface defect, and the diameter of the frequency domain coil is d. coil .

[0081] Step 5: Combining the data obtained from the time-domain electromagnetic method and the frequency-domain electromagnetic method, the following classification and discussion are conducted:

[0082] (a) If r≤2d coilIf r > 2d, it is considered that the near-surface defect at the position is a small-size defect, and the near-surface defect depth d n is determined by formula (1). For the position where there is no near-surface defect or only small-size defect, the far-surface defect depth is:

[0083] d f = D - D t (10)

[0084] wherein d f is the far-surface defect depth of the metal sample, D is the standard thickness of the metal sample without defect, and D t is the thickness of the metal sample at the position detected by the time-domain method.

[0085] (b) If r > 2d coil , it is considered that the near-surface defect at the position is a large-size defect, and the following examples are discussed in combination with Figure 5 . As shown in the figure, the standard thickness D = 10 mm, the bd segment has a large-size near-surface defect with a depth of 2 mm, and the ce segment has a large-size far-surface defect with a depth of 2 mm. Figure 5

[0086] When the time-frequency dual-mode detection is performed, the thickness D t of each position is determined by the time-domain method, i.e., the ab and ef segments are 10 mm thick, the bc and de segments are 8 mm thick, and the cd segment is 6 mm thick.

[0087] The bc segment: the frequency-domain method detects that the bd segment has a large-size near-surface defect, and the ab segment has no far-surface defect. According to the continuity of the large-size corrosion defect, it is considered that the bc segment should be a near-surface large-size defect, d n = D - D t = 2 mm.

[0088] The cd segment: the frequency-domain detection voltage of the cd segment has no obvious change (the difference is less than U fdmin ) compared with that of the bc segment, which indicates that the thickness of the near-surface defect of the cd segment has no change compared with that of the bc segment, i.e., the thickness change is caused by the far-surface defect. Therefore, d n = 2 mm, d f = D - D t - d n = 2 mm.

[0089] The de segment: the frequency-domain method detects that the de segment has no large-size near-surface defect, and therefore the thickness change of the de segment is caused by the far-surface defect, i.e., d f = D - D t = 2 mm.

[0090] wherein d n ​d is the depth of the near-surface defect of the metal test piece f D is the depth of the far-surface defect of the metal test piece t D is the thickness of the metal pipe at the location detected by the time domain method.

Claims

1. A method for detecting defects in metal pipes based on time-frequency domain electromagnetic methods, comprising the following steps: (1) Establish a time-frequency domain dual-mode electromagnetic sensor test system with parallel time-frequency domain coils, and measure and obtain electromagnetic response parameters through two modes respectively; (2) Frequency domain method measurement to obtain the depth d of small-sized defects near the surface. n Regarding the surface size r of small defects and the difference in induced voltage U fd The functional relationship; (3) Time-domain measurement yields the functional relationship between attenuation rate k and thickness D; (4) Data fusion of time-domain and frequency-domain methods for the optimization of detection of different types of defects: 1) Calibration parameter U fdmin That is, the minimum voltage difference where near-surface defects exist. During the scanning process, when U fd >U fdmin If the condition is not met, it is determined that there is a near-surface defect at the corresponding location; otherwise, it is determined that there is no near-surface defect. 2) Measuring U in the frequency domain fd >U fdmin The coordinate range is used to obtain the surface size r of the near-surface defect. Let the diameter of the frequency domain receiving coil be d. coil There are two situations: (a) Determine the threshold coefficient h based on the size of the frequency domain receiving coil, if r ≤ hd coil If the near-surface defect is considered to be a small-sized defect, then the near-surface defect depth is d as measured by the frequency domain method. n ; For locations with no defects or only small defects near the surface, the defect depth on the far surface is: d f =D-D t in, d f D represents the depth of the defect on the far surface of the metal pipe, and D represents the standard thickness of the defect-free metal pipe. t The thickness of the metal pipe at this location is obtained by time-domain detection. (b) If r > hd coil If the near-surface defect is large, its depth is calculated based on the time-domain measurement results. d n =D-D t -d f Where, d n D is the depth of near-surface defects in the metal pipe, and D is the standard thickness of the defect-free metal pipe. t d represents the thickness of the metal pipe at this location, as determined by the time-domain method. f This represents the depth of defects on the far surface of the metal pipe.

