Buried liquid-filled water supply pipeline guided wave detection frequency and mode adaptive optimization method

Abstract

The application discloses a buried liquid-filled water supply pipeline guided wave detection frequency and mode adaptive optimization method, which can quickly calculate the dispersion characteristics and attenuation characteristics of the buried liquid-filled water supply pipeline according to the parameters of the to-be-detected pipeline geometry, material and service environment, and automatically output a recommended excitation frequency, guided wave mode, detection time window and maximum detection distance estimation method, thereby providing a unified basis for subsequent defect identification, detection distance improvement and field layout.

Description

Technical Field

[0001] This invention relates to an adaptive optimization method for frequency and mode selection in guided wave testing of buried liquid-filled water supply pipelines, belonging to the technical fields of non-destructive testing of municipal water supply pipelines, ultrasonic guided wave testing, guided wave dispersion analysis, and health monitoring of underground pipe network structures. Background Technology

[0002] Urban water supply pipelines operate in complex environments, including underground, water-filled, and pressurized conditions. Affected by soil conditions, moisture intrusion, and pipeline aging, these pipelines are prone to structural damage such as external corrosion and fatigue cracks. Because the pipelines are buried underground, structural defects are often concealed. The accumulation of structural damage can eventually lead to pipeline leaks, causing economic losses, water waste, and potentially resulting in road flooding or collapse, equipment damage, traffic congestion, and urban water supply interruptions. Guided wave detection technology holds great potential for screening such pipeline structural damage. Its principle involves exciting low-frequency ultrasonic guided waves propagating axially along the pipe wall through a sensor ring. When these guided waves encounter defects such as changes in pipe cross-section or localized cracks, they generate echoes. Based on the echo propagation speed, arrival time, and signal amplitude, the distance between the sensor ring and the defect can be calculated, enabling defect localization.

[0003] Existing guided wave testing cases mostly focus on exposed pipelines (such as process pipelines). However, urban water supply pipelines are buried, filled with water, and operate under pressure for extended periods. The outer wall of the pipe is constrained by the soil, cladding layer, and groundwater environment, resulting in significant fluid-structure interaction between the fluid inside the pipe and the pipe wall. Compared to empty pipes above ground or in laboratory settings, guided wave propagation in buried liquid-filled water supply pipelines exhibits increased modal numbers, enhanced dispersion, energy leakage through radiation to the external soil, and a significantly shortened propagation distance.

[0004] In current guided wave testing engineering practices, testing personnel typically select excitation parameters based on the dispersion curve of empty pipes, empirical frequencies, or equipment-recommended frequency bands. This method fails to fully account for the coupling effects of factors such as pipe diameter, wall thickness, pipe material, water inside the pipe, external soil impedance, burial depth, and cladding layer. This leads to problems on-site, including inappropriate target mode selection, low signal-to-noise ratio of defect echoes, large deviations in propagation distance estimation, and unstable applicability of the same frequency in different pipe sections.

[0005] Semi-analytical finite element method can efficiently solve the dispersion curve of waveguide structure, but traditional applications are mostly concentrated on hollow cylinders, above-ground pipelines or single boundary conditions. For the combined working conditions of pipe wall, water body, soil and cladding layer in municipal buried liquid-filled water supply pipelines, there is still a lack of an adaptive optimization method that can be oriented towards field detection tasks and uniformly complete coupled dispersion calculation, mode tracking, index scoring, frequency band recommendation and detection distance estimation.

[0006] Therefore, there is an urgent need for a method that can quickly calculate the dispersion and attenuation characteristics of buried liquid-filled water supply pipelines based on the geometry, materials, and service environment parameters of the pipeline to be inspected, and automatically output recommended excitation frequency, guided wave mode, detection time window, and maximum detection distance estimation, so as to provide a unified basis for subsequent defect identification, detection distance improvement, and field deployment. Summary of the Invention

[0007] The technical problem to be solved by this invention is to provide a rapid adaptive optimization method for frequency and mode selection in guided wave detection of buried liquid-filled water supply pipelines, so as to solve the problems in existing guided wave detection, such as frequency and mode selection relying on experience, difficulty in considering the influence of pipe-soil-water coupling, inaccurate detection distance estimation, and difficulty in quickly configuring on-site detection parameters.

