A pulse eddy current thickness evaluation method based on multi-feature fusion

Abstract

This invention relates to the field of nondestructive testing technology, solving the problem of large thickness assessment errors caused by the combined effects of the cladding layer lift-off effect and the shielding effect of the metal protective layer in existing technologies. Specifically, it relates to a pulsed eddy current thickness assessment method based on multi-feature fusion. This method extracts multi-dimensional feature parameters such as centroid time, information diffusion energy, and information diffusion width as inputs, and performs normalization and weighted fusion on each feature to construct a fused feature. This fused feature comprehensively characterizes the time distribution, energy distribution, and diffusion characteristics of the signal, achieving high-precision quantitative assessment of the wall thickness of clad equipment. This invention uses multi-dimensional features such as centroid time, information diffusion energy, and information diffusion width as inputs, and comprehensively characterizes the signal through feature normalization and weighted fusion methods. It fully utilizes the overall information of the pulsed eddy current signal, improving the accuracy and stability of wall thickness assessment, thereby achieving accurate measurement of the wall thickness of clad pipes.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing technology, and in particular to a pulsed eddy current thickness evaluation method based on multi-feature fusion. Background Technology

[0002] Ferromagnetic equipment is widely used in the petroleum, chemical, and power industries, with carbon steel pipelines and pressure vessels being typical examples. Their structural integrity directly affects operational safety, and wall thickness assessment is a crucial aspect of condition monitoring. To reduce heat loss and improve corrosion resistance, the equipment surface is typically covered with an insulation layer and a metal protective layer, forming a clad structure. Pulsed eddy current detection (PECT) shows promising application prospects in wall thickness assessment due to its ability to penetrate thick cladding layers for non-contact far-field detection. However, in practical clad structures, the lift-off effect and the shielding effect caused by the metal protective layer still present a coupling problem, leading to severe signal attenuation, increased measurement errors, and difficulty in achieving high-precision wall thickness assessment. Existing methods often rely on single-point characteristics such as peak value and zero-crossing time, which are easily affected by lift-off variations and shielding effects, resulting in insufficient stability and quantitative capabilities, making it difficult to meet the high-precision detection requirements under complex operating conditions.

[0003] To address the aforementioned issues, it is necessary to construct robust feature parameters based on the overall signal distribution characteristics. However, due to the significant temporal diffusion characteristics of PECT signals and the coupling of multiple influencing factors, feature extraction faces challenges such as weak anti-interference capabilities, insufficient information utilization, and incomplete feature representation. Therefore, constructing multidimensional features based on the centroid, information diffusion energy, and diffusion width, and achieving comprehensive signal representation through feature fusion, is of great significance for mitigating the effects of lift-off, reducing shielding effects, and improving the accuracy of pipe thickness assessment. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a pulsed eddy current thickness evaluation method based on multi-feature fusion, which solves the problem of large evaluation thickness errors caused by the combined effects of the cladding layer lift-off effect and the metal protective layer shielding effect in existing technologies.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a pulsed eddy current thickness evaluation method based on multi-feature fusion, comprising the following steps: S1. Simulation of a coated pipe based on a pulsed eddy current analytical model, equating the coated pipe to a four-layer flat plate structure and establishing an induced voltage signal. Mathematical model; S2. Pulse eddy current testing is performed on a standard specimen of known thickness using square wave excitation, and based on the induced voltage signal. The mathematical model performs differential operations on the detection signal of the receiving coil to obtain the differential signal. ; S3, for differential signals Perform feature extraction to obtain features including the centroid. Information diffusion energy and the breadth of information dissemination Multidimensional features; S4. Perform max-min normalization on the multidimensional features and then perform linear weighting according to the preset weights to obtain the fused features; S5, regarding the centroid Information dissemination width Linear fitting of fusion features and information diffusion energy. By performing quadratic function fitting, the fitting curve equations of multidimensional features and fusion features corresponding to the thickness of the standard specimen are obtained; S6. Place the pulsed eddy current probe on the surface of the pipe to be tested, which is the same material as the standard specimen, and perform pulsed eddy current detection to obtain the multidimensional features and fusion features of the corresponding pipe to be tested. Calculate the true thickness of the pipe to be tested based on the fitted curve equation.

