A method for quantitatively evaluating the debonding of a polyethylene pipe and a related device

Abstract

The application discloses a polyethylene pipe debonding quantitative evaluation method and a related device. The method provided by the application is used for debonding quantitative evaluation of a polyethylene pipe to be detected by using a debonding quantitative evaluation model. The debonding quantitative evaluation model is constructed as follows: firstly, a microwave detection simulation model is constructed, and the debonding depth simulation data and the debonding thickness simulation data are obtained by setting the electromagnetic parameters, the debonding depth range and the debonding thickness range of the polyethylene pipe on the microwave detection simulation model; and the debonding quantitative evaluation model is obtained by training the debonding depth simulation data and the debonding thickness simulation data. The debonding quantitative evaluation model can be used for high-precision nondestructive identification of internal debonding defects of the polyethylene pipe and accurate prediction of the debonding depth and the debonding thickness, and has strong practical application potential in quantitative evaluation of internal debonding of the polyethylene pipe.

Description

Technical Field

[0001] This application belongs to the field of pipeline safety assessment technology, specifically relating to a quantitative assessment method and related apparatus for debonding of polyethylene pipes. Background Technology

[0002] Polyethylene (PE) pipes are widely used in gas transmission and water supply systems due to their high cost-effectiveness, strong corrosion resistance, and excellent durability. They are a key component of modern infrastructure and are of great significance for ensuring the stable operation of infrastructure. However, long-term use of PE pipes can easily lead to hidden internal defects such as debonding and porosity. Traditional metal pipe inspection methods are not applicable due to the limitations of material properties, making it difficult to detect defects early and accurately, which may lead to serious consequences such as leaks and malfunctions.

[0003] Microwave nondestructive testing technology is a feasible solution to this problem. It utilizes the high resolution and strong penetration of high-frequency electromagnetic waves in non-metallic materials to identify internal defects by analyzing the interaction between electromagnetic waves and the material's dielectric constant and magnetic permeability. Furthermore, when combined with the COMSOL finite element simulation model, this technology can simulate the interaction based on the material's physical parameters, thus assisting in achieving accurate detection.

[0004] In existing methods for evaluating polyethylene pipes, on the one hand, the dielectric constant and magnetic permeability of polyethylene pipes according to the GB15558.1-2015 standard are not fully understood, and the influence mechanism of the depth and thickness of debonding inside the pipe on the multidimensional characteristics of microwave signals is not elucidated; on the other hand, existing studies are mostly based on measurement data analysis or the construction of quantitative models using pure microwave S-parameters, which makes it difficult to capture the influence of subtle changes in debonding depth and thickness on the multidimensional characteristic parameters of microwave signals, thus limiting the detection accuracy and quantitative evaluation effect. Summary of the Invention

[0005] In view of the fact that the existing methods for analyzing the debonding defects of buried polyethylene pipes for gas (GB15558.1-2015) are difficult to capture the signal influence of subtle changes in debonding, which limits the detection accuracy and quantitative evaluation effect, this application provides a quantitative evaluation method and related device for debonding of polyethylene pipes.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] The first aspect of this application provides a method for quantitatively assessing the debonding of polyethylene pipes, including:

[0008] Obtain the microwave S-parameters of the polyethylene pipe to be tested;

[0009] Based on the microwave S-parameters of the polyethylene pipe to be tested, a quantitative debonding assessment model is used to obtain the quantitative debonding assessment results of the polyethylene pipe; the quantitative debonding assessment results include debonding depth and debonding thickness.

[0010] The quantitative assessment model for deadhesion is obtained through the following steps:

[0011] A microwave detection simulation model is constructed, wherein the simulation objects of the microwave detection simulation model include a rectangular waveguide measuring device and a polyethylene pipe;

[0012] Parameters are set for the microwave detection simulation model to obtain simulation data of debonding depth and debonding thickness; the parameters include the electromagnetic parameters of the polyethylene pipe, the debonding depth range, and the debonding thickness range; the electromagnetic parameters of the polyethylene pipe are obtained by inversion calculation of the microwave S-parameters of the defect-free polyethylene pipe.

[0013] The quantitative evaluation model for debonding was trained using simulation data of debonding depth and debonding thickness.

[0014] Furthermore, the quantitative assessment model for deadhesion is as follows:

[0015]

[0016] in, This represents the predicted debonding depth of the sample under test. The covariance matrix represents the debonding depth eigenvectors between simulation data with the same debonding thickness but different debonding depths and the test sample. This represents the covariance matrix formed by the eigenvectors of debonding depth among simulation data with the same debonding thickness but different debonding depths. This indicates the debonding depth in the simulation data; This represents the predicted debonding thickness of the sample to be tested; The covariance matrix represents the debonding thickness eigenvectors between simulation data with the same debonding depth but different debonding thicknesses and the test sample. The covariance matrix represents the eigenvectors of debonding thickness from simulation data with the same debonding depth but different debonding thicknesses. For noise variance; This represents the debonding thickness in the simulation data; Represents the identity matrix.

[0017] Furthermore, the electromagnetic parameters of the polyethylene pipe are obtained by inverting the microwave S-parameters of the defect-free polyethylene pipe, specifically as follows:

[0018] exist Within the frequency band, microwave signals of the same duration are used to perform several measurements on defect-free polyethylene pipes of different thicknesses to obtain several sets of microwave S-parameters, which are used as microwave S-parameters of defect-free polyethylene pipes.

[0019] The microwave S-parameters of defect-free polyethylene pipes are inverted and calculated, and the electromagnetic parameters of the polyethylene pipes include the relative permittivity and relative permeability.