[0008] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: an adaptive optimization method for guided wave detection frequency and mode of buried liquid-filled water supply pipelines, comprising the following steps: Step 1: Collect the basic parameters of the water supply pipeline to be tested, including four categories: (a) Pipeline geometric parameters: pipe diameter, wall thickness; (b) Pipeline material parameters: elastic modulus, Poisson's ratio, density; (c) Fluid parameters inside the pipe: water density, sound velocity, operating pressure; (d) External soil and covering layer parameters: soil density, material damping ratio, shear modulus, complex shear modulus, burial depth, covering layer thickness, and material parameters; Enter the above parameters into the parameter input module; Step 2: In the pipe-soil-water coupling modeling module, establish a three-level comparative working condition model based on the above parameters. The three-level comparative working conditions include an empty pipe model, a liquid-filled pipe model, and a buried liquid-filled pipe model. In the buried liquid-filled pipe model, set the water acoustic domain, pipe wall elastic domain, covering layer domain, and external soil equivalent impedance boundary. Step 3: Dispersion calculation; In the SAFE dispersion solution module, the pipe cross-section is discretized based on the semi-analytical finite element method. Analytical waveforms are used in the axial direction, where represents the radial, circumferential, and axial coordinates, respectively, t represents time, k represents the wave number, and represents the frequency. The pipe-soil-water coupled dispersion eigenvalue problem is constructed, and the complex wave number at different frequencies is solved. Step 4: Modal tracking; Calculate the phase velocity, group velocity, attenuation coefficient, dispersion broadening, wall energy proportion, water energy proportion, soil leakage energy proportion, and modal morphology correlation coefficient for each guided wave mode based on the complex wave number; Step 5: Using the modal scoring module, establish a comprehensive scoring function for each candidate mode; the comprehensive scoring function S can be expressed as formula (1): (1) in For low decay score, For weak dispersion scoring, Assess the energy concentration at the pipe wall. For equivalent defect sensitivity scoring, Scoring the project's incentive potential. to These are the weighting coefficients; The scoring weights are adjusted according to the type of inspection task or engineering application scenario. When the inspection task is long-distance screening, the weights of low attenuation score and weak dispersion score are increased. When the inspection task is pitting corrosion or local wall thinning defect identification, the weights of equivalent defect disturbance sensitivity score and pipe wall energy concentration score are increased. Among these, the weights corresponding to complex wave number disturbance index and candidate defect area strain energy concentration index can be increased. When the inspection task is rapid on-site inspection, the weight of engineering excitability score is increased. Step 6: The detection parameter output module outputs the recommended results, which include the preferred mode, recommended frequency range, center frequency, bandwidth, excitation waveform suggestion, reception time window, theoretical maximum detection distance, high-frequency dispersion areas to be avoided, and on-site calibration suggestions. When the candidate frequency is located near the mode cutoff frequency, in the mode crossover area, in the group velocity change area, in the attenuation peak area, or in the low-efficiency frequency band of the equipment, the candidate frequency is marked as a forbidden frequency band or a reduced-weighted frequency band.

[0009] As a preferred approach, step 7 is also included: after the guided wave detection is completed, the calculation results are corrected by experimental or field baseline signals, the measured group velocity and attenuation coefficient are extracted using end echo, weld echo or known structural features, and the attenuation score, dispersion score and the weighting coefficients of the attenuation score and dispersion score in step 5 are corrected.

[0010] As a preferred embodiment, the establishment of the three-level comparative working condition model specifically includes: Level 1, Empty Pipe Model: Only considering the elastic domain of the pipe wall Stress in the pipe wall -strain The relationship satisfies the linear elastic constitutive equation ,in Represents the fourth-order elastic tensor; The second level, the liquid-filled tube model: This adds an acoustic domain to the water inside the tube, based on the empty tube design. Sound pressure field of water body Satisfying the Helmholtz equation ,in On the inner surface of the pipe wall The fluid-structure interaction boundary conditions are satisfied at this location. ; Level 3, Buried Liquid-Filled Pipe Model: Adding a Coating Layer Domain Equivalent impedance boundary with external soil Equivalent impedance boundary conditions are applied to the outer surface of the coating layer using a viscoelastic equivalent impedance boundary model. The inner surface of the pipe wall satisfies the fluid-structure interaction boundary condition, the outer surface of the cladding layer satisfies the equivalent impedance boundary condition, and the contact surface between the pipe wall and the cladding layer satisfies the displacement and stress continuity condition.

[0011] As a preferred embodiment, the SAFE dispersion solving module solves the dispersion eigenvalue problem of the pipe-soil-water coupled system based on the semi-analytical finite element method. The specific process is as follows: For the elastic domain and the cladding domain of the pipe wall, the displacement field at any point on the pipe cross-section is expressed as: ,That Represented by finite element discretization as For the acoustic domain of a water body, the sound pressure field is expressed as: ; Substituting the semi-analytical displacement field and sound pressure field into the governing equations and coupled boundary conditions of each domain, the dispersion eigenvalue problem is obtained through the Galerkin weighted residual method: (2) in , , Here is the stiffness matrix. For the quality matrix, For a given angular frequency, the vector represents the generalized degrees of freedom. Solving for a set of complex wave numbers yields a set of complex wave numbers. ; Calculate the dispersion parameters of each mode based on the complex wavenumber: Phase velocity:

[0012] Group speed:

[0013] Attenuation coefficient: (dB / m) Dispersion broadening: =

[0014] Pipe wall energy percentage: .

[0015] As a preferred embodiment, the modal tracking module is used to establish continuous tracking relationships between modes within the frequency domain; for adjacent frequency points... and Calculate the modal morphology correlation coefficient (MSCC): (3) in This represents the confidence level between the m-th mode and the n-th mode. and They represent the frequencies respectively. The next m First-order mode vector, Represents frequency Next n The modal vector is of order H, where H represents the conjugate transpose. A modal matching cost function is constructed based on the modal morphology correlation coefficient, phase velocity continuity, group velocity continuity, and pipe wall energy proportion continuity. Candidate modes at adjacent frequency points are consecutively numbered.

