Magnetic ultrasonic guided wave high-temperature chemical pipeline defect detection system and method

Abstract

The invention provides a high-temperature chemical pipeline defect detection system and method based on magnetostrictive ultrasonic guided waves, and relates to the technical field of high-temperature chemical pipeline defect detection.The method comprises the steps that magnetostrictive excitation signals are applied, the propagation process of the guided waves in a pipeline is calculated through numerical simulation, and the amplitude and phase information of guided wave signals is obtained; performing time-frequency domain analysis on the guided wave signal, extracting group velocities and phase velocities of guided waves in different modes, determining frequency dispersion characteristics of the guided waves, and evaluating the influence of a high-temperature environment on guided wave propagation; processing the received guided wave signal by adopting a modal decomposition and reconstruction algorithm to obtain a corrected guided wave signal; and carrying out feature extraction and parameter estimation on the corrected guided wave signal to obtain the position, size and type information of the pipeline defect. According to the invention, the problem of multilayer heterogeneous pipeline guided wave detection in a high-temperature environment is solved, the accuracy and reliability of defect detection are improved, and an effective method is provided for pipeline health monitoring and safety assessment.

Description

Technical Field

[0001] The present invention relates to the technical field of high-temperature chemical pipeline defect detection, and in particular to a high-temperature chemical pipeline defect detection system and method using magneto-ultrasonic guided waves. Background Art

[0002] When magnetostrictive ultrasonic guided waves propagate in multi-layer heterogeneous high-temperature chemical pipeline structures, due to the complexity of the pipeline structure and the influence of the high-temperature environment, the propagation characteristics of the guided waves show obvious multi-modal and dispersion characteristics. The propagation speeds of guided waves of different modes in the pipeline are different, and they change with the frequency, resulting in distortion and distortion of the guided wave signal. In addition, the sound velocity and attenuation coefficient of the pipeline material change in the high-temperature environment, further affecting the propagation behavior of the guided wave. Therefore, to accurately obtain the propagation characteristics of guided waves in multi-layer heterogeneous high-temperature pipelines, it is necessary to comprehensively consider factors such as pipeline structure, material properties, and temperature field distribution, establish a complete mathematical and physical model, and combine numerical solution methods for simulation analysis. At the same time, it is also necessary to design experiments in a targeted manner to study the changing laws of guided wave signals under different excitation frequencies and guided wave modes, and select the best detection parameters. Only on the basis of a deep understanding of the guided wave propagation mechanism can the defect location and evaluation of multi-layer heterogeneous high-temperature pipelines be achieved to ensure the safe operation of the pipeline. Summary of the invention

[0003] The present invention provides a high-temperature chemical pipeline defect detection method using magnetically induced ultrasonic guided waves, which mainly includes:

[0004] A discrete mathematical model of a multi-layer heterogeneous high-temperature pipeline is established to obtain the temperature field distribution data of the pipeline under a high-temperature environment, calculate the sound velocity and attenuation coefficient of the pipeline material at different temperatures and assign them to each grid unit of the discrete mathematical model; a magnetostrictive excitation signal is applied, and the propagation process of the guided wave in the pipeline is calculated by numerical simulation to obtain the amplitude and phase information of the guided wave signal; the guided wave signal is analyzed in the time and frequency domains to extract the group velocity and phase velocity of the guided wave in different modes, determine the dispersion characteristics of the guided wave, and evaluate the influence of the high-temperature environment on the guided wave propagation; the received guided wave signal is processed by a modal decomposition and reconstruction algorithm to obtain a corrected guided wave signal; feature extraction and parameter estimation are performed on the corrected guided wave signal to obtain the location, size and type information of the pipeline defect. If the detected defect parameter exceeds the preset threshold, it is judged that the pipeline has a serious defect; the detection results under different excitation frequencies and guided wave modes are comprehensively analyzed, and the most sensitive detection parameter combination to the pipeline defect is selected to form a high-temperature chemical pipeline defect detection scheme suitable for magnetostrictive ultrasonic guided waves.

[0005] Furthermore, the establishment of a discrete mathematical model of a multi-layer heterogeneous high-temperature pipeline includes: establishing a three-dimensional geometric model of the pipeline according to the structural parameters and material properties of the multi-layer heterogeneous high-temperature pipeline; meshing the three-dimensional geometric model using a finite element method to obtain a discrete mathematical model of the pipeline structure; establishing a dynamic equation of the pipeline structure based on the boundary conditions and initial conditions of the pipeline; solving the dynamic equation using a finite element method to obtain the propagation characteristics of ultrasonic guided waves in the pipeline structure, including phase velocity and group velocity, and drawing a dispersion curve; selecting a suitable frequency range according to the dispersion curve, avoiding severe dispersion areas, and determining the excitation frequency; and using a modal decomposition method to decompose the ultrasonic guided waves into different guided wave modes to obtain the vibration shape and energy distribution of each mode.

[0006] Furthermore, the method of obtaining the temperature field distribution data of the pipeline in a high temperature environment, calculating the sound velocity and attenuation coefficient of the pipeline material at different temperatures and assigning them to each grid unit of the discretized mathematical model includes: obtaining the boundary conditions of the pipeline in a high temperature environment, including the convective heat transfer coefficient and the ambient temperature of the inner and outer surfaces of the pipeline; using a finite element thermal analysis module, considering the thermal conductivity and specific heat capacity parameters of the pipeline material, calculating the temperature field distribution of the pipeline under steady-state conditions, and obtaining the temperature data of each grid unit of the pipeline; using a finite element thermal stress analysis module according to the temperature field distribution data, calculating the stress and strain distribution of the pipeline at different temperatures; using a finite element acoustic module based on the elastic modulus and Poisson's ratio parameters of the pipeline material and the temperature field distribution data, solving the sound wave propagation equation, and calculating the sound velocity distribution of the pipeline material at different temperatures; introducing the damping characteristics of the pipeline material into the sound wave propagation equation, considering the viscous damping coefficient of the material, and calculating the sound wave attenuation coefficient distribution of the pipeline material at different temperatures.

[0007] Furthermore, the applying of magnetostrictive excitation signal and the calculation of the propagation process of the waveguide in the pipeline by numerical simulation to obtain the amplitude and phase information of the waveguide signal include: applying magnetostrictive excitation signal at the incident end of the discretized mathematical model for different waveguide excitation frequencies and modes, wherein the parameters of the magnetostrictive excitation signal include center frequency, bandwidth, amplitude and duration; calculating the propagation process of the waveguide in the discretized mathematical model by finite element numerical simulation method to obtain the amplitude and phase information of the waveguide signal at different positions; performing time-frequency analysis on the waveguide signal by short-time Fourier transform to obtain the dispersion curve and group velocity curve of the waveguide propagation in the pipeline; analyzing the dispersion curve and group velocity curve to determine the propagation characteristics of the waveguide under different frequencies and modes, and determining the detection frequency and mode; performing modal decomposition on the waveguide signal by empirical mode decomposition method to obtain waveguide signal components of different modes; analyzing the amplitude and phase characteristics of waveguide signals of different modes to determine the propagation characteristics and attenuation laws of each mode.

[0008] Furthermore, the waveguide signal is subjected to time-frequency domain analysis to extract the group velocity and phase velocity of different modal waveguides, determine the dispersion characteristics of the waveguide, and evaluate the influence of the high temperature environment on the waveguide propagation, including: based on the established pipeline mechanics model, a semi-analytical finite element method is used to derive the dispersion equation of the ultrasonic waveguide in the pipeline, and the dispersion equation is solved to obtain the dispersion curves of the waveguides of different modes, including the relationship between the group velocity and the phase velocity and the frequency; by means of numerical simulation software, the propagation process of the ultrasonic waveguide in the multi-layer heterogeneous pipeline is simulated to obtain the displacement field and stress field distribution of the waveguides of different modes when propagating in the pipeline; the waveguide signal obtained by numerical simulation is subjected to time-frequency domain analysis by means of numerical calculation software, and the time-frequency characteristics of the waveguides of different modes are extracted by short-time Fourier transform to judge the dispersion characteristics and multimodal characteristics of the waveguide signal; according to the time-frequency characteristics of the waveguide signal, the curves of the group velocity and phase velocity of the waveguides of different modes and the frequency are determined, and compared with the dispersion curves calculated theoretically to verify the numerical simulation results.