[0020] Furthermore, the relative permittivity and relative permeability are obtained in the following manner:

[0021]

[0022] In the formula, Represents the equivalent reflection coefficient; Indicates the reflection coefficient; Indicates the forward transmission coefficient; Indicates the equivalent transmission coefficient;

[0023] The negative refractive index and characteristic impedance are obtained based on the equivalent reflection coefficient and the equivalent transmission coefficient, as shown below:

[0024]

[0025] In the formula, Indicates negative refractive index; Represents the imaginary unit; Represents the free space wavenumber. ,in Wavelength in free space; This indicates the physical thickness of the sample being measured in the direction of electromagnetic wave propagation. Indicates characteristic impedance;

[0026] The relative permittivity and relative permeability are determined based on the negative refractive index and the characteristic impedance, as shown below:

[0027]

[0028] In the formula, Represents the relative permittivity; This represents the relative permeability.

[0029] Furthermore, the method of training the quantitative evaluation model of debonding using simulation data of debonding depth and debonding thickness is specifically as follows:

[0030] The combination of microwave response characteristics was determined based on simulation data of debonding depth and debonding thickness.

[0031] For each sample, a debonding depth feature vector and a debonding thickness feature vector are constructed based on the combination of microwave response features.

[0032] The debonding depth feature vector and debonding thickness feature vector of all samples constitute the training set, which is used to train the quantitative evaluation model of debonding.

[0033] Furthermore, the optimal combination of microwave response features for each sample is determined through cross-validation.

[0034] Furthermore, the microwave response characteristic combination includes return loss, Phase, Phase, impedance, group delay, and standing wave ratio.

[0035] A second aspect of this application provides a quantitative assessment system for debonding of polyethylene pipes, comprising:

[0036] Evaluation parameter acquisition unit, used to acquire microwave S-parameters of the polyethylene pipe to be tested;

[0037] The debonding quantitative assessment unit is used to obtain the debonding quantitative assessment result of the polyethylene pipe based on the microwave S-parameters of the polyethylene pipe to be tested using the debonding quantitative assessment model; the debonding quantitative assessment result includes debonding depth and debonding thickness.

[0038] The quantitative assessment model for deadhesion is obtained through the following steps:

[0039] A microwave detection simulation model is constructed, wherein the simulation objects of the microwave detection simulation model include a rectangular waveguide measuring device and a polyethylene pipe;

[0040] Parameters are set for the microwave detection simulation model to obtain simulation data of debonding depth and debonding thickness; the parameters include the electromagnetic parameters of the polyethylene pipe, the debonding depth range, and the debonding thickness range; the electromagnetic parameters of the polyethylene pipe are obtained by inversion calculation of the microwave S-parameters of the defect-free polyethylene pipe.

[0041] The quantitative evaluation model for debonding was trained using simulation data of debonding depth and debonding thickness.

[0042] A third aspect of this application provides an electronic device, comprising: a memory and one or more processors; the memory being coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the steps of the polyethylene pipe debonding quantitative assessment method as described above.

[0043] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the quantitative assessment method for debonding of polyethylene pipes as described above.

[0044] Compared with the prior art, this application has the following beneficial technical effects:

[0045] This application combines microwave detection technology with finite element simulation. First, the electromagnetic parameters of polyethylene pipe material (GB15558.1-2015) are obtained by inverting experimentally measured microwave S-parameters. These inverted electromagnetic parameters are then embedded into the microwave detection simulation model, ensuring a precise match between the simulated microwave S-parameters and experimental measurement data. This allows for the construction of a high-fidelity finite element simulation model that accurately reflects the electromagnetic response characteristics of polyethylene pipes. Based on this simulation model, this application further systematically analyzes the influence mechanism of changes in the debonding depth (D) and debonding thickness (H) of internal debonding defects in polyethylene pipes on the multidimensional characteristic parameters of the microwave response signal. It extracts highly sensitive feature vectors for debonding depth and thickness, and constructs a quantitative assessment model for debonding based on multidimensional microwave response characteristics. Experimental verification shows that this model possesses high prediction accuracy and successfully achieves high-precision non-destructive identification of internal debonding defects in polyethylene pipes and accurate prediction of geometric parameters (depth and thickness). It has strong practical application potential in the field of quantitative assessment of internal debonding defects in polyethylene pipes. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 Flowchart of the quantitative assessment method for debonding of polyethylene pipes provided in this application;

[0048] Figure 2 Flowchart of the quantitative assessment model for debonding constructed for this application;

[0049] Figure 3 This is a schematic diagram of the microwave testing system used for evaluating polyethylene pipes in this application;

[0050] Figure 4 This is a schematic diagram of the rectangular waveguide measurement device provided in this application;

[0051] Figure 5The curves showing the real and imaginary parts of the relative permittivity as a function of frequency for polyethylene pipes of different thicknesses in this application are shown, where (a) represents the real part of the relative permittivity and (b) represents the imaginary part of the relative permittivity.

[0052] Figure 6 A microwave detection simulation model was constructed for this application;

[0053] Figure 7 The figures show the simulated and measured microwave S-parameter curves for polyethylene pipe samples M1-M6 in this application. The left side of rows 1-6 represents the frequency range of samples M1-M6 from 12GHz to 18GHz. Comparison curve of simulated and measured parameter values; the right side represents the sample values ​​respectively. Comparison curve of simulated and measured parameter values;

[0054] Figure 8 The above are the characteristic curves of microwave response when the debonding depth is constant at 1.0 mm according to the embodiments of this application. (a) to (f) represent the return loss, Phase, Characteristic curves of the average normalized parameters of phase, impedance, group delay and VSWR as a function of thickness in the range of 0.1 mm to 3.0 mm;

[0055] Figure 9 The above are the characteristic curves of microwave response when the debonding thickness is constant at 0.2 mm according to the embodiments of this application. (a) to (f) represent the return loss, Phase, Characteristic curves of the average normalized parameters of phase, impedance, group delay and VSWR as a function of depth in the range of 0.5 mm to 3.6 mm;

[0056] Figure 10 The simulated and measured microwave S-parameter curves are shown for a polyethylene pipe sample with a debonding depth of 1.0 mm and a debonding thickness of 0.2 mm according to an embodiment of this application. Where (a) represents... Curve (b) represents curve;

[0057] Figure 11 The evaluation results of the debonding thickness of the polyethylene pipe in the embodiments of this application;

[0058] Figure 12 The evaluation results of the debonding depth of the polyethylene pipe in the embodiments of this application;

[0059] Figure 13 This is a schematic diagram of the structure of the polyethylene pipe debonding quantitative assessment system according to a preferred embodiment of this application;

[0060] Figure 14 This is a schematic diagram of the electronic device structure according to a preferred embodiment of this application;

[0061] In the figure, 1 is a rectangular waveguide measuring device; 2 is a polyethylene pipe. Detailed Implementation

[0062] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0063] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0064] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0065] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0066] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0067] First, let's introduce the relevant terminology used in the embodiments of this application:

[0068] A frequency band is a portion of the microwave frequency range in the electromagnetic spectrum, typically between 12.0 GHz and 18 GHz.