[0016] As a preferred approach, the specific process for establishing a comprehensive scoring function for each candidate mode is as follows: Before scoring, constraints are first screened, and those that do not meet the minimum pipe wall energy percentage are excluded. Maximum permissible attenuation coefficient Constrained modes are directly excluded; For the modes that pass the constraint screening, calculate five sub-scores: (a) Decay score: ; (b) Dispersion score: ; (c) Energy score: , in These are the weights for the proportion of pipe wall energy and the weights for soil leakage inhibition, respectively. (d) Equivalent Defect Disturbance Sensitivity Scoring: For pitting, localized corrosion, or wall thinning defects to be identified, an equivalent micro-wall thickness disturbance is set in the candidate defect region of the pipe wall; the equivalent micro-wall thickness disturbance is used to characterize the local stiffness and mass changes caused by the defect, without the need to establish a complete defect scattering model; the defect-free baseline model and the model after applying the equivalent micro-wall thickness disturbance are solved respectively to obtain the first... Candidate modes at frequency Reference complex wavenumber at the location and the complex wave number after perturbation And calculate the normalized complex wavenumber perturbation index: (4) In the formula, This is the equivalent wall thickness reduction amount. This represents the original wall thickness of the pipe. Calculate the modal strain energy lumped index of the candidate defect region: (5) In the formula, This is a candidate defect region. For the entire pipe wall area, For the first Strain energy density of the first mode in the tube wall; Calculate the percentage of normal displacement on the candidate defect surface: (6) In the formula, The pipe wall surface where the candidate defect is located. This represents the normal displacement of the surface; when the candidate defect is located on the inner or outer surface of the pipe, the normal displacement of the corresponding surface is used respectively. No. The equivalent defect perturbation sensitivity score for the first mode is: (7) In the formula, , and These are the weighting coefficients for the proportions of complex wavenumber disturbance, strain energy concentration, and normal displacement, respectively, and they satisfy the following: (8) The normalization function is shown in formula (9); The larger the value, the more sensitive the corresponding frequency and mode are to the target defect; (9) in, and These represent the current candidate frequency and the minimum and maximum values ​​in the modality set, respectively. A positive number set to prevent the denominator from being zero; (e) Motivational rating: ,in Exciter spatial distribution and the first The normalized coupling efficiency of the mode shape is related to the actual sensor loop, excitation direction, circumferential order, and device bandwidth. The weights of the comprehensive scoring function are adaptively adjusted according to the type of inspection task: a weight vector is generated based on the type of inspection task, the type of target defect, the equipment bandwidth, and the on-site calibration error. And normalize the weights so that ; Meanwhile, for the three-level comparison working conditions, the design working condition difference indicators are as follows: (10) (11) in,, , and These represent the group velocities of guided waves under three operating conditions: empty pipe, liquid-filled pipe, and buried liquid-filled pipe. , and These represent the attenuation of guided waves under three operating conditions: empty pipe, liquid-filled pipe, and buried liquid-filled pipe. When the operating condition difference index of a certain frequency band exceeds the threshold, the frequency band is marked as an environmentally sensitive frequency band, and its recommended priority is reduced.

[0017] As a preferred embodiment, the theoretical maximum detection distance estimate in single-ended pulse echo mode is: (12) The estimated theoretical maximum detection distance for transmission detection is as follows: (13) in and All are expressed in dB. The equivalent amplitude reflection coefficient of the defect. The unit propagation attenuation coefficient is expressed in dB / m.

[0018] As a preferred embodiment, step 7 involves correcting the theoretical results using experimental or field baseline signals, specifically including: When the group velocity relative error and the relative error of the attenuation coefficient If all parameters are less than the preset threshold, the final recommended parameters are output; otherwise, the soil impedance, covering layer parameters, or coupling state parameters are updated, and the dispersion solution and modal scoring are re-executed. If the single-point correction method is used, then the end reflection and correction factor are utilized. After correction ; ,in Represents the measured peak amplitude. Represents the theoretical peak amplitude. Represents the corrected attenuation coefficient. Represents the theoretical attenuation coefficient. This represents the distance between the broadcast excitation point and the end of the pipeline; If a multi-point correction method is used, a more accurate attenuation coefficient is fitted using the least squares method by utilizing multiple known structural echoes.

[0019] As a preferred approach, for a new pipe segment to be tested, the data storage and feedback module calculates the normalized working condition distance between it and historical pipe segments based on pipe diameter, wall thickness, material parameters, fluid parameters, burial depth, soil parameters, and cladding layer parameters. Based on the working condition distance from smallest to largest, similar historical working conditions are selected from the database, and the recommended frequency range, preferred modes, and correction parameters of these similar historical working conditions are called as the initial values ​​for current dispersion calculation, mode tracking, and the establishment of a comprehensive scoring function. Guided wave detection frequency and mode selection then begin from step 3.