[0006] Furthermore, the four-layer flat plate structure consists of, from bottom to top, the air inside the pipe, the pipe wall, the insulation layer, and the metal protective layer, with the pulsed eddy current probe located above the four-layer flat plate structure.

[0007] Furthermore, the centroid Differential signal The weighted average position in the time domain; the information diffusion energy. Differential signal Energy accumulation; the information diffusion width Differential signal Distribution range in the time domain.

[0008] Furthermore, in step S2, the specific process includes the following steps: S21. Mark a certain area of ​​the standard specimen as the reference area and perform pulsed eddy current testing on it. Use the obtained test signal as the reference signal. ; S22. Mark other areas of the standard specimen as the detection area and perform pulsed eddy current testing on them. The obtained detection signal is used as the detection signal. ; S23, Reference signal With detection signal Differential signal obtained by differential analysis The expression is: ; In the formula, the subscript , These represent the reference area and the detection area, respectively.

[0009] Furthermore, in step S3, the specific process includes the following steps: S31, convert the differential signal Discretization is performed to obtain a discrete signal sequence. ; S32. Calculating discrete signal sequences based on weighted average of signal amplitude. center of mass This is used to describe the center of signal energy distribution on the time axis, and its expression is: ; in, For the first Each sampling time, ; The total number of samples; S33, Based on the centroid The information diffusion width, used to characterize the discreteness of a signal on the time axis, is calculated using the following expression: ; in, For information dissemination width; S34, From discrete signal sequence The information diffusion energy used to characterize the overall energy level of the signal is extracted, and its expression is: ; in, This indicates the energy of signal diffusion.

[0010] Furthermore, the fusion feature The expression is: ; in, , and These are the weighting coefficients corresponding to the centroid, information diffusion energy, and information diffusion width, respectively, and satisfy the following conditions: ; , , These are the normalized centroid, information diffusion energy, and information diffusion width, respectively.

[0011] Furthermore, the weighting coefficients , , .

[0012] Furthermore, the expression for the fitted curve equation is: ; ; ; ; In the formula, Indicates pipe thickness; , , , , , , , , The fitting coefficients are denoted as .

[0013] By employing the above technical solution, the present invention provides a pulsed eddy current thickness evaluation method based on multi-feature fusion, which has at least the following beneficial effects: 1. This invention uses multi-dimensional features such as centroid time, information diffusion energy, and information diffusion width as inputs. It comprehensively characterizes the signal through feature normalization and weighted fusion methods, making full use of the overall information of the pulse eddy current signal, thereby improving the accuracy and stability of wall thickness assessment and achieving accurate measurement of the wall thickness of the cladding layer pipe.

[0014] 2. By constructing a multi-feature fusion mechanism, this invention effectively weakens the coupling effect of the lift-off effect caused by the coating layer and the shielding effect caused by the metal protective layer on the detection signal, thereby enhancing the anti-interference ability of the method and enabling stable and reliable detection results to be obtained even under complex working conditions.

[0015] 3. In the feature extraction process, the present invention adopts normalization processing and weighted fusion strategy, which reduces the dependence on single features, reduces the error caused by different working conditions and signal fluctuations, and improves the applicability and consistency of the method, thereby ensuring the reliability of the thickness evaluation results and the engineering application value. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of the pulsed eddy current analytical model in this invention; Figure 2 This is a schematic diagram of the differential signal in this invention; Figure 3 This is a schematic diagram showing the fitting relationship between the centroid, information diffusion width, information diffusion energy, and fusion characteristics and the pipe thickness in this invention. Figure 4This is a schematic diagram showing the changes in centroid, information diffusion width, information diffusion energy, and fusion characteristics with pipe thickness under different lift-off conditions obtained in this invention. Figure 5 This is a schematic diagram showing the changes in centroid, information diffusion width, information diffusion energy, and fusion characteristics with pipe thickness under different metal protective layer conditions obtained in this invention. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. This will allow for a full understanding of how the present application uses technical means to solve technical problems and achieve technical effects, and to facilitate its implementation.