[0069] The Nicolson-Ross-Weir (NRW) method is a technique for extracting the electromagnetic properties of materials from scattering parameters (S-parameters), and it is widely used in microwave engineering and electromagnetics. Its core principle is to invert the equivalent permittivity and equivalent permeability of a material by measuring its reflection and transmission characteristics.

[0070] Mapminmax normalization is a function in MATLAB used for data normalization. Its core function is to scale data row by row to a specified range (usually [-1,1] or [0,1]) to eliminate the difference in units between different features. It is suitable for scenarios such as pattern recognition and machine learning.

[0071] Polyethylene (PE) pipes are a vital component of modern infrastructure, widely used in gas transmission and water supply systems due to their cost-effectiveness, corrosion resistance, and superior durability. Despite these advantages, the structural integrity of PE pipes can be compromised by long-term operational stress, environmental impacts, and potential manufacturing defects, leading to internal defects such as debonding and porosity, which pose significant challenges for detection and quantification. Traditional inspection methods for metallic pipes are no longer applicable due to the non-metallic nature of PE pipes and the embedding of defects within the material. Unlike surface defects, these internal anomalies are often difficult to detect precisely until they cause serious consequences such as leaks, pressure losses, or catastrophic failures. Therefore, early identification of internal debonding in PE pipes is crucial for preventing such failures and ensuring the long-term reliability of PE pipelines. Traditional non-destructive testing (NDT) methods, including ultrasonic testing (UT) and X-ray imaging, have limited sensitivity to early millimeter-scale defects in non-metallic materials such as PE. Therefore, there is an urgent need for a more accurate quantitative inspection and assessment method capable of effectively detecting cavity-like defects within PE pipes, enabling more proactive pipeline health management.

[0072] Microwaves are high-frequency electromagnetic waves with high resolution and strong penetrating power in non-metallic materials such as polyethylene, enabling the detection of internal defects in pipes, including debonding defects. Therefore, microwave non-destructive testing technology holds promise as a solution to this challenge and is well-suited for detecting hidden internal defects in polyethylene pipes. The working principle of microwave technology is to analyze the interaction between electromagnetic waves and material properties such as dielectric constant and permeability. The presence of defects in the pipe alters these properties, allowing for the detection of these changes and the identification of the defect's location and geometry.

[0073] In the research on debonding detection of polyethylene pipes, the combination of microwave non-destructive testing (MNDT) and COMSOL finite element simulation model is an important method for achieving accurate defect detection. Microwave testing technology has the advantages of non-destructiveness and strong penetration for non-metallic materials, and can effectively detect internal defects in materials. COMSOL finite element simulation, based on the physical parameters of the material, can predict and analyze the electromagnetic response characteristics of the material by simulating the interaction between microwaves and the material.

[0074] Despite some progress in recent years, the application of microwave nondestructive testing (NDT) technology in defect detection of polyethylene (PE) pipes still faces several key challenges. The differences in the electrical properties of various composite materials affect the accurate quantification of their internal defects, and the dielectric constant and magnetic permeability of PE pipes conforming to the GB15558.1-2015 standard are not yet fully understood. Furthermore, the mechanisms by which the debonding depth (D) and debonding thickness (H) within PE pipes affect the multidimensional characteristics of microwave signals are not fully elucidated. These issues hinder the quantitative characterization of PE pipe debonding and pose challenges to inverse modeling. Existing research mainly relies on measurement data analysis or quantitative models built based on pure S-parameters, making it difficult to capture the influence mechanism of subtle changes in debonding depth (D) and debonding thickness (H) on multidimensional characteristic parameters. These limitations in measurement data and modeling ultimately restrict the detection accuracy and quantitative assessment of internal debonding. Therefore, this application uses measured data to inversely deduce the electromagnetic parameters of the material, thereby accurately constructing a high-fidelity finite element simulation model that conforms to the actual influence of debonding defects on microwave signals. In the simulation model, the variation law of debonding depth and thickness was further studied, and a quantitative evaluation method for debonding of polyethylene pipes that integrates actual measurement and finite element simulation was proposed to address these challenges.

[0075] like Figure 1 As shown, based on this, this application provides a method for quantitatively assessing the debonding of polyethylene pipes, including:

[0076] S1, obtain the microwave S-parameters of the polyethylene pipe to be tested;

[0077] S2, based on the microwave S-parameters of the polyethylene pipe to be tested, a quantitative debonding assessment model is used to obtain the quantitative debonding assessment results of the polyethylene pipe; the quantitative debonding assessment results include debonding depth and debonding thickness;

[0078] Among them, such as Figure 2 As shown, the quantitative assessment model for deadhesion is obtained through the following steps:

[0079] S201, Construct a microwave detection simulation model, wherein the simulation objects of the microwave detection simulation model include a rectangular waveguide measuring device and a polyethylene pipe;

[0080] S202, set parameters for the microwave detection simulation model to obtain simulation data of debonding depth and debonding thickness; the parameters include the electromagnetic parameters of the polyethylene pipe, the debonding depth range, and the debonding thickness range; the electromagnetic parameters of the polyethylene pipe are obtained by inversion calculation of the microwave S-parameters of the defect-free polyethylene pipe.