[0020] As a preferred approach, after completing on-site testing, the measured group velocity is determined... and measured attenuation coefficient Calculate the correction error: (14) (15) In the formula, and These represent the calculated group velocity and decay coefficient, respectively. when or If the error exceeds the corresponding error threshold, update the soil equivalent impedance, the covering layer damping parameters or the scoring weight, and re-execute the dispersion solution and modal scoring; if neither error exceeds the corresponding error threshold, write the final optimization result and correction parameters of the current pipe segment into the database for subsequent similar pipe segments to use.

[0021] The beneficial effects of this invention are: it can transform the frequency and mode selection process, which originally relied on experience, into a calculable, comparable, and calibrable engineering process; it can avoid directly extrapolating the parameters of empty pipes or above-ground pipelines to buried liquid-filled water supply pipelines; and it can provide unified frequency, mode, and attenuation boundary parameters for pitting corrosion identification, improved detection distance, and multi-well deployment in the field. Attached Figure Description

[0022] Figure 1 System architecture module diagram.

[0023] Figure 2 : Diagram of the cross-sectional model of the buried liquid-filled pipeline coupled with soil and water.

[0024] Figure 3 Schematic diagram of SAFE cross-section discretization and axial analysis.

[0025] Figure 4 Example of a comparison chart of dispersion curves under different operating conditions (the black dashed line represents the first mode of torsional wave T(0,1), the black solid line represents the first mode of longitudinal wave L(0,1), and the gray solid line represents the second mode of longitudinal wave L(0,2)).

[0026] Figure 5 Example of attenuation curve comparison under different operating conditions (the black dashed line represents the first mode of torsional wave T(0,1), the black solid line represents the first mode of longitudinal wave L(0,1), and the gray solid line represents the second mode of longitudinal wave L(0,2)).

[0027] Figure 6 Experimental calibration procedure. Detailed Implementation

[0028] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings; like Figure 1-6 As shown, a method for adaptive optimization of guided wave detection frequency and mode for buried liquid-filled water supply pipelines includes the following steps: Step 1: Collect the basic parameters of the water supply pipeline to be tested, including four categories: (a) Pipe geometric parameters: pipe diameter Wall thickness (b) Pipe material parameters: elastic modulus Poisson's ratio ,density (c) Fluid parameters inside the pipe: water density Speed ​​of sound Operating pressure (d) External soil and covering layer parameters: soil density Material damping ratio shear modulus Complex shear modulus burial depth Coating thickness And material parameters; enter the above parameters into the parameter input module.

[0029] Step 2: In the pipe-soil-water coupling modeling module, establish a three-level comparative working condition model based on the above parameters. The three-level comparative working conditions include an empty pipe model, a liquid-filled pipe model, and a buried liquid-filled pipe model. In the buried liquid-filled pipe model, set the water acoustic domain, the pipe wall elastic domain, the coating layer domain, and the external soil equivalent impedance boundary.

[0030] The establishment of the three-level comparative working condition model specifically includes: Level 1, Empty Pipe Model: Only considering the elastic domain of the pipe wall Stress in the pipe wall -strain The relationship satisfies the linear elastic constitutive equation ,in Represents the fourth-order elastic tensor; The second level, the liquid-filled tube model: This adds an acoustic domain to the water inside the tube, based on the empty tube design. Sound pressure field of water body Satisfying the Helmholtz equation ,in On the inner surface of the pipe wall The fluid-structure interaction boundary conditions are satisfied at this location. ; Level 3, Buried Liquid-Filled Pipe Model: Adding a Coating Layer Domain Equivalent impedance boundary with external soil Equivalent impedance boundary conditions are applied to the outer surface of the coating layer using a viscoelastic equivalent impedance boundary model. The inner surface of the pipe wall satisfies the fluid-structure interaction boundary condition, the outer surface of the cladding layer satisfies the equivalent impedance boundary condition, and the contact surface between the pipe wall and the cladding layer satisfies the displacement and stress continuity condition.

[0031] Step 3: Dispersion Calculation; In the SAFE dispersion solver module, for the pipe wall elastic domain and the cladding layer domain, the displacement field at any point on the pipe cross-section is expressed as: ,in These represent radial, circumferential, and axial coordinates, respectively; t represents time; and k represents the wavenumber. Represents frequency, its Represented by finite element discretization as For the acoustic domain of a water body, the sound pressure field is expressed as: ; Substituting the semi-analytical displacement field and sound pressure field into the governing equations and coupled boundary conditions of each domain, the dispersion eigenvalue problem is obtained through the Galerkin weighted residual method: (2) in , , Here is the stiffness matrix. For the quality matrix, For a given angular frequency, the vector represents the generalized degrees of freedom. Solving for a set of complex wave numbers yields a set of complex wave numbers. ; Calculate the dispersion parameters of each mode based on the complex wavenumber: Phase velocity:

[0032] Group speed:

[0033] Attenuation coefficient: (dB / m) Dispersion broadening: =

[0034] Pipe wall energy percentage: .