[0018] Example: This embodiment proposes a pulsed eddy current thickness assessment method based on multi-feature fusion. It extracts multi-dimensional feature parameters such as centroid, information diffusion energy, and information diffusion width as input, and performs normalization and weighted fusion on each feature to construct a fused feature. This fused feature comprehensively characterizes the signal's temporal distribution, energy distribution, and diffusion characteristics, effectively reducing the influence of lift-off and shielding effects, thereby achieving high-precision quantitative assessment of the coating equipment wall thickness. The method includes the following steps: S1. Simulation of a coated pipe based on a pulsed eddy current analytical model, equating the coated pipe to a four-layer flat plate structure and establishing an induced voltage signal. The mathematical model is described in this embodiment. The pulsed eddy current analytical model refers to deriving the electromagnetic diffusion equation based on Maxwell's equations, applying boundary conditions to the multi-layered flat / cylindrical structure, obtaining the analytical solution through Laplace transform (or Fourier transform), and then obtaining the time-domain response induced voltage signal through inverse transform. The theoretical model.

[0019] like Figure 1 As shown, the four-layer flat structure, from bottom to top, consists of the air inside the pipe, the pipe wall, the insulation layer, and the metal protective layer, with the pulsed eddy current probe located above the four-layer flat structure. The large-area corrosion defects on the pipe wall surface are equivalent to uniform wall thickness reduction defects; the pulsed eddy current probe, composed of a coaxially placed hollow excitation coil and a receiving coil, is located in the fifth layer above the four-layer flat structure.

[0020] Furthermore, establish the induced voltage signal The mathematical model in this embodiment includes the following process: With the axis of symmetry of the excitation coil as Axis, with A cylindrical coordinate system is established with the intersection of the axis and the upper surface of the coating layer as the origin. The induced voltage signal of the receiving coil. The expression is: ; ; In the above formula, The imaginary unit; The angular frequency of the harmonic; The amplitude of the harmonic current; Permeability of free space; For the first The eigenvalues ​​are calculated using the truncated region expansion method; This refers to the lift-off distance between the pulsed eddy current probe and the pipe surface. To characterize the probe lift-off response to the induced voltage signal The lift-off coefficient of the effect; The number of eigenvalues; It is a first-order Bessel function of the first kind; The height of the coil; For the first eigenvalues The coil coefficient, characterizing the eigenvalue For induced voltage signal Contributions; It represents the generalized reflection coefficients of the first to fourth layers in a four-layer flat plate structure.

[0021] Among them, the coil coefficient The expression is: ; In the above formula, ; Indicates the coil height; and These represent the inner and outer radii of the coil, respectively. Number of coil turns; subscript and These represent the excitation coil and the receiving coil, respectively.

[0022] Generalized reflectance The expression for is obtained through the following formula: ; ; In the formula, , ; This refers to the number of layers in a four-layer flat panel structure. , , These represent the thickness of the pipe, the insulation layer, and the metal protective layer, respectively. For the first Layer and first Reflection coefficient between layers; For the 1st to the 1st The generalized reflectance coefficient of the layer, ; and These are the first and second type Bessel functions, respectively. The longitudinal wave number of the electromagnetic wave; and The first The magnetic permeability and electrical conductivity of the layer.

[0023] S2. Pulse eddy current testing is performed on a standard specimen of known thickness using square wave excitation, and based on the induced voltage signal. The mathematical model performs differential operations on the detection signal of the receiving coil to obtain the differential signal. .

[0024] The centroid Differential signal The weighted average position in the time domain; the information diffusion energy. Differential signal Energy accumulation; the information diffusion width Differential signal Distribution range in the time domain.