[0009] Furthermore, the modal decomposition and reconstruction algorithm is used to process the received waveguide signal to obtain a corrected waveguide signal, including: using a semi-analytical finite element method to solve the dispersion equation of ultrasonic guided waves in the pipeline to obtain dispersion curves and modal shapes of different modal guided waves; determining the detection frequency range according to the dispersion curve, and selecting a mode with a long propagation distance and high sensitivity as the detection mode; using wavelet transform to perform time-frequency analysis on the received waveguide signal to extract arrival time, amplitude and frequency characteristic parameters of different modes; using continuous wavelet transform to perform time-frequency representation on the waveguide signal to obtain wavelet coefficient matrices at different scales; identifying waveguide signal components of different modes by analyzing the wavelet coefficient matrix; using empirical mode decomposition to decompose the waveguide signal into multiple intrinsic mode functions to achieve separation of different modal signals; performing dispersion compensation on each separated modal signal to correct the waveform distortion caused by dispersion and restore the original form of the signal.

[0010] Furthermore, the feature extraction and parameter estimation of the corrected guided wave signal are performed to obtain the location, size and type information of the pipeline defect. If the detected defect parameter exceeds the preset threshold, it is judged that the pipeline has serious defects, including: using the ultrasonic guided wave of the selected frequency and mode to excite the pipeline, and through simulation analysis of the propagation characteristics of the ultrasonic guided wave in the defective pipeline, extracting the time domain waveform and spectrum characteristics of the defect reflection signal; using the signal processing method to reduce the noise and enhance the characteristics of the extracted defect reflection signal to improve the signal-to-noise ratio and recognition of the signal; using the support vector machine to classify and identify the processed defect reflection signal, and judge the type and size of the pipeline defect according to the signal characteristics; calculating the axial position and radial depth of the defect according to the time difference and amplitude of the defect reflection signal, so as to achieve accurate positioning and quantitative evaluation of the pipeline defect; comparing the defect parameters obtained by the evaluation with the preset defect severity criterion to determine the hazard level of the defect; using the D-S evidence theory to fuse the defect information obtained from multiple detection positions and multiple modes to improve the accuracy and reliability of defect identification and positioning.

[0011] Further, the comprehensive analysis of the detection results under different excitation frequencies and guided wave modes, the optimization of the detection parameter combination most sensitive to pipeline defects, and the formation of a high-temperature chemical pipeline defect detection scheme suitable for magneto-ultrasonic guided waves include: analyzing the slope of the dispersion curve and the change of the group velocity, judging the dispersion characteristics of the guided waves under different frequencies, and selecting the frequency range with smaller dispersion as the excitation frequency; analyzing the displacement and stress distribution of the guided waves of different modes according to the modal characteristics, and determining the guided wave mode sensitive to defects; introducing defects of different types and sizes into the pipeline model, and simulating the defects by setting corresponding material properties and geometric parameter changes in the finite element model; using the time-domain finite difference method to solve the guided wave field in the defective pipeline model, and iteratively solving the evolution of the guided wave field in time and space; obtaining the guided wave signal of the defective pipeline, and extracting the time-frequency characteristics of the defect reflection signal; by analyzing the parameters of the defect reflection signal under different excitation frequencies and guided wave modes, evaluating the sensitivity of the detection parameters to the defects; comprehensively comparing the defect detection results under different detection parameter combinations, and using the genetic algorithm to search for the optimal parameter combination to obtain a high-temperature chemical pipeline defect detection scheme suitable for magneto-ultrasonic guided waves.

[0012] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0013] The present invention discloses a high-temperature chemical pipeline defect detection system and method using magnetostrictive ultrasonic guided waves. The method first establishes a three-dimensional geometric model of the pipeline and performs finite element meshing to obtain the temperature field distribution in a high-temperature environment. The thermoacoustic coupling equation is used to calculate the acoustic parameters of the pipeline material at different temperatures, and the results are assigned to the discretized model. A magnetostrictive excitation signal is applied to the incident end of the model, and the guided wave propagation process is calculated by numerical simulation. The simulated signal is subjected to time-frequency analysis to extract the guided wave dispersion characteristics. The received signal is processed by a modal decomposition and reconstruction algorithm, and each mode is subjected to distortion correction. Pipeline defect information is obtained by feature extraction and parameter estimation. If the detected defect parameter exceeds a preset threshold, it is determined that a serious defect exists. The detection results under different excitation parameters are comprehensively analyzed, and the most sensitive detection parameter combination is selected. The present invention solves the difficult problem of guided wave detection of multi-layer heterogeneous pipelines under high-temperature environments, improves the accuracy and reliability of defect detection, and provides an effective method for pipeline health monitoring and safety assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flow chart of a high-temperature chemical pipeline defect detection system and method using magneto-ultrasonic guided waves. DETAILED DESCRIPTION

[0015] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0016] like Figure 1 In this embodiment, a high-temperature chemical pipeline defect detection method using magnetically induced ultrasonic guided waves may specifically include:

[0017] Step S101, according to the structural parameters and material properties of the multi-layer heterogeneous high-temperature pipeline, a three-dimensional geometric model of the pipeline is established, and the model is meshed using the finite element method to obtain a discretized mathematical model of the pipeline structure.

[0018] According to the structural parameters and material properties of multi-layer heterogeneous high-temperature pipelines, a three-dimensional geometric model of the pipeline is established. According to the boundary conditions and initial conditions of the pipeline, the dynamic equation of the pipeline structure is established. The finite element method is used to solve the dynamic equation, and the propagation characteristics of ultrasonic guided waves in the pipeline structure are obtained, including the phase velocity and group velocity of ultrasonic guided waves at different frequencies, and the dispersion curve is drawn to analyze the dispersion characteristics. According to the dispersion curve, the appropriate frequency range is selected to avoid the severe dispersion area and determine the excitation frequency. The modal decomposition method is used to decompose the ultrasonic guided waves into different guided wave modes, and the vibration shape and energy distribution of each mode are obtained. Through modal synthesis, the propagation characteristics of different modes are analyzed, including propagation velocity, attenuation coefficient and energy distribution, and the dominant mode is determined. According to the characteristics of the dominant mode, the genetic algorithm is used to optimize the sensor layout scheme, and the signal-to-noise ratio of the sensor receiving signal is used as the optimization target, so that the sensor can effectively excite and receive the dominant mode and improve the detection sensitivity. The finite element method is used to simulate the propagation process of ultrasonic guided waves in multi-layer heterogeneous pipelines, analyze the influence of the interface on the guided wave propagation, and obtain the propagation law. By combining theoretical analysis with numerical simulation, a mathematical model of ultrasonic guided wave propagation in multi-layer heterogeneous high-temperature pipelines is established, including dynamic equations and constitutive equations, laying the foundation for subsequent defect detection.

[0019] Specifically, ultrasonic guided wave testing of multi-layer heterogeneous high-temperature pipelines is a complex engineering technology. First, a three-dimensional geometric model of the pipeline needs to be established, which is usually based on the structural parameters and material properties of the pipeline. For example, for a three-layer composite pipeline, the inner layer may be stainless steel, the middle layer is insulation material, and the outer layer is carbon steel. The thickness, diameter and other parameters of each layer need to be accurately modeled. Establishing the dynamic equation is the next key step. Here, the boundary conditions and initial conditions of the pipeline need to be considered. The boundary conditions may include fixed or free states at both ends of the pipeline, and the initial conditions may be static states. The dynamic equation is usually a partial differential equation that describes the motion law of the pipeline structure under the action of external forces. Using the finite element method to solve the dynamic equation is a common numerical method. This method discretizes the continuous pipeline structure into a finite number of units, and then solves the equation on each unit. In this way, the characteristics of ultrasonic guided wave propagation in the pipeline can be obtained, including phase velocity and group velocity. The phase velocity reflects the phase propagation velocity of the wave, while the group velocity represents the energy propagation velocity. The dispersion curve is an important tool for analyzing the propagation characteristics of ultrasonic guided waves. It shows the relationship between wave velocity and frequency. By drawing the dispersion curve, we can intuitively see in which frequency ranges the propagation characteristics of the wave are relatively stable. For example, we may find that in the frequency range of 20-50kHz, the group velocity of a certain mode changes little, which is an ideal excitation frequency range. Modal decomposition is an effective way to understand complex waves. In multi-layer heterogeneous pipes, ultrasonic guided waves may contain multiple modes at the same time. Through modal decomposition, complex waves can be decomposed into the superposition of several simple modes. Each mode has its own specific vibration shape and energy distribution. For example, we may find that the energy distribution of a certain axisymmetric mode on the pipe wall is the most concentrated, which may be an ideal detection mode. Modal synthesis is the inverse process of modal decomposition. By analyzing the propagation characteristics of different modes, such as propagation velocity, attenuation coefficient and energy distribution, we can determine which modes dominate the propagation process. These dominant modes usually have lower attenuation and higher energy, so they are more suitable for long-distance detection. Optimization of sensor layout is the key to improving detection results. Genetic algorithm is a commonly used optimization method that simulates the process of biological evolution. Here, the position, angle and other parameters of the sensor can be encoded as "genes", and the signal-to-noise ratio of the received signal can be used as "fitness". Through multiple generations of iterations, an optimized sensor layout scheme can be finally obtained. The influence of the interface on guided wave propagation is an important issue in multi-layer heterogeneous pipelines. The interface between different materials will cause reflection and transmission of waves, affecting the propagation of waves. Through finite element simulation, these effects can be analyzed in detail. For example, it may be found that at certain frequencies, the interface will cause serious energy loss, while at other frequencies, the effect of the interface is relatively small. Finally, through the combination of theoretical analysis and numerical simulation, a complete mathematical model can be established to describe the propagation behavior of ultrasonic guided waves in multi-layer heterogeneous high-temperature pipelines.This model includes not only the dynamic equations, but also the constitutive equations that describe the material behavior. For example, for high temperature environments, it may be necessary to consider the nonlinear thermal expansion effect of the material. This comprehensive model provides a theoretical basis for subsequent defect detection and can be used to predict the impact of different types of defects on ultrasonic guided wave propagation, thereby improving the accuracy and reliability of detection.