[0035] Step 4: Modal tracking; based on complex wavenumbers Calculate the phase velocity of each guided wave mode. Group speed attenuation coefficient , frequency dispersion broadening Pipe wall energy ratio Water body energy ratio Percentage of energy leaked from soil and modal morphology correlation coefficient; For adjacent frequency points and Calculate the modal morphology correlation coefficient (MSCC): (3) in This represents the confidence level between the m-th mode and the n-th mode. and They represent the frequencies respectively. The next m First-order mode vector, Represents frequency Next n The modal vector is of order H, where H represents the conjugate transpose. A modal matching cost function is constructed based on the modal morphology correlation coefficient, phase velocity continuity, group velocity continuity, and pipe wall energy proportion continuity. Candidate modes at adjacent frequency points are consecutively numbered.

[0036] Step 5: Using the modal scoring module, establish a comprehensive scoring function for each candidate mode; the comprehensive scoring function S is expressed as formula (1): (1) in For low decay score, For weak dispersion scoring, Assess the energy concentration at the pipe wall. For equivalent defect sensitivity scoring, Scoring the project's incentive potential. to These are the weighting coefficients; The scoring weights are adjusted according to the type of inspection task or engineering application scenario. When the inspection task is long-distance screening, the weights of low attenuation score and weak dispersion score are increased. When the inspection task is pitting corrosion or local wall thinning defect identification, the weights of equivalent defect disturbance sensitivity score and pipe wall energy concentration score are increased. Among these, the weights corresponding to complex wave number disturbance index and candidate defect area strain energy concentration index can be increased. When the inspection task is rapid on-site inspection, the weight of engineering excitability score is increased.

[0037] Before scoring, constraints are first screened, and those that do not meet the minimum pipe wall energy percentage are excluded. Maximum permissible attenuation coefficient Constrained modes are directly excluded; For the modes that pass the constraint screening, calculate five sub-scores: (a) Decay score: ; (b) Dispersion score: ; (c) Energy score: , in These are the weights for the proportion of pipe wall energy and the weights for soil leakage inhibition, respectively. (d) Equivalent Defect Disturbance Sensitivity Scoring: For pitting, localized corrosion, or wall thinning defects to be identified, an equivalent micro-wall thickness disturbance is set in the candidate defect region of the pipe wall; the equivalent micro-wall thickness disturbance is used to characterize the local stiffness and mass changes caused by the defect, without the need to establish a complete defect scattering model; the defect-free baseline model and the model after applying the equivalent micro-wall thickness disturbance are solved respectively to obtain the first... Candidate modes at frequency Reference complex wavenumber at the location and the complex wave number after perturbation And calculate the normalized complex wavenumber perturbation index: (4) In the formula, This is the equivalent wall thickness reduction amount. This represents the original wall thickness of the pipe. Calculate the modal strain energy lumped index of the candidate defect region: (5) In the formula, This is a candidate defect region. For the entire pipe wall area, For the first Strain energy density of the first mode in the tube wall; Calculate the percentage of normal displacement on the candidate defect surface: (6) In the formula, The pipe wall surface where the candidate defect is located. This represents the normal displacement of the surface; when the candidate defect is located on the inner or outer surface of the pipe, the normal displacement of the corresponding surface is used respectively. No. The equivalent defect perturbation sensitivity score for the first mode is: (7) In the formula, , and These are the weighting coefficients for the proportions of complex wavenumber disturbance, strain energy concentration, and normal displacement, respectively, and they satisfy the following: (8) The normalization function is shown in formula (9); The larger the value, the more sensitive the corresponding frequency and mode are to the target defect; (9) in, and These represent the current candidate frequency and the minimum and maximum values ​​in the modality set, respectively. A positive number set to prevent the denominator from being zero; (e) Motivational rating: ,in Exciter spatial distribution and the first The normalized coupling efficiency of the mode shape is related to the actual sensor loop, excitation direction, circumferential order, and device bandwidth. The weights of the comprehensive scoring function are adaptively adjusted according to the type of inspection task: a weight vector is generated based on the type of inspection task, the type of target defect, the equipment bandwidth, and the on-site calibration error. And normalize the weights so that ; Meanwhile, for the three-level comparison working conditions, the design working condition difference indicators are as follows: (10) (11) in, , and These represent the group velocities of guided waves under three operating conditions: empty pipe, liquid-filled pipe, and buried liquid-filled pipe. , and These represent the attenuation of guided waves under three operating conditions: empty pipe, liquid-filled pipe, and buried liquid-filled pipe. When the operating condition difference index of a certain frequency band exceeds the threshold, the frequency band is marked as an environmentally sensitive frequency band and its recommended priority is reduced. This can transform the "empty pipe parameters cannot be directly extrapolated" into a clear technical means. Step 6: The detection parameter output module outputs the recommended results, which include the preferred mode, recommended frequency range, center frequency, bandwidth, excitation waveform suggestion, reception time window, theoretical maximum detection distance, high-frequency dispersion areas to be avoided, and on-site calibration suggestions. When the candidate frequency is located near the mode cutoff frequency, in the mode crossover area, in the group velocity change area, in the attenuation peak area, or in the low-efficiency frequency band of the equipment, the candidate frequency is marked as a forbidden frequency band or a reduced-weighted frequency band.