[0025] Furthermore, in step S2, the specific process includes the following steps: S21. Mark a certain area of ​​the standard specimen as the reference area and perform pulsed eddy current testing on it. Use the obtained test signal as the reference signal. ; S22. Mark other areas of the standard specimen as the detection area and perform pulsed eddy current testing on them. The obtained detection signal is used as the detection signal. ; S23, Reference signal With detection signal Differential signal obtained by differential analysis The expression is: ; In the formula, the subscript , These represent the reference area and the detection area, respectively.

[0026] S3, for differential signals Perform feature extraction to obtain features including the centroid. Information diffusion energy and the breadth of information dissemination The multidimensional features. In step S3, the specific process includes the following steps: S31, convert the differential signal Discretization is performed to obtain a discrete signal sequence. ; S32. Calculating discrete signal sequences based on weighted average of signal amplitude. center of mass This is used to describe the center of signal energy distribution on the time axis, and its expression is: ; in, For the first Each sampling time, ; This represents the total number of samples taken.

[0027] This feature reflects the diffusion delay characteristics of eddy currents in a conductor. As the thickness of the pipe increases, the electromagnetic diffusion process is prolonged, and the centroid time usually shows a corresponding change.

[0028] S33, Based on the centroid The information diffusion width, used to characterize the discreteness of a signal on the time axis, is calculated using the following expression: ; in, This refers to the breadth of information dissemination.

[0029] This feature reflects the diffusion range of eddies from the surface to the interior of the material. The greater the thickness, the longer the eddy diffusion path, and the wider the signal distribution on the time axis.

[0030] S34, From discrete signal sequence The information diffusion energy used to characterize the overall energy level of the signal is extracted, and its expression is: ; in, This indicates the energy of signal diffusion.

[0031] This feature has a certain noise suppression effect and can reflect the response intensity of the measured structure to the excitation magnetic field.

[0032] S4. Perform max-min normalization on the multidimensional features and then linearly weight them according to preset weights to obtain the fused features. This embodiment obtains the fused features by adjusting the centroid. Information diffusion energy and the breadth of information dissemination Perform max-min normalization on each feature to obtain the normalized feature parameters. The expression for max-min normalization is: ; in, Represents the original feature quantity. and These represent the maximum and minimum values ​​of the characteristic quantity, respectively.

[0033] normalized center of mass Information diffusion energy and the breadth of information dissemination The fusion features are constructed by linear weighting according to preset weights. The expression is: ; In the formula, , and These are the weighting coefficients corresponding to the centroid, information diffusion energy, and information diffusion width, respectively, and satisfy the following conditions: Specifically, the weighting coefficients , , .

[0034] S5, regarding the centroid Information dissemination width Linear fitting of fusion features and information diffusion energy. By performing quadratic function fitting, the fitting curve equations corresponding to the thickness of the standard specimen are obtained for the multidimensional features and the fusion features, respectively.

[0035] S6. Place the pulsed eddy current probe on the surface of the pipe to be tested, which is the same material as the standard specimen, and perform pulsed eddy current detection to obtain the multidimensional features and fusion features of the corresponding pipe to be tested. Calculate the true thickness of the pipe to be tested based on the fitted curve equation, thereby realizing the assessment of the pipe thickness.

[0036] This embodiment illustrates the process of extracting the centroid, information diffusion width, information diffusion energy, and fusion features of the preprocessed signal obtained in a single sampling period using this method. For example... Figure 2 As shown, the differential signal obtained after differential processing of the detection signal acquired under square wave excitation conditions is presented.

[0037] Verification example: This validation example verifies the high sensitivity of the centroid, information diffusion width, information diffusion energy, and fusion characteristics to changes in pipe thickness. It also verifies the linear correspondence between the centroid, information diffusion width, and fusion characteristics and wall thickness, and the quadratic function relationship of the information diffusion energy. Simultaneously, under conditions of lift-off effects caused by variations in cladding thickness and shielding effects caused by different metal protective layer materials, the anti-interference capability and robustness of the multidimensional characteristics are verified. Furthermore, pipe thickness inversion is performed on a 40.00 mm thick specimen to verify the evaluation accuracy and reliability of the method of this invention.