[0020] Step S102, obtain the temperature field distribution data of the pipeline in a high temperature environment, and use the thermoacoustic coupling equation to calculate the sound velocity and attenuation coefficient of the pipeline material at different temperatures according to the thermophysical parameters of the pipeline material. The thermoacoustic coupling equation includes the heat conduction equation and the sound wave propagation equation, and considers the influence of temperature on acoustic parameters. The calculation results are assigned to each grid unit of the discretized mathematical model.

[0021] According to the geometric dimensions and material properties of the pipeline, ANSYS is used to establish a three-dimensional finite element model of the pipeline and divide the hexahedral grid units. The boundary conditions of the pipeline in a high temperature environment, including the convection heat transfer coefficient of the inner and outer surfaces of the pipeline and the ambient temperature, are obtained as the load conditions of the finite element model. The steady-state thermal analysis module of ANSYS is used to consider the thermal physical parameters such as thermal conductivity and specific heat capacity of the pipeline material, calculate the temperature field distribution of the pipeline under steady-state conditions, and obtain the temperature data of each grid unit of the pipeline. According to the thermal expansion coefficient and temperature field distribution data of the pipeline material, the thermal stress analysis module of ANSYS is used to calculate the stress and strain distribution of the pipeline at different temperatures. Based on the acoustic parameters such as elastic modulus and Poisson's ratio of the pipeline material, as well as the temperature field distribution data, the acoustic module of COMSOL Multiphysics is used to solve the sound wave propagation equation to calculate the sound velocity distribution of the pipeline material at different temperatures. The damping characteristics of the pipeline material are introduced into the sound wave propagation equation, and the viscous damping coefficient of the material is considered to calculate the sound wave attenuation coefficient distribution of the pipeline material at different temperatures. The calculated sound velocity and attenuation coefficient distribution are compared with the preset convergence conditions. If the convergence conditions are not met, return to step 5, adjust the acoustic parameters, and recalculate the sound velocity and attenuation coefficient until the convergence conditions are met. The calculated sound velocity and attenuation coefficient of the pipeline material at different temperatures are assigned to each grid unit of the pipeline finite element model to form an acoustic parameter distribution that takes into account the influence of temperature, providing material property data for subsequent sound wave propagation analysis.

[0022] Specifically, finite element modeling of multi-layer heterogeneous high-temperature pipelines is a complex engineering task. Taking a three-layer composite pipeline as an example, the inner layer may be a high-temperature resistant alloy, the middle layer is an insulation material, and the outer layer is carbon steel. When using ANSYS software to build a three-dimensional model, it is necessary to accurately input the geometric dimensions and material properties of each layer. For example, the inner layer is 5mm thick, 200mm in diameter, and the material is Inconel625; the middle layer is 10mm thick, the material is ceramic fiber; the outer layer is 8mm thick, and the material is P91 steel. Meshing is a key step in finite element analysis. For this multi-layer structure, hexahedral units are usually used because they can better capture the stress concentration at the interface between layers. In the axial direction of the pipeline, finer mesh division may be required to accurately simulate the propagation of sound waves. For example, in critical areas such as near welds, the mesh size may need to be controlled at around 1mm. The setting of boundary conditions directly affects the accuracy of the analysis results. For high-temperature pipelines, the inner surface may face a high-temperature fluid of 600℃, and the convective heat transfer coefficient may reach 1000W / (m 2 ·K); the outer surface may be exposed to an environment of 50°C, and the convection heat transfer coefficient is about 10W / (m 2 ·K). These parameters need to be adjusted according to the actual working conditions. Steady-state thermal analysis is the basis for understanding the temperature distribution of the pipeline. Considering that the thermal conductivity of the material changes with temperature, temperature-related material properties need to be input. For example, the thermal conductivity of Inconel625 increases from 9.8W / (m·K) to 22.8W / (m·K) in the range of 20-800℃. The analysis results may show that the inner wall temperature is close to 600℃, while the outer wall temperature may drop to around 100℃. Thermal stress analysis reveals the impact of temperature gradients on the pipeline structure. Due to the different thermal expansion coefficients of the inner and outer materials, significant thermal stress may result. For example, at 600℃, the thermal expansion coefficient of Inconel625 is approximately 16×10 -6 / K, while P91 steel is about 12×10 -6 / K. This difference may produce stresses of up to 100MPa at the interface. The calculation of the sound velocity distribution needs to take into account the effect of temperature on the elastic modulus of the material. Taking P91 steel as an example, its elastic modulus may drop from 220GPa at room temperature to 180GPa at 600℃. This will cause the sound velocity to drop from about 5900m / s at room temperature to about 5400m / s. The damping properties of the material have an important influence on the propagation of sound waves. At high temperatures, the viscous damping coefficient of the material usually increases. For example, the damping coefficient of P91 steel may increase from 0.001 at room temperature to 0.005 at 600℃. This will cause a significant increase in the sound wave attenuation coefficient, which may rise from 0.1dB / m to 0.5dB / m. Iterative calculation is the key to ensure the accuracy of the results. It may be necessary to adjust the material parameters multiple times until the changes in the sound velocity and attenuation coefficient are less than the preset threshold, such as 1%. This process may take 5-10 iterations to converge. The final acoustic parameter distribution lays the foundation for subsequent analysis. For example, near the inner wall of the pipe, due to the higher temperature, the speed of sound may decrease by 10%, while the attenuation coefficient may increase by 3 times. Such data are crucial for accurately predicting the propagation behavior of sound waves, helping to optimize ultrasonic testing strategies and improve the accuracy of defect detection.

[0023] Step S103, for different guided wave excitation frequencies and modes, a magnetostrictive excitation signal is applied at the incident end of the discretized mathematical model. The parameters of the magnetostrictive excitation signal include center frequency, bandwidth, amplitude and duration. The propagation process of the guided wave in the pipeline is calculated by a numerical simulation method to obtain the amplitude and phase information of the guided wave signal at different positions.

[0024] According to the material properties and geometric dimensions of the multi-layer heterogeneous pipeline, the finite element method is used to establish a discretized mathematical model of the pipeline, including the different material properties and interface conditions of the inner layer, outer layer and middle layer of the pipeline. For different guided wave excitation frequencies and modes, a magnetostrictive excitation signal is applied at the incident end of the discretized mathematical model. The parameters of the magnetostrictive excitation signal include center frequency, bandwidth, amplitude and duration. The propagation process of the guided wave in the pipeline is calculated by the finite element numerical simulation method, and the amplitude and phase information of the guided wave signal at different positions are obtained. The short-time Fourier transform is used to perform time-frequency analysis on the guided wave signal to obtain the dispersion curve and group velocity curve of the guided wave propagation in the pipeline. The characteristics of the dispersion curve and the group velocity curve are analyzed, and the propagation characteristics of the guided wave under different frequencies and modes are judged by comparing the propagation distance and attenuation of the guided wave under different frequencies and modes, and the optimal detection frequency and mode are determined. The empirical mode decomposition method is used to perform modal decomposition on the guided wave signal to obtain the guided wave signal components of different modes. The amplitude and phase characteristics of different modal waveguide signals are analyzed, and the propagation characteristics and attenuation laws of each mode are determined by comparing the propagation distances and attenuation of different modal waveguide signals. Based on the analysis results of the comprehensive dispersion characteristics and multimodal characteristics, the genetic algorithm is used to optimize the parameters of the magnetostrictive excitation signal. By constructing a fitness function, the optimization goal is to maximize the propagation distance of the waveguide and minimize the attenuation, so that the magnetostrictive excitation signal can excite the waveguide mode with good propagation characteristics and low attenuation. The optimized magnetostrictive excitation signal is verified by finite element simulation, and the waveguide propagation characteristics excited in the multi-layer heterogeneous pipeline are analyzed. By comparing the waveguide propagation distance and attenuation before and after optimization, it is determined that the optimized magnetostrictive excitation signal can effectively excite the waveguide mode with good propagation characteristics, laying the foundation for subsequent defect detection.