[0038] The theoretical maximum detection distance estimate, in single-ended pulse echo mode, is as follows: (12) The estimated theoretical maximum detection distance for transmission detection is as follows: (13) in and All are expressed in dB. The equivalent amplitude reflection coefficient of the defect. The unit propagation attenuation coefficient is expressed in dB / m.

[0039] After the guided wave detection is completed, the calculation results are corrected by experimental or field baseline signals. The measured group velocity and attenuation coefficient are extracted using end echoes, weld echoes or known structural features (such as pipe joints, flanges, etc.). The attenuation score, dispersion score, and the weighting coefficients of the attenuation score and dispersion score in step 5 are then corrected.

[0040] Theoretical results are corrected using experimental or field baseline signals, specifically including: When the group velocity relative error and the relative error of the attenuation coefficient If all parameters are less than the preset threshold, the final recommended parameters are output; otherwise, the soil impedance, covering layer parameters, or coupling state parameters are updated, and the dispersion solution and modal scoring are re-executed. If the single-point correction method is used, then the end reflection and correction factor are utilized. After correction ; ,in Represents the measured peak amplitude. Represents the theoretical peak amplitude. Represents the corrected attenuation coefficient. Represents the theoretical attenuation coefficient. This represents the distance between the broadcast excitation point and the end of the pipeline.

[0041] If a multi-point correction method is used, a more accurate attenuation coefficient is fitted using the least squares method by utilizing multiple known structural echoes.

[0042] Subsequently, for a new pipe segment to be tested, the data storage and feedback module calculates the normalized working condition distance between it and historical pipe segments based on pipe diameter, wall thickness, material parameters, fluid parameters, burial depth, soil parameters, and coating layer parameters. Based on the working condition distance from smallest to largest, similar historical working conditions are selected from the database, and the recommended frequency range, preferred mode, and correction parameters of similar historical working conditions are called as the initial values ​​for the current dispersion calculation, mode tracking, and establishment of the comprehensive scoring function. The guided wave detection frequency and mode selection will begin from step 3.

[0043] And based on the measured group velocity and measured attenuation coefficient Calculate the correction error: (14) (15) In the formula, and These represent the calculated group velocity and decay coefficient, respectively. when or If the error exceeds the corresponding error threshold, update the soil equivalent impedance, the covering layer damping parameters or the scoring weight, and re-execute the dispersion solution and modal scoring; if both errors are not greater than the corresponding error threshold, write the final optimization results of the current pipe segment (optimized mode, recommended frequency range, center frequency, bandwidth, excitation waveform suggestions, receiving time window, theoretical maximum detection distance, high-frequency dispersion areas to be avoided and field calibration suggestions, etc.) and correction parameters (correction factor and error) into the database for subsequent similar pipe segments to call.

[0044] The above embodiments are merely illustrative of the principles and effects of the present invention, as well as some examples of its application, and are not intended to limit the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept of the present invention, and these modifications and improvements are all within the scope of protection of the present invention.

Claims

1. A method for adaptive optimization of guided wave detection frequency and mode in buried liquid-filled water supply pipelines, comprising the following steps: Step 1: Collect the basic parameters of the water supply pipeline to be tested, including four categories: (a) Pipe geometric parameters: pipe diameter Wall thickness ; (b) Pipe material parameters: elastic modulus Poisson's ratio ,density (c) Fluid parameters inside the pipe: water density Speed ​​of sound Operating pressure (d) External soil and covering layer parameters: soil density Material damping ratio shear modulus Complex shear modulus burial depth Coating thickness and material parameters; enter the above parameters into the parameter input module; Step 2: In the pipe-soil-water coupling modeling module, establish a three-level comparative working condition model based on the above parameters. The three-level comparative working conditions include an empty pipe model, a liquid-filled pipe model, and a buried liquid-filled pipe model. In the buried liquid-filled pipe model, set the water acoustic domain, pipe wall elastic domain, covering layer domain, and external soil equivalent impedance boundary. Step 3: Dispersion Calculation; In the SAFE dispersion solution module, the pipe cross-section is discretized based on the semi-analytical finite element method, and analytical waveforms are used in the axial direction. ,in These represent radial, circumferential, and axial coordinates, respectively; t represents time; and k represents the wavenumber. Representing frequencies, constructing a pipe-soil-water coupled dispersion eigenvalue problem, and solving for different frequencies. The complex wave number below; Step 4: Modal tracking; based on complex wavenumbers Calculate the phase velocity of each guided wave mode. Group speed attenuation coefficient , frequency dispersion broadening Pipe wall energy ratio Water body energy ratio Percentage of energy leaked from soil and modal morphology correlation coefficient; Step 5: Using the modal scoring module, establish a comprehensive scoring function for each candidate mode; the comprehensive scoring function S can be expressed as formula (1): （1） in For low decay score, For weak dispersion scoring, Assess the energy concentration at the pipe wall. For equivalent defect sensitivity scoring, Scoring the project's incentive potential. to These are the weighting coefficients; The scoring weights are adjusted according to the type of inspection task or engineering application scenario. When the inspection task is long-distance screening, the weights of low attenuation score and weak dispersion score are increased. When the inspection task is pitting corrosion or local wall thinning defect identification, the weights of equivalent defect disturbance sensitivity score and pipe wall energy concentration score are increased. Among these, the weights corresponding to complex wave number disturbance index and candidate defect area strain energy concentration index can be increased. When the inspection task is rapid on-site inspection, the weight of engineering excitability score is increased. Step 6: The detection parameter output module outputs the recommended results, which include the preferred mode, recommended frequency range, center frequency, bandwidth, excitation waveform suggestion, reception time window, theoretical maximum detection distance, high-frequency dispersion areas to be avoided, and on-site calibration suggestions. When the candidate frequency is located near the mode cutoff frequency, in the mode crossover area, in the group velocity change area, in the attenuation peak area, or in the low-efficiency frequency band of the equipment, the candidate frequency is marked as a forbidden frequency band or a reduced-weighted frequency band.