[0081] S203 uses simulation data of debonding depth and debonding thickness to train the quantitative evaluation model of debonding.

[0082] This application obtains the true electromagnetic parameters of polyethylene pipes through inversion of measured data, and constructs a high-fidelity microwave detection simulation model based on this. Based on the microwave detection simulation model, a quantitative evaluation model for debonding based on multi-feature selection is constructed, which achieves high-precision prediction of the debonding depth and debonding thickness of debonding defects in polyethylene pipes.

[0083] The electromagnetic parameters of the polyethylene pipe were obtained by inversion calculation of the microwave S-parameters of the defect-free polyethylene pipe. The specific acquisition process is as follows:

[0084] To ensure that the measured performance of the polyethylene pipe samples is representative of actual engineering applications, the polyethylene pipes used were sourced from buried polyethylene pipes for gas applications conforming to the GB15558.1-2015 standard. The preparation process of the polyethylene pipe samples was strictly controlled, employing CNC (Computerized Numerical Control) precision cold machining to prevent any thermal or mechanical deformation that could alter the dielectric properties. This machining method ensured tight tolerances, minimized thickness variations, and maintained the uniformity of the material properties of the polyethylene pipe samples.

[0085] To accurately characterize the transmission and reflection characteristics of polyethylene pipe samples within the target frequency range, this application establishes a high-resolution microwave scanning and measurement system. For example... Figure 3 As shown, the system includes a Keysight ENA E5080B Vector Network Analyzer (VNA), a low-noise coaxial cable, a coaxial waveguide converter, a custom open rectangular waveguide (RWG) measurement fixture, a data fusion processing unit, and a finite element simulation unit. Figure 4 As shown, a defect-free polyethylene pipe 2 is fitted and placed inside a rectangular waveguide measuring device 1, and then a vector network analyzer is used to measure the pipe. The frequency band is used to repeatedly measure each sample multiple times with a microwave signal of fixed step size to obtain several sets of discrete microwave S-parameters. The microwave S-parameters include and , Indicates the reflection coefficient; This represents the forward transmission coefficient.

[0086] In some specific implementations, six PE pipe samples with different thicknesses (1mm, 1.5mm, 2.0mm, 2.5mm, 3.0mm, 3.5mm) were obtained after CNC precision cold machining pretreatment. Each sample had a length and width of 15.799mm × 7.899mm and was numbered M1 to M6, with three samples for each number, as shown in Table 1. Using a vector network analyzer, 90 sets of discrete microwave S-parameters were obtained by repeatedly measuring each sample with the same number five times in the 12GHz–18GHz frequency band with a microwave signal step size of 0.0305GHz.

[0087] Table 1. Geometric parameters of defect-free polyethylene pipe samples.

[0088]

[0089] In some specific implementations, noise reduction processing is performed on the microwave S-parameters of defect-free polyethylene pipes. Specifically, this involves averaging several sets of discrete microwave S-parameters to obtain the average value of the microwave S-parameters for each polyethylene pipe sample as the target data, thereby reducing the unstable fluctuations in the original data caused by measurement errors. The averaging formula is as follows:

[0090]

[0091] in, Indicates a certain sample in At frequency point or average value; This represents the average of the real parts; Represents the imaginary unit; This represents the average value of the imaginary part.

[0092] In some specific embodiments, the electromagnetic parameters of the polyethylene pipe are obtained by inverting the microwave S-parameters of the defect-free polyethylene pipe. Specifically, the electromagnetic parameters are obtained by inverting the denoised microwave S-parameters using the NRW algorithm, including the following calculation process:

[0093] Considering the multiple reflection effect produced by polyethylene samples with debonding defects, the reflection coefficient and transmission coefficient are redefined as follows:

[0094] (1)

[0095] (2)

[0096] in, Represents the equivalent reflection coefficient; Indicates the reflection coefficient; Indicates the forward transmission coefficient; This represents the equivalent transmission coefficient.

[0097] Then, the negative refractive index and characteristic impedance are extracted based on the equivalent reflection coefficient and equivalent transmission coefficient, as shown below:

[0098] (3)

[0099] (4)

[0100] In the formula, Indicates negative refractive index; Represents the imaginary unit; Represents the free space wavenumber. ,in Wavelength in free space; This indicates the physical thickness of the sample being measured in the direction of electromagnetic wave propagation. This represents the characteristic impedance.

[0101] The relative permittivity and relative permeability are determined based on the negative refractive index and characteristic impedance, as shown below:

[0102] (5)

[0103] (6)

[0104] In the formula, Represents the relative permittivity; This represents the relative permeability.

[0105] The microwave S-parameters of 90 groups of polyethylene pipes were processed using the NRW algorithm described above, such as... Figure 5 As shown, the preliminary relationship between the real and imaginary parts of the relative permittivity and the frequency and the thickness of the polyethylene pipe was obtained, where, Figure 5 In figure (a), the relationship between the real part of the relative permittivity and the frequency and the thickness of the polyethylene pipe is shown. Figure 5 In diagram (b), the relationship between the real part of the relative permittivity and the frequency and thickness of the polyethylene pipe is shown. The relative permittivity of the polyethylene pipe used in this embodiment was then calculated. relative permeability .

[0106] In some specific implementations, COMSOL finite element simulation is used to simulate and construct a microwave detection simulation model of a rectangular waveguide measurement device and a polyethylene pipe in a real-world scenario, such as... Figure 6 As shown, its geometric dimensions are the same as those of the experimental measuring device.

[0107] In some specific implementations, parameters are set for the microwave detection simulation model to obtain simulation data on debonding depth and debonding thickness. The specific process is as follows:

[0108] First, the electromagnetic parameters obtained from the inversion calculation are assigned to the polyethylene pipe in the microwave detection simulation model.