[0025] Specifically, the finite element modeling of multi-layer heterogeneous pipelines requires an accurate description of the properties and geometric dimensions of each layer of material. Taking a three-layer composite pipeline as an example, the inner layer may be a high-temperature resistant alloy Inconel625 with a thickness of 5mm; the middle layer is a ceramic fiber insulation material with a thickness of 10mm; and the outer layer is P91 steel with a thickness of 8mm. In finite element software such as ANSYS, it is necessary to define parameters such as elastic modulus, Poisson's ratio, density, etc. for each layer of material, and consider the changes of these parameters with temperature. For example, the elastic modulus of Inconel625 is about 205GPa at room temperature, but it may drop to 180GPa at 600℃. The setting of interface conditions is also crucial, and it is usually assumed that the layers are completely bonded. The design of magnetostrictive excitation signals directly affects the excitation effect of guided waves. A typical excitation signal may be a Gaussian modulated sine wave, whose center frequency, bandwidth, amplitude, and duration are all key parameters. For example, for a pipeline with a diameter of 200mm, a center frequency of 30kHz and a duration of 5 cycles may be selected, so that a guided wave mode with a wavelength equivalent to the diameter of the pipeline can be excited. The choice of bandwidth affects the time domain resolution of the signal, and a Gaussian envelope of 3-5 cycles is usually selected. When calculating the guided wave propagation process by finite element simulation, the time step and grid size need to be set reasonably. The general principle is that the time step should be less than 1 / 10 of the time required for the wave to propagate through the smallest grid unit, and the grid size should be less than 1 / 10 of the shortest wavelength. For a 30kHz guided wave, the wavelength in steel is about 180mm, so the grid size may need to be controlled below 15mm. The simulation results include time domain signals such as displacement and stress at different positions. Short-time Fourier transform (STFT) is an effective tool for analyzing the time-frequency characteristics of guided wave signals. By selecting a suitable time window function (such as Hanning window) and window length, a balance can be achieved between time and frequency resolution. STFT results can intuitively display the dispersion characteristics of various modes. For example, for a steel pipe with a wall thickness of 10mm, dispersion curves of multiple modes such as L(0,1), L(0,2) and T(0,1) may be observed in the range of 20-100kHz. Analysis of dispersion curves and group velocity curves helps to select the best detection frequency and mode. Generally, non-dispersive or weakly dispersive modes are more suitable for long-distance propagation. For example, the T (0, 1) mode exhibits non-dispersive characteristics in most frequency ranges, and the group velocity is close to the shear wave velocity (about 3200m / s), so it is often used for pipeline detection. The L (0, 2) mode also exhibits weak dispersion characteristics in certain frequency ranges, and the propagation speed is faster (up to 5000m / s), which is suitable for rapid scanning. The empirical mode decomposition (EMD) method can decompose complex guided wave signals into multiple intrinsic mode functions (IMFs). Each IMF may correspond to a specific guided wave mode or frequency component. By analyzing the energy distribution and propagation characteristics of each IMF, the main guided wave modes can be identified.For example, in a typical pipeline waveguide signal, it may be found that the first IMF corresponds to the high-frequency L(0, 2) mode, while the second IMF corresponds to the low-frequency L(0, 1) mode. When the genetic algorithm optimizes the magnetostrictive excitation signal, it is necessary to define a suitable fitness function. A possible fitness function is the ratio of the amplitude of the waveguide signal at a specific propagation distance (such as 10m) to the initial amplitude. The optimization parameters may include the center frequency (such as 20-100kHz), the signal bandwidth (such as 2-10 cycles), and the excitation duration (such as 3-10 cycles). Through multiple generations of iterations, it may be found that the T(0, 1) mode with the best propagation characteristics can be excited by using a 5-cycle Gaussian modulated sine wave near 45kHz. The optimized magnetostrictive excitation signal needs to be verified by finite element simulation. Comparing the waveguide propagation distance and attenuation before and after optimization, it may be found that the optimized signal still maintains 80% of the initial amplitude after a propagation distance of 10m, while it may only be 50% before optimization. This improvement is of great significance for improving the signal-to-noise ratio and detection sensitivity in actual detection. Through this series of analysis and optimization processes, a solid theoretical foundation can be laid for the guided wave detection technology of multi-layer heterogeneous pipelines, and the reliability and effectiveness of detection can be improved.

[0026] Step S104, perform time-frequency domain analysis on the waveguide signal obtained by numerical simulation, extract the curves of group velocity and phase velocity of different modal waveguides with frequency, determine the dispersion characteristics of the waveguide, analyze the distortion and degree of the waveguide signal, and evaluate the impact of the high temperature environment on the waveguide propagation.

[0027] According to the material properties and geometric dimensions of the multi-layer heterogeneous pipeline, a mathematical model of the pipeline is established to determine the material parameters such as density, elastic modulus, Poisson's ratio of each layer of the pipeline and the geometric parameters such as inner diameter, outer diameter, wall thickness of the pipeline. Based on the established pipeline mechanical model, the semi-analytical finite element method is used to derive the dispersion equation of ultrasonic guided waves in the pipeline, and the dispersion equation is solved to obtain the dispersion curves of different modal guided waves, including the relationship between group velocity and phase velocity with frequency. The propagation process of ultrasonic guided waves in multi-layer heterogeneous pipelines is simulated by numerical simulation software COMSOL Multiphysics, and the displacement field and stress field distribution of different modal guided waves propagating in the pipeline are obtained. MATLAB is used to perform time-frequency domain analysis on the guided wave signals obtained by numerical simulation, and short-time Fourier transform (STFT) is used to extract the time-frequency characteristics of different modal guided waves to determine the dispersion characteristics and multimodal characteristics of the guided wave signals. According to the time-frequency characteristics of the guided wave signals, the curves of group velocity and phase velocity of different modal guided waves with frequency are determined, and compared with the dispersion curves calculated theoretically to verify the accuracy of the numerical simulation results. A high temperature environment is introduced into the numerical simulation. By setting the temperature load in the simulation model and considering the influence of material parameters with temperature, the influence of high temperature on the propagation characteristics of ultrasonic guided waves in multi-layer heterogeneous pipelines is analyzed, and the time-frequency characteristics of the guided wave signals at different temperatures are obtained. Signal processing methods such as wavelet denoising and Hilbert transform are used to denoise and enhance the guided wave signals in the high temperature environment, improve the signal-to-noise ratio and resolution of the signal, and facilitate the extraction of the dispersion characteristics and multimodal characteristics of the guided wave signals. By comparing and analyzing the differences in the time-frequency characteristics of the guided wave signals under normal and high temperature environments, the degree of influence of the high temperature environment on the guided wave propagation is evaluated, and the degree of distortion and distortion of the guided wave signals is determined. According to the results of the dispersion characteristics and multimodal characteristics analysis, combined with the expected type and size of pipeline defects, the ultrasonic guided wave detection parameters are optimized, such as selecting the excitation frequency range and guided wave mode where the guided wave signal energy is concentrated and sensitive to defects, so as to improve the sensitivity and reliability of defect detection in multi-layer heterogeneous pipelines.

[0028] Specifically, mathematical modeling of multi-layer heterogeneous pipelines requires accurate description of the properties and geometric parameters of each layer of materials. Taking a three-layer composite pipeline as an example, the inner layer may be a high-temperature resistant alloy Inconel625, the middle layer is a ceramic fiber insulation material, and the outer layer is P91 steel. For Inconel625, the density at room temperature is about 8440kg / m 3 , elastic modulus is about 205GPa, Poisson's ratio is about 0.3. The density of ceramic fiber may be in the range of 2000-3000kg / m3, and the elastic modulus is between 50-100GPa. The density of P91 steel is about 7700kg / m 3, the elastic modulus is about 210GPa, and the Poisson's ratio is about 0.3. In terms of geometric parameters, it is assumed that the inner diameter of the pipeline is 200mm, the thickness of the Inconel625 layer is 5mm, the thickness of the ceramic fiber layer is 10mm, and the thickness of the P91 steel layer is 8mm. Based on these parameters, the dispersion equation can be derived using the semi-analytical finite element method. This method combines the accuracy of the analytical method and the flexibility of the finite element method, and is particularly suitable for the analysis of multi-layer structures. By solving the dispersion equation, the dispersion curves of different modal guided waves can be obtained. For example, for the above three-layer pipeline, multiple modes such as L(0, 1), L(0, 2) and T(0, 1) may be observed in the frequency range of 20-100kHz. The T(0, 1) mode usually exhibits non-dispersive characteristics, and the group velocity is close to the shear wave velocity (about 3200m / s). The L(0, 2) mode may exhibit weak dispersion characteristics in certain frequency ranges, and the propagation speed can reach 5000m / s. Using