2. The adaptive optimization method for guided wave detection frequency and mode of buried liquid-filled water supply pipeline as described in claim 1, characterized in that: It also includes step 7, after the guided wave detection is completed, the calculation results are corrected by experimental or field baseline signals, the measured group velocity and attenuation coefficient are extracted by end echo, weld echo or known structural features, and the attenuation score, dispersion score and the weighting coefficient of the attenuation score and dispersion score in step 5 are corrected.

3. The adaptive optimization method for guided wave detection frequency and mode of a buried liquid-filled water supply pipeline as described in claim 1, characterized in that: The establishment of the three-level comparative working condition model specifically includes: Level 1, Empty Pipe Model: Only considering the elastic domain of the pipe wall Stress in the pipe wall -strain The relationship satisfies the linear elastic constitutive equation ,in Represents the fourth-order elastic tensor; The second level, the liquid-filled tube model: This adds an acoustic domain to the water inside the tube, based on the empty tube design. Sound pressure field of water body Satisfying the Helmholtz equation ,in On the inner surface of the pipe wall The fluid-structure interaction boundary conditions are satisfied at this location. ; Level 3, Buried Liquid-Filled Pipe Model: Adding a Coating Layer Domain Equivalent impedance boundary with external soil Equivalent impedance boundary conditions are applied to the outer surface of the coating layer using a viscoelastic equivalent impedance boundary model. The inner surface of the pipe wall satisfies the fluid-structure interaction boundary condition, the outer surface of the cladding layer satisfies the equivalent impedance boundary condition, and the contact surface between the pipe wall and the cladding layer satisfies the displacement and stress continuity condition.

4. The adaptive optimization method for guided wave detection frequency and mode of buried liquid-filled water supply pipeline as described in claim 1, characterized in that: The SAFE dispersion solving module solves the dispersion eigenvalue problem of the pipe-soil-water coupled system based on the semi-analytical finite element method. The specific process is as follows: For the elastic domain and the cladding domain of the pipe wall, the displacement field at any point on the pipe cross-section is expressed as: ,That Represented by finite element discretization as For the acoustic domain of a water body, the sound pressure field is expressed as: ; Substituting the semi-analytical displacement field and sound pressure field into the governing equations and coupled boundary conditions of each domain, the dispersion eigenvalue problem is obtained through the Galerkin weighted residual method: （2） in , , Here is the stiffness matrix. For the quality matrix, For a given angular frequency, the vector represents the generalized degrees of freedom. Solving for a set of complex wave numbers yields a set of complex wave numbers. ; Calculate the dispersion parameters of each mode based on the complex wavenumber: Phase velocity: ; Group speed: ; Attenuation coefficient: ; Dispersion broadening: = ;; Pipe wall energy percentage: .

5. The adaptive optimization method for guided wave detection frequency and mode of buried liquid-filled water supply pipeline as described in claim 1, characterized in that: The modal tracking module is used to establish continuous tracking relationships between modes in the frequency domain; for adjacent frequency points... and Calculate the modal morphology correlation coefficient (MSCC): （3） in This represents the confidence level between the m-th mode and the n-th mode. and They represent the frequencies respectively. The next m First-order mode vector, Represents frequency Next n The order mode vector, where H represents the conjugate transpose; A mode matching cost function is constructed based on the modal morphology correlation coefficient, phase velocity continuity, group velocity continuity, and tube wall energy proportion continuity, and candidate modes at adjacent frequency points are consecutively numbered.