[0109] To verify the response of a microwave testing simulation model to microwave signals using electromagnetic parameters of defect-free polyethylene pipes, a microwave signal with the same frequency band and step size as the actual measurement was used to simulate the testing of the polyethylene pipes. Multiple sets of simulated microwave S-parameters for different thicknesses were obtained and compared with measured microwave S-parameters to verify the authenticity of the microwave testing simulation model and the reliability of the electromagnetic parameters. In microwave nondestructive testing, an average amplitude error between simulated and measured microwave S-parameters of ≤0.5dB is considered excellent; ≤0.5~1.0dB is generally considered a good match; 1~2dB is still acceptable when material loss, assembly, and calibration tolerances exist. The relative permittivity is... relative permeability A polyethylene pipe was defined as a model for microwave detection. Simulated microwave S-parameters were obtained under microwave signals of the same frequency band (12GHz~18GHz) and step size (0.0305GHz), and compared with the measured microwave S-parameter curves. Figure 7 As shown, the simulated microwave S-parameters and the measured microwave S-parameters generally show a consistent trend, with an error within 0.5 dB. This verifies that the constructed microwave detection simulation model can accurately reflect the microwave response of actual defect-free polyethylene pipes.

[0110] Then, in the microwave detection simulation model, the debonding depth range and debonding thickness range are set, and simulation data of debonding depth and debonding thickness are obtained respectively.

[0111] Two sets of simulation data were used to study the influence mechanism of debonding defects on the microwave S-parameters of polyethylene pipes. In the microwave detection simulation model, the debonding depth of the polyethylene pipe was set to vary within the range of 0.5 mm to 3.6 mm, and 28 sets of simulation data on debonding depth were obtained. Similarly, the debonding thickness of the polyethylene pipe was set to vary within the range of 0.1 mm to 3.0 mm, and 28 sets of simulation data on debonding thickness were obtained.

[0112] Based on simulation data of debonding depth, the correlation between microwave response characteristics and debonding depth is analyzed. Based on simulation data of debonding thickness, the correlation between microwave response characteristics and debonding thickness is analyzed to determine the combination of microwave response characteristics used for training the quantitative evaluation model of debonding. Microwave response characteristics include return loss (RL). Phase ( ), Phase ( ),impedance( Group delay () The six microwave response characteristics are: 1) microwave wave ratio (SWR).

[0113] The mechanism between the normalized parameters of the average value of the microwave response characteristics and the changes in debonding depth or debonding thickness was analyzed. Based on the physical sensitivity of the microwave response characteristics to changes in the internal structure of polyethylene pipes, a combination of microwave response characteristics was determined for training the quantitative assessment model of debonding, providing a reliable basis for subsequent debonding quantification and inversion modeling. The calculation formulas for the six microwave response characteristics and their average values ​​are as follows:

[0114] (7)

[0115] (8)

[0116] (9)

[0117] (10)

[0118] (11)

[0119] (12)

[0120] (13)

[0121] (14)

[0122] (15)

[0123] (16)

[0124] in, Indicates in Return loss at a given frequency point; Indicates in Reflection coefficient at a given frequency; This represents the average return loss of a single sample. Indicates the number of samples; Indicates in Frequency point Phase or Phase; Indicates in Frequency point or ; The average of a single sample Phase; The average of a single sample Phase; Indicates in Impedance at a given frequency; Indicates the reference characteristic impedance; This represents the average impedance of a single sample; Indicates in Group delay at a given frequency point; exist A little below the frequency Phase or Phase; and Indicates different frequency points; This represents the average group delay for a single sample; Indicates in VSWR at a given frequency; It represents the average standing wave ratio of a single sample.

[0125] like Figure 8 As shown, when the debonding depth in the simulation data is constantly 1.0 mm, the debonding thickness is in the range of 0.1 mm to 3.0 mm. Figure 8 From (a) to (f), we can see the return loss. Phase, The average normalized parameters of phase, impedance, group delay, and VSWR are strongly correlated with the debonding thickness; for example... Figure 9 As shown, when the debonding thickness in the simulation data is constantly 0.2 mm, the debonding depth is in the range of 0.5 mm to 3.6 mm. Figure 9 From (a) to (f), we can see the return loss. Phase, The average normalized parameters of phase, impedance, group delay, and VSWR show a strong correlation with the depth of debonding, highlighting the physical sensitivity of the selected microwave response characteristics to changes in the internal structure of polyethylene pipes, and providing a reliable basis for debonding quantification and modeling. Therefore, the combination of microwave response characteristics used for training the quantitative evaluation model of debonding in this application includes return loss, Phase, Phase, impedance, group delay, and standing wave ratio.

[0126] For each sample, a debonding depth feature vector and a debonding thickness feature vector are constructed based on the combination of microwave response features, as follows:

[0127] From the Six microwave response feature vectors were extracted from the simulated debonding depth and debonding thickness data of each sample. These features collectively constitute the debonding depth feature vector or debonding thickness feature vector of a single sample. Specifically, the average value of each microwave response feature vector in a single sample was calculated, and this average value was expressed as the debonding depth feature vector or debonding thickness feature vector of the corresponding single sample.

[0128]

[0129] In the formula, This represents the debonding depth feature vector or debonding thickness feature vector for a single sample. , , , , , Use respectively , , , , , Correspondence representation; Indicates the first sample, This represents the total number of samples.

[0130] To mitigate the impact of scale differences between different feature dimensions on the construction of a quantitative assessment model for deadhesion, this application adopted... Normalization, The average eigenvalues ​​of each sample are linearly scaled to the range [0,1]. The normalization parameters used to construct the quantitative assessment model for deadhesion are determined by simulation data to ensure the consistency and comparability of all input data. For the ... Six microwave response feature vectors corresponding to each sample The normalized calculation expression is as follows:

[0131] (18)

[0132] in, Represented as the first The first sample Normalized values ​​of microwave response eigenvectors; Represented as the first The first sample Each microwave response feature vector; The first and second training samples are respectively the first and second training samples. The minimum and maximum values ​​of a microwave response eigenvector.