[0029] Numerical simulation using COMSOL Multiphysics can intuitively demonstrate the propagation process of guided waves in multilayer pipes. By setting the appropriate grid size (usually less than 1 / 10 of the shortest wavelength) and time step (less than 1 / 10 of the time required for the wave to propagate through the smallest grid unit), accurate simulation results can be obtained. For example, for a 30kHz guided wave, the wavelength in steel is about 180mm, so the grid size may need to be controlled below 15mm. The simulation results include the displacement field and stress field distribution at different positions, which can be exported for subsequent signal processing and analysis. Short-time Fourier transform (STFT) is an effective tool for analyzing the time-frequency characteristics of guided wave signals. When implementing STFT in MATLAB, it is necessary to select a suitable time window function (such as Hanning window) and window length. The choice of window length affects the time and frequency resolution, and it is usually necessary to trade off between the two. For example, for a center frequency of 30kHz, a window length of 1ms may be selected, which can obtain reasonable frequency resolution while maintaining sufficient time resolution. The STFT results can intuitively display the dispersion characteristics of various modes, which helps to identify different modes and analyze their propagation characteristics. High temperature environment has a significant impact on the propagation characteristics of guided waves. In COMSOL simulation, high temperature environment can be simulated by setting temperature load (such as 600℃) and considering the change of material parameters with temperature. For example, the elastic modulus of Inconel625 may drop to 180GPa at 600℃. By comparing the simulation results at room temperature and high temperature, the changes in guided wave velocity and attenuation characteristics can be observed. High temperature may cause the group velocity of some modes to decrease and increase the attenuation of the signal. Wavelet denoising is an effective method for processing guided wave signals in high temperature environment. Selecting appropriate wavelet basis functions (such as Daubechies wavelet) and decomposition layers can effectively remove thermal noise. Hilbert transform can be used to extract the envelope of the signal, which helps to analyze the energy distribution and attenuation characteristics of the signal. These treatments can improve the signal-to-noise ratio of the signal, so that the dispersion characteristics and multimodal characteristics of the guided wave signal can be accurately extracted even in high temperature environment. By comparing and analyzing the differences in the time-frequency characteristics of the guided wave signal under normal temperature and high temperature environment, the degree of influence of high temperature on guided wave propagation can be evaluated. For example, it may be found that at 600°C, the group velocity of the T(0,1) mode decreases by 5%, while the signal amplitude attenuates by 20% after propagating for 10m. This information is essential for optimizing detection strategies in high temperature environments. Based on the results of dispersion characteristics and multimodal characteristics analysis, the ultrasonic guided wave detection parameters can be optimized. For example, for detecting corrosion defects on the inner wall of a pipeline, the excitation of the L(0,2) mode in the range of 40-50kHz may be selected because this mode is more sensitive to inner wall defects in this frequency range. For detecting outer wall defects, the T(0,1) mode in the range of 30-40kHz may be more appropriate because it has good penetration ability and sensitivity to outer wall defects.Through this optimization, the sensitivity and reliability of defect detection in multi-layer heterogeneous pipelines can be significantly improved.

[0030] Step S105, according to the multi-modal dispersion characteristics of the waveguide, the received waveguide signal is processed using a modal decomposition and reconstruction algorithm. The modal decomposition uses the empirical mode decomposition method to decompose the signal into multiple intrinsic mode functions. The reconstruction algorithm performs distortion correction and compensation on each modal signal based on the dispersion compensation principle to obtain a corrected waveguide signal.

[0031] According to the material properties and geometric dimensions of the multi-layer heterogeneous pipeline, a mathematical model of the pipeline is established, including parameters such as the density, elastic modulus, and Poisson's ratio of each layer. The semi-analytical finite element method is used to solve the dispersion equation of ultrasonic guided waves in the pipeline, and the dispersion curves and modal shapes of different modal guided waves are obtained. By analyzing the results of the dispersion equation, the dispersion curve diagram is drawn to observe the dispersion characteristics and energy distribution of different modal guided waves. According to the dispersion curve diagram, the appropriate detection frequency range is determined, avoiding the high-frequency dispersion area and high attenuation area, and the mode with long propagation distance and high sensitivity is selected as the detection mode. According to the modal shape of the selected detection mode, the arrangement and excitation mode of the transducer are optimized so that it can efficiently excite and receive the guided wave signal of the target mode. The wavelet transform is used to perform time-frequency analysis on the received guided wave signal, and the characteristic parameters such as arrival time, amplitude and frequency of different modes are extracted. The continuous wavelet transform (CWT) is used to represent the guided wave signal in time and frequency, and the wavelet coefficient matrix at different scales is obtained. By analyzing the wavelet coefficient matrix, the waveguide signal components of different modes are identified. The waveguide signal is decomposed into multiple intrinsic mode functions (IMFs) using the empirical mode decomposition method (EMD) to achieve the separation of different modal signals. The time-frequency analysis results of the wavelet transform are compared with the IMFs obtained by EMD decomposition to verify the effectiveness of modal separation. Dispersion compensation is performed on each separated modal signal to correct the waveform distortion caused by dispersion and restore the original form of the signal. The back propagation model based on the dispersion curve is used to perform dispersion compensation on each modal signal. The phase velocity and group velocity of different frequency components are calculated according to the dispersion curve, and the phase and amplitude of the frequency components of the signal are adjusted to eliminate the dispersion effect. The effect of dispersion compensation is evaluated by comparing the changes in the characteristic parameters of the signal before and after compensation, and the detection sensitivity is calibrated. The propagation characteristics and defect response characteristics of each modal signal are comprehensively analyzed to construct a multi-modal fusion defect location and evaluation method. The multi-modal equation group for defect location is established using the arrival time difference and amplitude difference of each modal signal. The least square method is used to solve the equations to obtain the position coordinates of the defects. The type and severity of the defects are evaluated based on the frequency changes and energy attenuation of each modal signal. The accuracy and reliability of the detection are improved by weighted fusion of the positioning results and evaluation results of each modality. Finally, the detection results are visualized to generate intuitive defect distribution maps and evaluation reports to provide support for pipeline maintenance and decision-making.

[0032] Specifically, mathematical modeling of multi-layer heterogeneous pipelines requires accurate description of the properties and geometric parameters of each layer of materials. Taking a three-layer composite pipeline as an example, the inner layer may be a high-temperature resistant alloy Inconel625, the middle layer is a ceramic fiber insulation material, and the outer layer is P91 steel. For Inconel625, the density at room temperature is about 8440kg / m 3, elastic modulus is about 205GPa, Poisson's ratio is about 0.3. The density of ceramic fiber may be 2000-3000kg / m 3 The elastic modulus is between 50 and 100 GPa. The density of P91 steel is about 7700 kg / m 3, the elastic modulus is about 210GPa, and the Poisson's ratio is about 0.3. In terms of geometric parameters, it is assumed that the inner diameter of the pipeline is 200mm, the thickness of the Inconel625 layer is 5mm, the thickness of the ceramic fiber layer is 10mm, and the thickness of the P91 steel layer is 8mm. Based on these parameters, the dispersion equation can be derived using the semi-analytical finite element method. This method combines the accuracy of the analytical method with the flexibility of the finite element method, and is particularly suitable for the analysis of multi-layer structures. By solving the dispersion equation, the dispersion curves of different modal guided waves can be obtained. For example, for the above three-layer pipeline, multiple modes such as L(0,1), L(0,2) and T(0,1) may be observed in the frequency range of 20-100kHz. The T(0,1) mode usually exhibits non-dispersive characteristics, and the group velocity is close to the shear wave velocity (about 3200m / s). The L(0,2) mode may exhibit weak dispersion characteristics in certain frequency ranges, and the propagation velocity can reach 5000m / s. According to the results of the dispersion curve analysis, the appropriate detection frequency and mode can be selected. For example, for detecting corrosion defects on the inner wall of the pipeline, the L(0,2) mode may be selected to be excited in the range of 40-50kHz, because this mode is more sensitive to inner wall defects in this frequency range. For detecting outer wall defects, the T(0,1) mode in the range of 30-40kHz may be more suitable because it has good penetration ability and sensitivity to outer wall defects. In order to efficiently excite and receive the target mode, the transducer arrangement needs to be optimized according to the modal shape. For example, for the T(0,1) mode, a ring array transducer can be used, and multiple transducer units are evenly arranged along the circumference of the pipeline and excited in phase to produce a pure torsional mode. For the L(0,2) mode, an axially vibrating piezoelectric transducer can be used, arranged at the end of the pipeline or arranged along the axial direction of the pipeline. The received guided wave signal usually contains multiple modal components, which need to be separated and identified. Wavelet transform is an effective time-frequency analysis tool. For example, for a L(0,2) modal signal with a center frequency of 50kHz, Morlet wavelet can be selected as the mother wavelet for continuous wavelet transform. By analyzing the wavelet coefficient matrix, the time-frequency characteristics of different modes can be identified. At the same time, empirical mode decomposition (EMD) can decompose the signal into multiple intrinsic mode functions (IMFs). For example, for a mixed signal containing two modes, L(0,2) and T(0,1), EMD may decompose it into 3-5 IMFs, of which the first two IMFs may correspond to these two modes respectively. Dispersion compensation is a key step in restoring the original form of the signal. Taking the L(0,2) mode as an example, assume that the signal is significantly distorted after propagating 100m. From the dispersion curve, it can be seen that in the range of 40-50kHz, the phase velocity of this mode changes from 5200m / s to 5000m / s. Using this information, an anti-propagation model can be constructed to adjust the phases of different frequency components of the signal.For example, for the frequency component of 45kHz, its phase velocity is about 5100m / s, and the propagation time is 19.6ms, and its phase needs to be delayed by 2π×45000×0.0196≈5549 radians. By performing similar processing on all frequency components, the dispersion effect can be effectively eliminated and the signal waveform can be more concentrated. Multimodal fusion is an effective way to improve detection accuracy. For example, suppose there is a transverse crack 50m away from the transmitting end. The L(0,2) and T(0,1) modes may arrive at the defect position at different times and generate reflected waves. By measuring the arrival time difference of the reflected waves of these two modes, a positioning equation can be established. Assuming that the reflected wave of the L(0,2) mode arrives 0.5ms earlier than the T(0,1) mode, the defect position can be estimated based on their group velocity difference (about 1800m / s). At the same time, by analyzing the frequency change and energy attenuation of the reflected waves of each mode, the severity of the defect can be evaluated. For example, if the main frequency of the L(0, 2) modal reflection wave drops from the original 45kHz to 40kHz, and the energy attenuation exceeds 50%, it may indicate that this is a deeper transverse crack. Finally, the inspection results are visualized to generate an intuitive defect distribution map. For example, a heat map can be used to plot the defect probability distribution along the axial and circumferential directions of the pipeline, with different colors representing different defect severity levels. This visual representation can help maintenance personnel quickly locate problem areas and provide strong support for pipeline maintenance and decision-making.