6. The adaptive optimization method for guided wave detection frequency and mode of buried liquid-filled water supply pipeline as described in claim 1, characterized in that: The specific process for establishing a comprehensive scoring function for each candidate mode is as follows: Before scoring, constraints are first screened, and those that do not meet the minimum pipe wall energy percentage are excluded. Maximum permissible attenuation coefficient The constrained modes are directly excluded; For the modes that pass the constraint screening, calculate five sub-scores: (a) Decay score: ; (b) Dispersion score: ; (c) Energy score: , in These are the weights for the proportion of pipe wall energy and the weights for soil leakage inhibition, respectively. (d) Equivalent Defect Disturbance Sensitivity Scoring: For pitting, localized corrosion, or wall thinning defects to be identified, an equivalent micro-wall thickness disturbance is set in the candidate defect region of the pipe wall; the equivalent micro-wall thickness disturbance is used to characterize the local stiffness and mass changes caused by the defect, without the need to establish a complete defect scattering model; the defect-free baseline model and the model after applying the equivalent micro-wall thickness disturbance are solved respectively to obtain the first... Candidate modes at frequency Reference complex wavenumber at the location and the complex wave number after perturbation And calculate the normalized complex wavenumber perturbation index: （4） In the formula, This is the equivalent wall thickness reduction amount. This represents the original wall thickness of the pipe. Calculate the modal strain energy lumped index of the candidate defect region: （5） In the formula, This is a candidate defect region. For the entire pipe wall area, For the first Strain energy density of the first mode in the tube wall; Calculate the percentage of normal displacement on the candidate defect surface: （6） In the formula, The pipe wall surface where the candidate defect is located. This represents the normal displacement of the surface; when the candidate defect is located on the inner or outer surface of the pipe, the normal displacement of the corresponding surface is used respectively. No. The equivalent defect perturbation sensitivity score for the first mode is: （7） In the formula, , and These are the weighting coefficients for the proportions of complex wavenumber disturbance, strain energy concentration, and normal displacement, respectively, and they satisfy the following: （8） The normalization function is shown in formula (9); The larger the value, the more sensitive the corresponding frequency and mode are to the target defect; （9） in, and These represent the current candidate frequency and the minimum and maximum values ​​in the modality set, respectively. A positive number set to prevent the denominator from being zero; (e) Motivational rating: ,in Exciter spatial distribution and the first The normalized coupling efficiency of the first mode shape is related to the actual sensor loop, excitation direction, circumferential order, and device frequency band. The weights of the comprehensive scoring function are adaptively adjusted according to the type of inspection task: a weight vector is generated based on the type of inspection task, the type of target defect, the equipment bandwidth, and the on-site calibration error. And normalize the weights so that ; Meanwhile, for the three-level comparison working conditions, the design working condition difference indicators are as follows: （10） （11） in, , and These represent the group velocities of guided waves under three operating conditions: empty pipe, liquid-filled pipe, and buried liquid-filled pipe. , and These represent the attenuation of guided waves under three operating conditions: empty pipe, liquid-filled pipe, and buried liquid-filled pipe. When the operating condition difference index of a certain frequency band exceeds the threshold, the frequency band is marked as an environmentally sensitive frequency band, and its recommended priority is reduced.

7. The adaptive optimization method for guided wave detection frequency and mode of a buried liquid-filled water supply pipeline as described in claim 1, characterized in that: The theoretical maximum detection distance estimate, in single-ended pulse echo mode, is as follows: （12） The estimated theoretical maximum detection distance for transmission detection is as follows: （13） in and All are expressed in dB. The equivalent amplitude reflection coefficient of the defect. The unit propagation attenuation coefficient is expressed in dB / m.

8. The adaptive optimization method for frequency and mode selection of guided wave detection in buried liquid-filled water supply pipelines as described in claim 2, characterized in that: In step 7, the theoretical results are corrected using experimental or field baseline signals, specifically including: When the group velocity relative error and the relative error of the attenuation coefficient If all parameters are less than the preset threshold, the final recommended parameters are output; otherwise, the soil impedance, covering layer parameters, or coupling state parameters are updated, and the dispersion solution and modal scoring are re-executed. If a single-point correction method is used, then end reflection and correction factor are utilized. After correction ; ,in Represents the measured peak amplitude. Represents the theoretical peak amplitude. Represents the corrected attenuation coefficient. Represents the theoretical attenuation coefficient. This represents the distance between the broadcast excitation point and the end of the pipeline; If a multi-point correction method is used, a more accurate attenuation coefficient is fitted using the least squares method by utilizing multiple known structural echoes.

9. The adaptive optimization method for frequency and mode selection of guided wave detection in buried liquid-filled water supply pipelines as described in any one of claims 1-8, characterized in that: For a new pipe segment to be tested, the data storage and feedback module calculates the normalized working condition distance between it and historical pipe segments based on pipe diameter, wall thickness, material parameters, fluid parameters, burial depth, soil parameters, and cladding layer parameters. Based on the working condition distance from smallest to largest, similar historical working conditions are selected from the database, and the recommended frequency range, preferred mode, and correction parameters of similar historical working conditions are called as the initial values ​​for the current dispersion calculation, mode tracking, and establishment of the comprehensive scoring function. The guided wave detection frequency and mode selection will begin from step 3.

10. The adaptive optimization method for frequency and mode selection of guided wave detection in buried liquid-filled water supply pipelines as described in claim 9, characterized in that: After completing the on-site testing, based on the measured group speed and measured attenuation coefficient Calculate the correction error: （14） （15） In the formula, and These represent the calculated group velocity and decay coefficient, respectively. when or If the error exceeds the corresponding error threshold, update the soil equivalent impedance, the covering layer damping parameters or the scoring weight, and re-execute the dispersion solution and modal scoring; if neither error exceeds the corresponding error threshold, write the final optimization result and correction parameters of the current pipe segment into the database for subsequent similar pipe segments to use.

Images