[0133] right After normalizing the samples, we get a 6-dimensional microwave response eigenvector matrix:

[0134] (19)

[0135] in, Indicates to The microwave response feature vector matrix is ​​obtained by normalizing the samples. Each row corresponds to one sample, and each column corresponds to one microwave response feature vector. This indicates that the matrix belongs to a dimension of The space of real matrix numbers.

[0136] The optimal combination of microwave response features for each sample was determined through cross-validation. Five-fold cross-validation was used to obtain the optimal combination of microwave response features, which was then used to train the quantitative debonding assessment model, as detailed below:

[0137] To rigorously evaluate the predictive performance of each microwave response feature combination and reduce the risk of overfitting, this application employs a five-fold cross-validation strategy for the debonding depth simulation data and the debonding thickness simulation data. The debonding depth simulation data and the debonding thickness simulation data are each randomly divided into five mutually exclusive subsets of approximately equal size, denoted as follows: For each subset subset Used as the simulation verification set, the other four subsets Combinations form the simulation training set. After excluding a degenerate subset, the total number of combinations is... Then, all candidate subsets are evaluated. , These represent the six microwave response feature vectors after normalization of a given sample; subset The dimensions range from 3 to 6. For each subset, a multidimensional feature quantization model of Gaussian Process Regression (GPR) is trained and validated using a five-fold cross-validation procedure; then, the root mean square error (RMSE) of the five subsets is calculated to evaluate the predictive performance of each microwave response feature combination. Prediction accuracy is quantized using RMSE; for the , A subset, defined as:

[0138] (20)

[0139] in, Indicates the first The root mean square error of each subset; The simulation verification set is represented by the first... Predicted values ​​for debonding thickness or depth for each sample. For the simulation training set The debonding thickness or depth value of each sample For the first The number of samples in the simulation validation set is determined. This process is repeated five times, with each subset serving as a validation set. After all iterations are completed, the average RMSE of the five iterations can be calculated using equation (21). To determine the optimal microwave response feature subset, the subset that produces the minimum average root mean square error is selected according to the criteria defined in equation (22). Feature combination .

[0140] (twenty one)

[0141] (twenty two)

[0142] Using the aforementioned five-fold cross-validation method, the model was trained on both the debonding depth simulation dataset and the debonding thickness simulation dataset to determine the optimal combination of microwave response features for each sample. Based on this optimal combination of microwave response features, the debonding depth feature vector and the debonding thickness feature vector were determined and used to train the quantitative debonding assessment model.

[0143] In some specific embodiments, this application employs a five-fold cross-validation strategy for the debonding depth simulation data and the debonding thickness simulation data. By training the debonding depth simulation data and the debonding thickness simulation data, the results shown in Table 2 are obtained. It can be seen that the RMSE value obtained based on the analysis of the four microwave response characteristics is the smallest. Therefore, the optimal combination of microwave response characteristics is shown in equations (23) and (24):

[0144]

[0145] In the formula, The optimal combination of microwave response characteristics representing the debonding depth; The optimal combination of microwave response characteristics represents the debonding thickness.

[0146] Table 2. Test results of the optimal combination of microwave response characteristics

[0147]

[0148] In the table, Represents the total number of combinations; The mean square error representing the debonding depth; This represents the average root mean square error of the debonding thickness.

[0149] To achieve quantitative detection of debonding, this application uses the debonding depth feature vector and debonding thickness feature vector of all samples in the simulation data to form a training set, and trains a multidimensional feature quantification model based on Gaussian process regression to obtain a quantitative evaluation model for debonding. The training objective is to learn a latent mapping function from the training data. and noise variance To account for the inherent noise in the experimental measurements, an additive Gaussian noise term is introduced. To explain the observed target With latent function The difference between them is shown in Equation (25), which enables the model to capture the potential functional relationships and associated uncertainties in the observed target.

[0150] (25)

[0151] in This represents the debonding depth feature vector or debonding thickness feature vector of the i-th sample, which can be... or ; Indicates the debonding depth or debonding thickness. Assume they are independent and identically distributed.

[0152] To measure the similarity between input data points, this application uses a radial basis function (RBF) kernel as the covariance function in the Gaussian process regression model. The RBF kernel is defined as shown in equation (26). This kernel function enables the Gaussian process regression model to flexibly capture nonlinear relationships while maintaining a smooth response surface. After training on two datasets corresponding to different debonding depth feature vectors and debonding thickness feature vector datasets, two different sets of optimal kernel parameters were obtained, as shown in Table 3.

[0153] (26)

[0154] in, and This represents the debonding depth feature vector or the debonding thickness feature vector of two samples. It is the overall scale of the signal variance output by the control function. It is the length scale parameter that controls the smoothness of the function.

[0155] Table 3. Hyperparameters obtained from training the quantitative evaluation model for deadhesion.

[0156]

[0157] for For each simulated sample, a covariance matrix is ​​constructed by calculating the eigenvectors between each pair of simulated data. Similarly, for The covariance matrix formed by the eigenvectors of the simulated data and the predicted data for each predicted sample is: .

[0158]

[0159] Based on these covariance matrices and the hyperparameters corresponding to the quantitative debonding assessment model, a quantitative debonding assessment model is constructed, as shown below:

[0160]

[0161] in, This represents the predicted debonding depth of the sample under test. The covariance matrix represents the debonding depth eigenvectors between simulation data with the same debonding thickness but different debonding depths and the test sample. This represents the covariance matrix formed by the eigenvectors of debonding depth among simulation data with the same debonding thickness but different debonding depths. This indicates the debonding depth in the simulation data; This represents the predicted debonding thickness of the sample to be tested; The covariance matrix represents the debonding thickness eigenvectors between simulation data with the same debonding depth but different debonding thicknesses and the test sample. The covariance matrix represents the eigenvectors of debonding thickness from simulation data with the same debonding depth but different debonding thicknesses. For noise variance; This represents the debonding thickness in the simulation data; Represents the identity matrix.