[0033] Step S106, by extracting features and estimating parameters of the corrected guided wave signal, the location, size and type information of the pipeline defect is obtained. The preset thresholds of the defect parameters are determined according to the pipeline design specifications and safety requirements, including the percentage of the defect depth to the pipe wall thickness, the defect length and width, etc. If the detected defect parameters exceed the preset thresholds, it is determined that the pipeline has serious defects.

[0034] The finite element method is used to model the multi-layer heterogeneous pipeline, simulate the propagation of ultrasonic guided waves in the pipeline, and obtain the dispersion curves and modal distribution at different frequencies. By analyzing the dispersion curves and modal distribution, the excitation frequency and mode with the smallest propagation attenuation of guided waves in the pipeline and the strongest interaction with defects are selected. The pipeline is excited by ultrasonic guided waves of the selected frequency and mode, and the propagation characteristics of ultrasonic guided waves in the pipeline with defects are simulated and analyzed to extract the time domain waveform and spectrum characteristics of the defect reflection signal. The extracted defect reflection signal is denoised and enhanced by signal processing methods such as wavelet denoising and bandpass filtering to improve the signal-to-noise ratio and recognition degree of the signal. The processed defect reflection signal is classified and identified by support vector machine, and the type (such as cracks, corrosion, deformation, etc.) and size (length, width, depth) of the pipeline defect are judged according to the signal characteristics. The axial position and radial depth of the defect are calculated according to the arrival time difference and amplitude of the defect reflection signal, so as to achieve accurate positioning and quantitative evaluation of the pipeline defect. The defect parameters (type, position, size) obtained by the evaluation are compared with the preset defect severity criteria to determine the hazard level of the defect. The D-S evidence theory is used to fuse the defect information obtained from multiple detection positions and multiple modes to improve the accuracy and reliability of defect identification and positioning. The fused defect information will be used to generate a pipeline health assessment report. Generate a pipeline defect detection report, which includes basic pipeline information, detection parameter settings, defect identification and positioning results (defect type, axial position, radial depth, size), defect severity assessment, pipeline health status score, etc., and is intuitively displayed in the form of charts to provide a basis for pipeline maintenance and decision-making.

[0035] Specifically, finite element modeling of multi-layer heterogeneous pipelines is a complex task that requires accurate description of the physical properties and geometric parameters of each layer of material. Taking a three-layer composite pipeline as an example, the inner layer may be a high-temperature resistant alloy Inconel625, the middle layer is a ceramic fiber insulation material, and the outer layer is P91 steel. When modeling, it is necessary to consider parameters such as the density, elastic modulus, Poisson's ratio of each layer of material, as well as the inner diameter of the pipeline and the thickness of each layer. Through finite element analysis software such as ANSYS, a three-dimensional model of the pipeline can be constructed and meshed. When simulating the propagation of ultrasonic guided waves, it is necessary to set appropriate boundary conditions and excitation sources, such as applying axial or radial displacement excitation at one end of the pipeline. The acquisition of dispersion curves and modal distribution is crucial for selecting the optimal detection parameters. By calculating the phase velocity and group velocity at different frequencies, a dispersion curve can be drawn. For example, for the above three-layer pipeline, multiple modes such as L(0, 1), L(0, 2) and T(0, 1) may be observed in the frequency range of 20-100kHz. The T(0,1) mode usually exhibits non-dispersive characteristics, and the group velocity is close to the shear wave velocity. The L(0,2) mode may exhibit weak dispersion characteristics in certain frequency ranges, and the propagation velocity is high. According to the results of the dispersion curve analysis, the appropriate detection frequency and mode can be selected. When ultrasonic guided waves propagate in a defective pipeline, reflection, scattering, and mode conversion will occur. Through simulation analysis, the characteristics of the defect reflection signal can be observed. For example, for a transverse crack with a depth of 30% of the pipe wall thickness, the reflection signal of the L(0,2) mode may show a significant increase in amplitude and a decrease in frequency. By extracting these time domain waveforms and spectral features, a basis can be provided for subsequent defect identification. Signal processing is a key step to improve the accuracy of defect detection. Wavelet denoising can effectively remove high-frequency noise in the signal while retaining useful defect information. For example, using db4 wavelet to decompose the signal into 5 layers and perform soft threshold processing on the high-frequency coefficients can significantly improve the signal-to-noise ratio of the signal. Bandpass filtering can further enhance the defect characteristics within a specific frequency range. Support vector machine (SVM) is a powerful classification tool that can be used to identify defect types. By extracting the time domain and frequency domain features of the defect reflection signal, such as peak amplitude, main frequency, energy distribution, etc., a feature vector is constructed. Using these feature vectors to train the SVM model, the classification of different types of defects such as cracks, corrosion, and deformation can be achieved. At the same time, by establishing a regression model, the size parameters of the defect can be estimated. The positioning and quantitative evaluation of defects are the core goals of pipeline inspection. The axial position of the defect can be calculated by using the arrival time difference of the defect reflection signal combined with the group velocity of the guided wave. For example, if the reflected signal arrives 1ms later than the transmitted signal, and the group velocity of the guided wave is 5000m / s, it can be inferred that the defect is about 2.5m away from the transmitting end. The radial depth of the defect can be estimated by the ratio of the amplitude of the reflected signal to the incident signal, which requires the establishment of an appropriate theoretical or empirical model.The hazard level determination of defects is crucial to the safe operation of pipelines. Evaluation criteria can be established based on the type, size and location of defects, combined with the design parameters and operating conditions of the pipeline. For example, for pressure pipelines, the ASMEB31G standard can be used to calculate the residual strength coefficient based on the ratio of defect depth to pipe wall thickness and the ratio of defect length to pipe diameter, thereby determining the hazard level of the defect. D-S evidence theory provides an effective information fusion method that can integrate the results of multiple detection positions and multiple modes to improve the reliability of defect identification. For example, if the L (0, 2) mode and the T (0, 1) mode give different judgment results for defects at the same position, they can be weighted and fused according to their respective credibility to obtain more accurate evaluation results. The final generated pipeline defect detection report should be comprehensive and intuitive. The report should contain basic information about the pipeline, such as material, size, operating parameters, etc. The detection parameter setting section needs to explain the waveguide mode, frequency, excitation mode, etc. used. The defect identification and positioning results should list the detailed information of each detected defect in a tabular form. Heat maps or three-dimensional models can be used to visualize the defect distribution, so that maintenance personnel can intuitively understand the health status of the pipeline. In addition, the overall health score of the pipeline should also be included to provide a basis for management decisions.