[0038] In this verification example, the cladding pipe is mostly made of carbon steel with an electrical conductivity of 1.6 MS / m, a relative permeability of 100, and a dielectric constant of 1. The insulation layer has an electrical conductivity of 0 MS / m, a relative permeability of 1, and a dielectric constant of 1. The metal protective layer is made of galvanized iron (with an electrical conductivity of 2.0 MS / m and a relative permeability of 300), aluminum (with an electrical conductivity of 21.6 MS / m and a relative permeability of 1), and stainless steel (with an electrical conductivity of 1.35 MS / m and a relative permeability of 1). The cladding thickness is set to decrease from 10 mm to 0 mm, with the metal cladding thickness set to 1 mm. The pipe wall thicknesses are 20.00 mm, 25.00 mm, 30.00 mm, 35.00 mm, 40.00 mm, 45.00 mm, and 50.00 mm, respectively, with the 50.00 mm pipe thickness serving as a reference thickness. The excitation signal is a square wave current with an amplitude of 4 A, a duty cycle of 50%, and a frequency of 1 Hz. The pulsed eddy current probe consists of an excitation coil and a receiving coil, and the parameters are shown in Table 1.

[0039] Table 1 Parameters of Pulsed Eddy Current Probe

[0040] To verify the relationship between the centroid, information diffusion width, information diffusion energy, fusion characteristics, and pipe thickness in this method, an analytical pulsed eddy current model was used to obtain the induced voltage signal. Pipe thicknesses were set to 20.00 mm, 25.00 mm, 30.00 mm, 35.00 mm, 40.00 mm, 45.00 mm, and 50.00 mm, with 50.00 mm used as the reference thickness. The detected signals at different thicknesses were then differentially processed with the signals at the reference thickness to obtain the differential signal. For example... Figure 2 This indicates that the processed signal waveform exhibits a clear evolution trend with the change in pipe thickness, providing a good foundation for subsequent feature extraction.

[0041] To investigate the quantitative relationship between centroid, information diffusion width, information diffusion energy, fusion characteristics, and pipe thickness, the centroid, information diffusion width, information diffusion energy, and fusion characteristics of specimens with different thicknesses were obtained. Figure 3 It can be seen that the centroid, information diffusion width, and fusion feature monotonically increase with increasing pipe thickness and exhibit a good linear relationship with thickness, indicating that the feature quantities are sensitive to pipe thickness and can be used for quantitative assessment of thickness. Meanwhile, the information diffusion energy exhibits a quadratic function relationship with pipe thickness. Therefore, the expression for the fitted curve equation is: ; ; ; ; In the formula, Indicates the thickness of the pipe.

[0042] Furthermore, to clarify the value range of the fitting coefficients for each fitted curve equation under different coating conditions, this invention performs feature extraction and fitting under different insulation layer thicknesses (0–40 mm, step size 10 mm) and different metal protective layers (galvanized iron, aluminum, and stainless steel, all with a thickness of 1 mm), and obtains the value range of each fitting coefficient as follows: The fitting coefficient range for different insulation layer thicknesses (0–40 mm) is as follows: a∈[291.66,303.65];b∈[-1.82,-1.14];c∈[1369.62,1390.66];d∈[-51.60,-46.95];e∈[1438.34,390691.76];f∈[-6141.11,-373.65];g∈[40.99,41.21];h∈[30.61, 30.65];k∈[17.94,17.98].

[0043] With a 40mm insulation layer, the range of fitting coefficients for different metal protective layer materials is as follows: 0-1mm galvanized iron sheet: a∈[279.17,291.66]; b∈[-4.57,-1.82]; c∈[1363.63,1369.62]; d∈[-60.16,-51.60]; e∈[390691.76,1128479.38]; f∈[-10353.56,-6141.11]; g∈[41.21, 41.45]; h∈[30.65, 30.68]; k∈[17.85, 17.94].

[0044] 0~1mm aluminum: a∈[290.27,291.66]; b∈[-3.33,-1.82]; c∈[1369.62,1385.26]; d∈[-60.16,-54.63]; e∈[390691.76,481154.87]; f∈[-6802.13,-6141.11]; g∈[41.21,41.25]; h∈[30.65,30.68]; k∈[17.91,17.94].