[0162] In some specific implementations, based on the set debonding depth and debonding thickness ranges of the debonding defects, two sets of polyethylene pipe samples were processed for actual measurement. One set had different debonding depths, and the other set had different debonding thicknesses. The sample scheme with debonding defects was as follows: one set consisted of six samples, numbered S1 to S6, with debonding thicknesses (H) of 0.2mm, 0.7mm, 1.2mm, 1.7mm, 2.2mm, and 2.7mm respectively, while keeping the debonding depth (D) constant at 1.0mm; this set served as the measured dataset for debonding thickness. The other set consisted of six samples, numbered N1 to N6, with debonding depths (D) of 1.0mm, 1.5mm, 2.0mm, 2.5mm, 3.0mm, and 3.5mm respectively, while keeping the debonding thickness (H) constant at 0.2mm; this set served as the measured dataset for debonding depth. The simulated microwave S-parameters of N1 (or S1) were compared with the measured microwave S-parameters, as follows... Figure 10 As shown; from Figure 10As can be seen from (a) and (b), the microwave response trend caused by debonding changes has a high degree of consistency between simulation and actual measurement, and the average amplitude error remains within the range of [missing data]. The results verified that the constructed microwave detection simulation model can realistically and accurately reflect the impact of debonding defects in actual polyethylene pipes.

[0163] The predictive performance of the trained quantitative debonding assessment models based on multidimensional microwave response characteristics was further evaluated: one model was used to predict the debonding depth (D), and another was used to predict the debonding thickness (H). The debonding depth prediction model trained using debonding depth simulation data was used to predict the debonding depth of samples N1~N6. Figure 11 As shown in Table 4, the predicted values ​​and measured values ​​are in excellent agreement, with a prediction accuracy between 94.54% and 97.16%, and a root mean square error (RMSE) of 0.0819 mm. This indicates that the debonding depth prediction model has strong accuracy and consistency for different debonding depths. Subsequently, the debonding thickness prediction model trained using debonding thickness simulation data was used to predict the debonding thickness of samples S1 to S6. Figure 12 As shown in Table 5, the prediction accuracy of the debonding thickness prediction model ranges from 72.50% to 99.682%, with a corresponding root mean square error (RMSE) of 0.0580 mm. The quantization accuracy is slightly lower for a debonding thickness of 0.2 mm, due to the limited training samples within this range and the inherent challenge of detecting extremely shallow defects. However, the prediction accuracy significantly improves with increasing debonding thickness. Notably, the prediction performance obtained in this application surpasses that of previously reported pure S-parameter quantization models, demonstrating the superior robustness and generalization ability of the proposed multidimensional microwave response feature fusion method in accurately quantifying internal debonding defects in polyethylene pipes.

[0164] Table 4. Evaluation results of debonding depth of PE pipe samples

[0165]

[0166] Table 5. Evaluation results of debonding thickness of PE pipe samples

[0167]

[0168] In some specific implementations, the microwave S-parameters of the polyethylene pipe to be tested are preprocessed in the same way as the microwave S-parameters of the defect-free polyethylene pipe in the construction of the debonding quantitative assessment model.

[0169] like Figure 13 As shown, another objective of this application is to provide a quantitative assessment system for debonding of polyethylene pipes, comprising:

[0170] Evaluation parameter acquisition unit, used to acquire microwave S-parameters of the polyethylene pipe to be tested;

[0171] The debonding quantitative assessment unit is used to obtain the debonding quantitative assessment result of the polyethylene pipe based on the microwave S-parameters of the polyethylene pipe to be tested using the debonding quantitative assessment model; the debonding quantitative assessment result includes debonding depth and debonding thickness.

[0172] The quantitative assessment model for deadhesion is obtained through the following steps:

[0173] A microwave detection simulation model is constructed, wherein the simulation objects of the microwave detection simulation model include a rectangular waveguide measuring device and a polyethylene pipe;

[0174] Parameters are set for the microwave detection simulation model to obtain simulation data of debonding depth and debonding thickness; the parameters include the electromagnetic parameters of the polyethylene pipe, the debonding depth range, and the debonding thickness range; the electromagnetic parameters of the polyethylene pipe are obtained by inversion calculation of the microwave S-parameters of the defect-free polyethylene pipe.

[0175] The quantitative evaluation model for debonding was trained using simulation data of debonding depth and debonding thickness.

[0176] Please see Figure 14 As shown, this application also provides an electronic device 100 for implementing a quantitative assessment method for debonding of polyethylene pipes; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0177] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the quantitative assessment method for debonding of polyethylene pipes by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device 100 (such as audio data), etc. In addition, the memory 101 may include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0178] The at least one processor 102 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor. The processor 102 is the control center of the electronic device 100, connecting various parts of the electronic device 100 via various interfaces and lines.

[0179] The memory 101 in the electronic device 100 stores multiple instructions to implement a quantitative assessment method for debonding of polyethylene pipes, and the processor 102 can execute the multiple instructions to achieve the following:

[0180] Obtain the microwave S-parameters of the polyethylene pipe to be tested;

[0181] Based on the microwave S-parameters of the polyethylene pipe to be tested, a quantitative debonding assessment model is used to obtain the quantitative debonding assessment results of the polyethylene pipe; the quantitative debonding assessment results include debonding depth and debonding thickness.

[0182] The quantitative assessment model for deadhesion is obtained through the following steps:

[0183] A microwave detection simulation model is constructed, wherein the simulation objects of the microwave detection simulation model include a rectangular waveguide testing device and a polyethylene pipe;

[0184] Parameters are set for the microwave detection simulation model to obtain simulation data of debonding depth and debonding thickness; the parameters include the electromagnetic parameters of the polyethylene pipe, the debonding depth range, and the debonding thickness range; the electromagnetic parameters of the polyethylene pipe are obtained by inversion calculation of the microwave S-parameters of the defect-free polyethylene pipe.

[0185] The quantitative evaluation model for debonding was trained using simulation data of debonding depth and debonding thickness.