[0036] Step S107, comprehensively analyze the detection results under different excitation frequencies and guided wave modes, and select the most sensitive detection parameter combination for pipeline defects. According to the defect detection results and parameter optimization results, a high-temperature chemical pipeline defect detection scheme suitable for magneto-ultrasonic guided waves is formed to guide pipeline health monitoring and safety assessment. For pipelines judged to have serious defects, repair or replacement suggestions are put forward.

[0037] According to the material properties and geometric dimensions of the multi-layer heterogeneous pipeline, the mathematical model of the pipeline is established, and the finite element method is used to discretize the pipeline. By solving the dispersion equation of ultrasonic guided waves in the pipeline, the dispersion curves and modal characteristics of the guided waves under different modes are obtained. The slope and group velocity changes of the dispersion curve are analyzed to determine the dispersion characteristics of the guided waves under different frequencies, and the frequency range with smaller dispersion is selected as the excitation frequency. The displacement and stress distribution of guided waves in different modes are analyzed according to the modal characteristics, and the guided wave modes sensitive to defects are determined. Defects of different types and sizes, such as cracks and corrosion, are introduced into the pipeline model, and the defects are simulated by setting the corresponding material properties and geometric parameter changes in the finite element model. The finite difference time domain method is used to solve the guided wave field in the defective pipeline model. Specifically, the wave equation is discretized and the time and space derivatives are approximated using the central difference format, and the evolution of the guided wave field in time and space is iteratively solved. The guided wave signal of the defective pipeline is obtained, and the time-frequency characteristics of the defect reflection signal are extracted. By analyzing the amplitude, phase and other parameters of the defect reflection signal under different excitation frequencies and guided wave modes, the sensitivity of the detection parameters to the defects is evaluated. The defect detection results under different detection parameter combinations are comprehensively compared, and the optimal parameter combination is searched using a genetic algorithm. By constructing an appropriate fitness function, considering multiple objectives such as detection sensitivity and resolution, multiple generations of evolutionary iterations are performed to obtain a high-temperature chemical pipeline defect detection scheme suitable for magneto-ultrasonic guided waves. According to the optimized detection scheme, the pipeline is tested by guided waves. If the detection results show that the pipeline has serious defects, a recommendation report for pipeline repair or replacement is generated to guide subsequent pipeline health monitoring and safety assessment work.

[0038] Specifically, the mathematical modeling of multi-layer heterogeneous pipelines needs to consider the physical properties and geometric parameters of each layer of materials. Taking a three-layer composite pipeline as an example, the inner layer may be a high-temperature resistant alloy Inconel625, the middle layer is a ceramic fiber insulation material, and the outer layer is P91 steel. When modeling, it is necessary to input parameters such as the density, elastic modulus, Poisson's ratio of each layer of material, as well as the inner diameter of the pipeline and the thickness of each layer. The finite element method is used to discretize the pipeline, and software such as ANSYS can be used to build a three-dimensional model and perform meshing. Solving the dispersion equation is the key to obtaining the propagation characteristics of guided waves. By calculating the phase velocity and group velocity at different frequencies, a dispersion curve can be drawn. For example, for the above three-layer pipeline, multiple modes such as L(0,1), L(0,2) and T(0,1) may be observed in the frequency range of 20-100kHz. The T(0,1) mode usually exhibits non-dispersive characteristics, and the group velocity is close to the shear wave velocity. However, the L(0,2) mode may exhibit weak dispersion characteristics in certain frequency ranges and have a higher propagation velocity. Choosing the right excitation frequency and mode is crucial for defect detection. By analyzing the slope of the dispersion curve and the change in group velocity, the frequency range with smaller dispersion can be determined. For example, for the L(0,2) mode, it may show smaller dispersion in the range of 50-70kHz. Modal feature analysis can reveal the displacement and stress distribution of guided waves in the pipe wall. For detecting outer wall defects, the mode with larger outer wall stress may be selected. Defect simulation can be achieved by changing the local material properties or geometric parameters in the finite element model. For example, when simulating corrosion defects, the unit thickness can be reduced in a specific area; when simulating cracks, the fracture surface can be introduced into the grid. The time-domain finite difference method is used to solve the guided wave field in the defective pipeline model. It is necessary to set an appropriate time step and spatial grid size to ensure the calculation stability and accuracy. The time-frequency feature extraction of the defect reflection signal is the basis for evaluating the sensitivity of the detection parameters. For example, for a transverse crack with a depth of 30% of the pipe wall thickness, the reflection signal of the L(0,2) mode may show a significant increase in amplitude and a decrease in frequency. By comparing the reflection signal characteristics under different excitation frequencies and modes, the best combination of detection parameters can be determined. The application of genetic algorithms can effectively optimize the detection scheme. The fitness function can comprehensively consider factors such as detection sensitivity, resolution and signal-to-noise ratio. For example, the signal-to-noise ratio of the defect reflection signal to the background noise can be used as an evaluation indicator, and the correlation between the reflection signal amplitude and the defect size can be used as another indicator. Through multiple generations of evolutionary iterations, the optimal combination of detection parameters under given conditions can be obtained. After the final detection scheme is implemented, if serious defects are found, a detailed repair or replacement recommendation report needs to be generated. The report should include information such as the location, type, size, etc. of the defect, as well as the evaluation results of the remaining strength of the pipeline.For example, for pressure pipelines, the ASMEB31G standard can be used to calculate the residual strength coefficient based on the ratio of defect depth to pipe wall thickness, and the ratio of defect length to pipe diameter, so as to determine the hazard level of the defect and make corresponding treatment suggestions.

[0039] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the concept of the present application. For example, the above features are replaced with the technical features with similar functions disclosed in the present application (but not limited to) to form a technical solution.

Claims

1. A method for high-temperature chemical pipeline defect detection using magnetically induced ultrasonic guided waves, characterized in that: include: Establish a discretized mathematical model of a multi-layer heterogeneous high-temperature pipeline, obtain the temperature field distribution data of the pipeline under a high-temperature environment, calculate the sound velocity and attenuation coefficient of the pipeline material at different temperatures and assign them to each grid unit of the discretized mathematical model; Apply magnetostrictive excitation signal, calculate the propagation process of guided wave in the pipeline through numerical simulation, and obtain the amplitude and phase information of guided wave signal; Performing time-frequency domain analysis on the waveguide signal, extracting the group velocity and phase velocity of different mode waveguides, determining the dispersion characteristics of the waveguide, and evaluating the impact of the high temperature environment on the waveguide propagation; The received waveguide signal is processed by using a modal decomposition and reconstruction algorithm to obtain a corrected waveguide signal; Perform feature extraction and parameter estimation on the corrected waveguide signal to obtain the location, size and type information of the pipeline defect. If the detected defect parameter exceeds a preset threshold, it is determined that the pipeline has a serious defect. The detection results under different excitation frequencies and guided wave modes are comprehensively analyzed, and the detection parameter combination most sensitive to pipeline defects is selected to form a high-temperature chemical pipeline defect detection solution suitable for magneto-ultrasonic guided waves.

2. The high-temperature chemical pipeline defect detection method using magnetic ultrasonic guided waves as claimed in claim 1 is characterized in that: The discrete mathematical model of the multi-layer heterogeneous high-temperature pipeline is established, including: According to the structural parameters and material properties of multi-layer heterogeneous high-temperature pipelines, a three-dimensional geometric model of the pipeline is established; The three-dimensional geometric model is meshed using a finite element method to obtain a discretized mathematical model of the pipeline structure; Based on the boundary conditions and initial conditions of the pipeline, the dynamic equation of the pipeline structure is established; The finite element method is used to solve the dynamic equation, obtain the propagation characteristics of ultrasonic guided waves in the pipeline structure, including phase velocity and group velocity, and draw a dispersion curve; According to the dispersion curve, a suitable frequency range is selected to avoid severe dispersion areas and determine the excitation frequency; The modal decomposition method is used to decompose the ultrasonic guided wave into different guided wave modes, and the vibration shape and energy distribution of each mode are obtained.