[0045] 0~1mm stainless steel: a∈[291.39,291.66]; b∈[-1.94,-1.82]; c∈[1369.62,1370.39]; d∈[-51.87,-51.60]; e∈[390691.76,441589.66]; f∈[-6527.50,-6141.11]; g∈[41.21,41.21]; h=30.65; k=17.94.

[0046] The above results show that the fitting coefficients of the fusion characteristics remain highly stable under different metal protective layer conditions. Specifically, under galvanized iron sheet conditions, the rate of change of the fitting coefficient h is only 0.10%, and the rate of change of the fitting coefficient k is only 0.50%; under aluminum conditions, the fitting coefficient h remains completely unchanged, and the rate of change of the fitting coefficient k is only 0.34%; under stainless steel conditions, both the fitting coefficients h and k remain completely unchanged. In contrast, the fitting coefficient e of the information diffusion energy changes by as much as 189% under galvanized iron sheet conditions, which is significantly affected by the shielding effect of high magnetic permeability. The above coefficient range provides a clear numerical basis for the selection of adaptive coefficients under different cladding conditions, thus ensuring the reliability and engineering practical value of the method under complex working conditions.

[0047] Fitting with linear equations Figure 3 (a) Figure 3 (b) Figure 3 (c) and Figure 3 The linear relationship shown in (d) is used to calculate the true thickness of the pipe under test based on the range of values ​​for each fitting coefficient determined above, through the fitting curve equation: ; ; ; .

[0048] To evaluate the thickness assessment accuracy of the above-mentioned fitted equation, a pipe with a thickness of 40.00 mm was selected as the verification object. Based on the obtained centroid, information diffusion width, information diffusion energy, and fusion feature values ​​of 0.135, 0.062, 0.0024, and 0.719, respectively, the predicted thicknesses calculated by substituting them into the above equation were 40.07 mm, 39.31 mm, 40.13 mm, and 40.01 mm, respectively. The relative errors were calculated to be 0.175%, 1.725%, 0.325%, and 0.025%, respectively. These errors are within the acceptable range, indicating that the extracted features can achieve quantitative assessment of pipe thickness with high measurement accuracy. Furthermore, the results show that the accuracy of fusion features in thickness assessment is better than that of single features.

[0049] ; In the formula, This is a relative error. This represents the actual pipe thickness.

[0050] To further verify the impact of lift-off variations caused by the coating layer on the characteristic quantities, this verification example uses an insulation layer with an electrical conductivity of 0 MS / m and a magnetic permeability of 1. By varying its thickness (0–40 mm in 10 mm increments), the effect of the lift-off effect caused by changes in the insulation layer thickness on the detection results is studied. Figure 4 (a) Figure 4 (b) Figure 4 (c) and Figure 4 Figure (d) shows the relationship between the centroid, information diffusion width, information diffusion energy, fusion characteristics, and pipe thickness under different lift-off conditions. The results indicate that under different lift-off conditions, each characteristic parameter exhibits a good linear correspondence with the wall thickness, and the coefficient of determination for fitting is high. All exceeded 0.998; among them, the fused features were least affected by the lift-off effect, further verifying the effectiveness of the multi-feature fusion method in suppressing lift-off interference.

[0051] To further verify the impact of the shielding effect caused by the metal protective layer on the characteristic quantities, this verification example, under the condition of no insulation layer and a metal protective layer thickness of 1 mm, selected galvanized iron, aluminum, and stainless steel as the metal protective layer materials for experiments. Figure 5 (a) Figure 5 (b) Figure 5 (c) and Figure 5 As shown in (d), under different metal protective layers, the characteristic parameters exhibit different response characteristics. Stainless steel and aluminum cladding have relatively little impact on the characteristic quantities; however, galvanized iron, due to its high permeability, significantly shields the low-frequency components of the transient electromagnetic field, causing signal changes and significantly affecting the centroid time, information diffusion width, and information diffusion energy. Figure 5 As shown in (d), the fusion feature basically coincides with the pipe thickness curve, indicating that the fusion feature is not sensitive to the presence of the metal protective layer and can effectively weaken the influence of the shielding effect, thereby achieving accurate quantitative assessment of the wall thickness.