[0186] If the modules / units integrated in the electronic device 100 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, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0187] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied 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.

[0188] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0189] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0190] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0191] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and not to limit them. Although this application has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation methods of this application.

Claims

1. A method for quantitative evaluation of debonding of polyethylene pipes, characterized by, The method comprises the following steps: obtaining the microwave S parameters of the polyethylene pipe to be detected; obtaining the debonding quantitative evaluation result of the polyethylene pipe according to the microwave S parameters of the polyethylene pipe to be detected by using a debonding quantitative evaluation model; the debonding quantitative evaluation result comprises a debonding depth and a debonding thickness; wherein the debonding quantitative evaluation model is obtained by the following steps: constructing a microwave detection simulation model, wherein the simulation object of the microwave detection simulation model comprises a rectangular waveguide measuring device and a polyethylene pipe; setting parameters of the microwave detection simulation model to obtain debonding depth simulation data and debonding thickness simulation data; the parameters comprise electromagnetic parameters of the polyethylene pipe, a debonding depth range and a debonding thickness range; the electromagnetic parameters of the polyethylene pipe are obtained by inverse calculation on the obtained microwave S parameters of the defect-free polyethylene pipe, and specifically: In Within the frequency band, several measurements are made on the polyethylene pipe with different thicknesses and no defects by using microwave signals with the same length, and several groups of microwave S parameters are obtained as the microwave S parameters of the polyethylene pipe with no defects. performing inverse calculation on the microwave S parameters of the defect-free polyethylene pipe to obtain electromagnetic parameters of the polyethylene pipe, wherein the electromagnetic parameters comprise a relative permittivity and a relative permeability; the relative permittivity and the relative permeability are obtained by the following method: wherein denotes the equivalent reflection coefficient; denotes the reflection coefficient; denotes the forward transmission coefficient; denotes the equivalent transmission coefficient; obtaining a negative refractive index and a characteristic impedance according to the equivalent reflection coefficient and the equivalent transmission coefficient, as shown below: wherein represents a negative refractive index; represents an imaginary unit; represents a free-space wave number, wherein is a free-space wavelength; represents a physical thickness of the measured sample in the direction of electromagnetic wave propagation; represents a characteristic impedance; determining the relative permittivity and the relative permeability according to the negative refractive index and the characteristic impedance, as shown below: In the formulae, denotes the relative dielectric constant; denotes the relative magnetic permeability; training the debonding quantitative evaluation model by using the debonding depth simulation data and the debonding thickness simulation data.

2. The polyethylene pipe debonding quantitative evaluation method according to claim 1, characterized by, The debonding quantitative evaluation model is as follows: wherein, represents a debonding depth prediction value of the sample under test, represents a covariance matrix of debonding depth feature vectors between the simulated data with the same debonding depth but different debonding thickness and the sample under test; represents a covariance matrix of debonding depth feature vectors between the simulated data with the same debonding depth but different debonding thickness; represents a debonding depth of the simulated data; represents a debonding thickness prediction value of the sample under test; represents a covariance matrix of debonding thickness feature vectors between the simulated data with the same debonding depth but different debonding thickness and the sample under test; represents a covariance matrix of debonding thickness feature vectors of the simulated data with the same debonding depth but different debonding thickness; is a noise variance; represents a debonding thickness of the simulated data; represents an identity matrix.

3. The polyethylene pipe debonding quantitative evaluation method according to claim 1, characterized by, the training of the debonding quantitative evaluation model by using the debonding depth simulation data and the debonding thickness simulation data is specifically as follows: determining a microwave response feature combination based on the debonding depth simulation data and the debonding thickness simulation data; for each sample, respectively constructing a debonding depth feature vector and a debonding thickness feature vector of the corresponding sample based on the microwave response feature combination; the debonding depth feature vectors and the debonding thickness feature vectors of all samples form a training set for training the debonding quantitative evaluation model.

4. The polyethylene pipe debonding quantitative evaluation method according to claim 3, characterized by The microwave response feature combination determines the optimal combination of the microwave response features of each sample through cross-validation.

5. The polyethylene pipe debonding quantitative evaluation method according to claim 3, characterized by, The combination of microwave response features includes return loss, phase, phase, impedance, group delay, and VSWR.

6. A polyethylene pipe debonding quantitative evaluation system characterized by comprising: The system performs the steps of the polyethylene pipe debonding quantitative evaluation method according to any one of claims 1-5, comprising: an evaluation parameter acquisition unit configured to obtain the microwave S parameters of the polyethylene pipe to be detected; a debonding quantitative evaluation unit configured to obtain the debonding quantitative evaluation result of the polyethylene pipe according to the microwave S parameters of the polyethylene pipe to be detected by using a debonding quantitative evaluation model; the debonding quantitative evaluation result comprises a debonding depth and a debonding thickness; wherein the debonding quantitative evaluation model is obtained by the following steps: constructing a microwave detection simulation model, wherein the simulation object of the microwave detection simulation model comprises a rectangular waveguide measuring device and a polyethylene pipe; setting parameters of the microwave detection simulation model to obtain debonding depth simulation data and debonding thickness simulation data; the parameters comprise electromagnetic parameters of the polyethylene pipe, a debonding depth range and a debonding thickness range; the electromagnetic parameters of the polyethylene pipe are obtained by inverse calculation on the obtained microwave S parameters of the defect-free polyethylene pipe; training the debonding quantitative evaluation model by using the debonding depth simulation data and the debonding thickness simulation data.

7. An electronic device, comprising: The method comprises the following steps: The electronic device comprises a memory, one or more processors; the memory is coupled with the processor; wherein the memory has computer program code stored therein, the computer program code comprises computer instructions, when the computer instructions are executed by the processor, the electronic device executes the steps of the polyethylene pipe debonding quantitative evaluation method as claimed in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, when the computer program is executed by the processor, the steps of the polyethylene pipe debonding quantitative evaluation method as claimed in any one of claims 1-5 are realized.

Images