3. The high-temperature chemical pipeline defect detection method using magnetic ultrasonic guided waves as claimed in claim 1 is characterized in that: The step of obtaining the temperature field distribution data of the pipeline in a high temperature environment, calculating the sound velocity and attenuation coefficient of the pipeline material at different temperatures and assigning them to each grid unit of the discretized mathematical model includes: Obtain the boundary conditions of the pipeline in a high temperature environment, including the convection heat transfer coefficient of the inner and outer surfaces of the pipeline and the ambient temperature; The finite element thermal analysis module is used to consider the thermal conductivity and specific heat capacity parameters of the pipeline material, calculate the temperature field distribution of the pipeline under steady-state conditions, and obtain the temperature data of each grid unit of the pipeline; According to the temperature field distribution data, using a finite element thermal stress analysis module, the stress and strain distribution of the pipeline at different temperatures is calculated; Based on the elastic modulus and Poisson's ratio parameters of the pipeline material and the temperature field distribution data, the finite element acoustic module is used to solve the sound wave propagation equation and calculate the sound velocity distribution of the pipeline material at different temperatures; The damping characteristics of the pipeline material are introduced into the sound wave propagation equation, and the viscous damping coefficient of the material is considered to calculate the distribution of the sound wave attenuation coefficient of the pipeline material at different temperatures.

4. The high-temperature chemical pipeline defect detection method using magnetic ultrasonic guided waves as claimed in claim 1 is characterized in that: The magnetostrictive excitation signal is applied, and the propagation process of the guided wave in the pipeline is calculated by numerical simulation to obtain the amplitude and phase information of the guided wave signal, including: According to different guided wave excitation frequencies and modes, a magnetostrictive excitation signal is applied to the incident end of the discretized mathematical model, and the parameters of the magnetostrictive excitation signal include center frequency, bandwidth, amplitude and duration; The propagation process of the guided wave in the discretized mathematical model is calculated by a finite element numerical simulation method to obtain the amplitude and phase information of the guided wave signal at different positions; Performing time-frequency analysis on the waveguide signal by short-time Fourier transform to obtain a dispersion curve and a group velocity curve of the waveguide propagating in the pipeline; Analyze the dispersion curve and group velocity curve, judge the propagation characteristics of the guided wave under different frequencies and modes, and determine the detection frequency and mode; Performing modal decomposition on the waveguide signal using an empirical mode decomposition method to obtain waveguide signal components of different modes; Analyze the amplitude and phase characteristics of different mode guided wave signals to determine the propagation characteristics and attenuation laws of each mode.

5. The high-temperature chemical pipeline defect detection method using magnetic ultrasonic guided waves as claimed in claim 1, characterized in that: The time-frequency domain analysis of the waveguide signal is performed to extract the group velocity and phase velocity of different mode waveguides, determine the dispersion characteristics of the waveguide, and evaluate the influence of the high temperature environment on the waveguide propagation, including: Based on the established pipeline mechanical model, the semi-analytical finite element method is used to derive the dispersion equation of ultrasonic guided waves in the pipeline, and the dispersion curves of guided waves of different modes are obtained by solving the dispersion equation, including the relationship between group velocity and phase velocity and frequency. The propagation process of ultrasonic guided waves in multi-layer heterogeneous pipelines is simulated by numerical simulation software, and the displacement field and stress field distribution of guided waves of different modes when propagating in the pipeline are obtained; The waveguide signal obtained by numerical simulation is analyzed in the time and frequency domain using numerical calculation software, and the time and frequency characteristics of different mode waveguides are extracted using short-time Fourier transform to determine the dispersion characteristics and multi-modal characteristics of the waveguide signal. According to the time-frequency characteristics of the waveguide signal, the curves of group velocity and phase velocity of different mode waveguides varying with frequency are determined and compared with the theoretically calculated dispersion curves to verify the numerical simulation results.

6. The high-temperature chemical pipeline defect detection method using magneto-ultrasonic guided waves according to claim 1, characterized in that: The method of using a modal decomposition and reconstruction algorithm to process the received waveguide signal to obtain a corrected waveguide signal includes: The semi-analytical finite element method is used to solve the dispersion equation of ultrasonic guided waves in the pipeline, and the dispersion curves and mode shapes of guided waves in different modes are obtained. According to the dispersion curve, the detection frequency range is determined, and a mode with a long propagation distance and high sensitivity is selected as the detection mode; Wavelet transform is used to perform time-frequency analysis on the received waveguide signal to extract the arrival time, amplitude and frequency characteristic parameters of different modes; Using continuous wavelet transform to perform time-frequency representation on the waveguide signal to obtain wavelet coefficient matrices at different scales; By analyzing the wavelet coefficient matrix, the waveguide signal components of different modes are identified; Decomposing the waveguide signal into a plurality of intrinsic mode functions by using an empirical mode decomposition method to separate signals of different modes; The separated modal signals are subjected to dispersion compensation to correct the waveform distortion caused by dispersion and restore the original form of the signal.

7. The high-temperature chemical pipeline defect detection method using magnetic ultrasonic guided waves as claimed in claim 1 is characterized in that: The feature extraction and parameter estimation of the corrected guided wave signal are performed to obtain the location, size and type information of the pipeline defect. If the detected defect parameter exceeds a preset threshold, it is determined that the pipeline has a serious defect, including: Using ultrasonic guided waves of selected frequency and mode to excite the pipeline, the propagation characteristics of ultrasonic guided waves in the pipeline with defects are analyzed through simulation, and the time domain waveform and spectrum characteristics of the defect reflection signal are extracted; Signal processing methods are used to reduce noise and enhance features of the extracted defect reflection signals to improve the signal-to-noise ratio and recognition degree of the signals; The processed defect reflection signal is classified and identified using support vector machine, and the type and size of pipeline defects are determined based on signal characteristics; The axial position and radial depth of the defect are calculated based on the arrival time difference and amplitude of the defect reflection signal, so as to achieve accurate positioning and quantitative evaluation of pipeline defects; Compare the evaluated defect parameters with the preset defect severity criteria to determine the hazard level of the defect; The D-S evidence theory is used to fuse the defect information obtained from multiple detection positions and multiple modes to improve the accuracy and reliability of defect identification and positioning.

8. The high-temperature chemical pipeline defect detection method using magnetic ultrasonic guided waves as claimed in claim 1, characterized in that: The comprehensive analysis of the detection results under different excitation frequencies and guided wave modes selects the most sensitive detection parameter combination for pipeline defects, and forms a high-temperature chemical pipeline defect detection scheme suitable for magneto-ultrasonic guided waves, including: Analyze the slope of the dispersion curve and the change of group velocity, determine the dispersion characteristics of the waveguide at different frequencies, and select the frequency range with smaller dispersion as the excitation frequency; Analyze the displacement and stress distribution of guided waves of different modes according to the modal characteristics, and determine the guided wave modes that are sensitive to defects; Introduce defects of different types and sizes into the pipeline model and simulate the defects by setting corresponding material properties and geometric parameter changes in the finite element model; The finite-difference time-domain method is used to solve the guided wave field in the defective pipeline model, and the evolution of the guided wave field in time and space is iteratively solved; Obtain the guided wave signal of the defective pipeline and extract the time-frequency characteristics of the defect reflection signal; By analyzing the parameters of defect reflection signals under different excitation frequencies and guided wave modes, the sensitivity of detection parameters to defects is evaluated; The defect detection results under different detection parameter combinations are comprehensively compared, and the optimal parameter combination is searched using a genetic algorithm to obtain a defect detection scheme for high-temperature chemical pipelines using magneto-ultrasonic guided waves.

9. A high-temperature chemical pipeline defect detection system based on magnetically induced ultrasonic guided waves, characterized in that: Including mathematical model building module, temperature field analysis module, sound velocity and attenuation coefficient calculation module, magnetostrictive excitation and simulation module, guided wave signal analysis module, defect detection and evaluation module and optimized detection plan generation module; Mathematical model building module, used to build a discretized mathematical model of multi-layer heterogeneous high-temperature pipelines; A temperature field analysis module is used to obtain temperature field distribution data of the pipeline in a high temperature environment, calculate the sound velocity and attenuation coefficient of the pipeline material at different temperatures, and assign them to each grid unit of the discretized mathematical model; The sound velocity and attenuation coefficient calculation module is used to apply magnetostrictive excitation signals, calculate the propagation process of guided waves in the pipeline through numerical simulation, and obtain the amplitude and phase information of the guided wave signal; A magnetostrictive excitation and simulation module is used to perform time-frequency domain analysis on the waveguide signal, extract the group velocity and phase velocity of different mode waveguides, determine the dispersion characteristics of the waveguide, and evaluate the impact of the high temperature environment on the waveguide propagation; The waveguide signal analysis module is used to process the received waveguide signal using a modal decomposition and reconstruction algorithm to obtain a corrected waveguide signal; A defect detection and evaluation module is used to extract features and estimate parameters of the corrected waveguide signal to obtain the location, size and type information of the pipeline defect. If the detected defect parameter exceeds a preset threshold, it is determined that the pipeline has a serious defect; The detection scheme generation module is optimized to comprehensively analyze the detection results under different excitation frequencies and guided wave modes, select the detection parameter combination that is most sensitive to pipeline defects, and form a high-temperature chemical pipeline defect detection scheme suitable for magneto-ultrasonic guided waves.

Images