[0052] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0053] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. Since the above embodiments are substantially similar to the method embodiments, their descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0054] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for thickness evaluation based on multi-feature fusion of pulsed eddy current, characterized in that, The method includes the following steps: S1. Simulation of a coated pipe based on a pulsed eddy current analytical model, equating the coated pipe to a four-layer flat plate structure and establishing an induced voltage signal. Mathematical model; S2. Pulse eddy current testing is performed on a standard specimen of known thickness using square wave excitation, based on the induced voltage signal. The mathematical model performs differential operations on the detection signal of the receiving coil to obtain the differential signal. ; S3, for differential signals Perform feature extraction to obtain features including the centroid. Information diffusion energy and the breadth of information dissemination Multidimensional features; S4. Perform max-min normalization on the multidimensional features and then perform linear weighting according to the preset weights to obtain the fused features; S5, regarding the centroid Information dissemination width Linear fitting of fusion features and information diffusion energy. By performing quadratic function fitting, the fitting curve equations of multidimensional features and fusion features corresponding to the thickness of the standard specimen are obtained; The expression for the fitted curve equation is: ； ； ； ； In the formula, Indicates pipe thickness; , , , , , , , , These are the fitting coefficients; S6. Place the pulsed eddy current probe on the surface of the pipe to be tested, which is the same material as the standard specimen, and perform pulsed eddy current detection to obtain the multidimensional features and fusion features of the corresponding pipe to be tested. Calculate the true thickness of the pipe to be tested based on the fitted curve equation.

2. The pulsed eddy current thickness evaluation method according to claim 1, characterized in that, The four-layer flat plate structure consists of, from bottom to top, the air inside the pipe, the pipe wall, the insulation layer, and the metal protective layer, with the pulsed eddy current probe located above the four-layer flat plate structure.

3. The pulsed eddy current thickness evaluation method according to claim 1, characterized in that, The centroid Differential signal The weighted average position in the time domain; the information diffusion energy. Differential signal Energy accumulation; the information diffusion width Differential signal Distribution range in the time domain.

4. The pulsed eddy current thickness evaluation method according to claim 3, characterized in that, In step S2, the specific process includes the following steps: S21. Mark a certain area of ​​the standard specimen as the reference area and perform pulsed eddy current testing on it. Use the obtained test signal as the reference signal. ; S22. Mark other areas of the standard specimen as the test area and perform pulsed eddy current testing on them. The obtained test signal is used as the test signal. ; S23, Reference signal With detection signal Differential signal obtained by differential analysis The expression is: ； In the formula, the subscript , These represent the reference area and the detection area, respectively.

5. The pulsed eddy current thickness evaluation method according to claim 1, characterized in that, In step S3, the specific process includes the following steps: S31, convert the differential signal Discretization is performed to obtain a discrete signal sequence. ; S32. Calculating discrete signal sequences based on weighted average of signal amplitude. center of mass This is used to describe the center of signal energy distribution on the time axis, and its expression is: ； in, For the first Each sampling time, ; The total number of samples; S33, Based on the centroid The information diffusion width, used to characterize the discreteness of a signal on the time axis, is calculated using the following expression: ； in, For information dissemination width; S34, From discrete signal sequence The information diffusion energy used to characterize the overall energy level of the signal is extracted, and its expression is: ； in, This indicates the energy of signal diffusion.

6. The pulsed eddy current thickness evaluation method according to claim 1, characterized in that, The fusion feature The expression is: ； in, , and These are the weighting coefficients corresponding to the centroid, information diffusion energy, and information diffusion width, respectively, and satisfy the following conditions: ; , , These are the normalized centroid, information diffusion energy, and information diffusion width, respectively.

7. The pulsed eddy current thickness evaluation method according to claim 6, characterized in that, The weighting coefficient , , .

Images