2. The method for detecting defects in metal pipes according to claim 1, characterized in that, The method for step (2) is as follows: For the metal pipe under test, the changes in the induced voltage of the frequency domain receiving coil at different locations are obtained by scanning. If a near-surface defect exists, the induced voltage difference U of the frequency domain receiving coil will be significant. fd The voltage changes with the scanning coordinates, and the extreme point of the voltage change during the scanning process is the location of the defect center, based on the induced voltage difference U. fd Determine the size of the defect surface by the range of changing coordinates; Let the depth of the near-surface small-sized defect be d. n Its surface size is r. Near-surface small dimensions with different surface sizes and depths are measured, and the corresponding induced voltage difference U is collected. fd Calibration was performed, and the depth d of small-sized defects near the surface was obtained through fitting. n Regarding the surface size r of small defects and the difference in induced voltage U fd The functional relationship.

3. The method for detecting defects in metal pipes according to claim 2, characterized in that, In step (2), the induced voltage difference U of the frequency domain receiving coil is obtained. fd The method is as follows: For measuring the standard thickness of a defect-free metal pipe, let the induced voltage of the frequency domain receiving coil be U. st ; For the metal pipe under test, let the induced voltage of the frequency domain receiving coil be U. f ; The induced voltage difference between the two is U fd =U f -U st .

4. The method for detecting defects in metal pipes according to claim 1, characterized in that, Depth d of small-sized defects near the surface n Regarding the surface size r of small defects and the difference in induced voltage U fd The functional relationship is as follows: In the formula, the constant a ij Let be the fitting coefficients, and m and n be U, respectively. fd And the highest power of r.

5. The method for detecting defects in metal pipes according to claim 1, characterized in that, In the frequency domain method measurement in step (2), the frequency domain coil consists of an excitation coil and a receiving coil. It is scanned along the extension direction of the pipe while maintaining a certain lift height from the pipe wall. During detection, the signal generator provides the excitation coil with a high-frequency sinusoidal alternating signal to excite the harmonic magnetic field, which forms an induced eddy current in the metal pipe being measured. The secondary induced magnetic field generated by the eddy current forms a secondary induced current in the receiving coil, and the amplitude of the induced voltage of the receiving coil is measured.

6. The method for detecting defects in metal pipes according to claim 1, characterized in that, The method for step (3) is as follows: Let the attenuation rate of the induced voltage signal in the later stage in the time domain be k, and the thickness of the metal pipe be D. Measure the metal pipes of different thicknesses, obtain the corresponding attenuation rates, calibrate them, and fit the functional relationship between the attenuation rate k and the thickness D. The attenuation rate k at different locations is obtained by scanning the metal pipe under test using a time-domain method coil. t Then, the thickness D of the metal pipe at different locations can be calculated. t .

7. The method for detecting defects in metal pipes according to claim 1, characterized in that, The functional relationship between attenuation rate k and thickness D in step (3) is as follows: D=g(k)=ak b In the formula, a and b are fitting coefficients.

8. The method for detecting defects in metal pipes according to claim 1, characterized in that, In step (3), the time-domain method coil consists of an excitation coil and a receiving coil nested together, with the excitation coil inside and the receiving coil outside. It maintains a certain lift-off height from the pipe wall and scans along the extension direction of the pipe. During detection, the signal generator provides a low-frequency pulse signal to the excitation coil, which excites a decaying pulse induced magnetic field, forming an induced eddy current in the metal pipe being tested. The secondary induced magnetic field generated by the eddy current forms a secondary induced current in the receiving coil, and the curve of the induced voltage signal of the receiving coil decaying with time is measured.

9. The method for detecting defects in metal pipes according to claim 1, characterized in that, h＝2。

Images