A pipeline damage positioning method and system based on ultrasonic guided wave multi-feature fusion

Abstract

The application belongs to the field of pipeline damage positioning, in order to solve the problem of poor positioning accuracy of pipeline damage, a pipeline damage positioning method and system based on ultrasonic guided wave multi-feature fusion are provided. Among them, the pipeline damage positioning method based on ultrasonic guided wave multi-feature fusion includes obtaining ultrasonic guided wave response signals under the conditions of no damage and damage, i.e. obtaining healthy signals and damage signals, selecting effective signal segments of the healthy signals and the damage signals, and screening out effective paths; ellipse path probability distribution functions and ring path probability distribution functions are constructed respectively; according to the product of the ellipse path probability distribution function and the ring path probability distribution function, the damage distribution probability on the pipeline plane is calculated, and then the damage distribution probabilities of the virtual plane and the actual plane are superimposed at the same position point, and finally the pipeline damage distribution is determined, which has the characteristics of high sensitivity, high positioning accuracy and good robustness.

Description

Technical Field

[0001] This invention belongs to the field of pipeline damage localization, and particularly relates to a pipeline damage localization method and system based on ultrasonic guided wave multi-feature fusion. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Pipelines, as crucial tools for fluid transportation, are widely used in the transport of oil, natural gas, and other fluids. However, in practical applications, pipelines may face harsh environments and human-caused damage. After long-term service, pipelines may develop problems such as cracks, perforations, corrosion, and deformation. Damage reduces the pipeline's stiffness and other properties, and in severe cases, can lead to leaks, thus affecting the pipeline's service life. Early detection of pipeline leaks is of great importance and is becoming an important application area for structural health monitoring technology.

[0004] Ultrasonic guided waves excited by piezoelectric ceramics can propagate along cylindrical pipes and are widely used in pipe damage detection due to their long propagation distance, high sensitivity, and full coverage. Ultrasonic guided wave imaging technology can accurately detect pipe sections over short distances. The probabilistic imaging method within ultrasonic guided wave imaging is widely used in locating damage in plate-like structures and also has some application in pipe damage location. However, due to the unique characteristics of pipe sensor arrays, pipe structures lack some paths perpendicular to the sensor array direction compared to plate-like structures, thus limiting the use of traditional probabilistic imaging methods. Furthermore, the lack of circumferential sensing paths in traditional probabilistic imaging methods leads to significant errors in location beyond the center of the monitoring area. Although existing technologies have improved probabilistic imaging methods by introducing time-of-flight parameters, the symmetry of pipe structures means that simply using time-of-flight for location can produce artifacts, thus affecting the accuracy of damage location in the pipe. Summary of the Invention

[0005] To address the technical problems mentioned above, this invention provides a pipeline damage localization method and system based on ultrasonic guided wave multi-feature fusion, which features high sensitivity, high positioning accuracy, and good robustness.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] The first aspect of the present invention provides a method for locating pipeline damage based on the fusion of multiple features of ultrasonic guided waves.

[0008] In one or more embodiments, a pipeline damage localization method based on ultrasonic guided wave multi-feature fusion includes:

[0009] The ultrasonic guided wave response signals under both undamaged and damaged states are obtained, i.e., healthy and damaged signals are obtained.

[0010] Based on the pipeline plane and velocity matching wave packets, effective signal segments of healthy and damaged signals are selected, thereby filtering out effective paths; wherein, the pipeline plane consists of the actual plane of the pipeline extending along the axial direction and half of the virtual plane of the pipeline extending to both sides;

[0011] Based on the correlation between health signals and damage signals along each effective path, an elliptical path probability distribution function is constructed; based on the flight time of the signal along each effective path, a circular path probability distribution function is constructed.

[0012] The damage distribution probability on the pipeline plane is calculated by multiplying the elliptical path probability distribution function and the circular path probability distribution function. Then, the damage distribution probabilities of the virtual plane and the actual plane are superimposed at the same location points to finally determine the pipeline damage distribution.

[0013] As one implementation method, before acquiring the ultrasonic guided wave response signals in both undamaged and damaged states, the method further includes:

[0014] Based on numerical simulation data of waveguide propagation mode and dispersion characteristics in the pipeline structure, the sensor network and excitation frequency are determined.

[0015] As one implementation method, in the process of determining the sensor network and excitation frequency, when the preset time and amplitude conditions of the signal between the two peaks of the two wave packets in the response signal are met, the current corresponding excitation frequency and sensor array distance are obtained.

[0016] As one implementation method, in the process of selecting the effective signal segments of the health signal and the damage signal, the length of the effective signal is determined based on the sensing path length and the group velocity of the wave of the selected mode.

[0017] As one implementation method, the process of filtering valid paths is as follows:

[0018] The corresponding scattering signal is obtained by subtracting the damage signal from the healthy signal on each sensing path.

[0019] The energy damage factor of each sensing path is obtained by comparing the energy of the scattered signal on each sensing path with the energy of the healthy signal.

[0020] Effective paths are selected by comparing the damage factor of each sensing path with a preset loss factor threshold.

[0021] A second aspect of the present invention provides a pipeline damage localization system based on ultrasonic guided wave multi-feature fusion.

[0022] In one or more embodiments, a pipeline damage localization system based on ultrasonic guided wave multi-feature fusion includes:

[0023] The response signal acquisition module is used to acquire ultrasonic guided wave response signals under both undamaged and damaged states, i.e., to obtain healthy and damaged signals.

[0024] The effective path filtering module is used to select effective signal segments of healthy and damaged signals based on the pipe plane and velocity matching wave packets, thereby filtering out effective paths; wherein, the pipe plane packet consists of the actual plane of the pipe extending along the axial direction and half of the virtual pipe plane extending to both sides;

[0025] The distribution function construction module is used to construct an elliptical path probability distribution function based on the correlation between health and damage signals on each effective path; and to construct a circular path probability distribution function based on the flight time of the signal on each effective path.

[0026] The damage distribution determination module is used to calculate the damage distribution probability on the pipeline plane based on the product of the elliptical path probability distribution function and the circular path probability distribution function. Then, the damage distribution probabilities of the virtual plane and the actual plane are superimposed at the same location points to finally determine the pipeline damage distribution.

[0027] As one implementation, the response signal acquisition module further includes, before acquiring the ultrasonic guided wave response signals in both undamaged and damaged states:

[0028] Based on numerical simulation data of waveguide propagation mode and dispersion characteristics in the pipeline structure, the sensor network and excitation frequency are determined.

[0029] In one implementation, in the effective path filtering module, during the selection of effective signal segments for healthy and damaged signals, the length of the effective signal is determined based on the sensing path length and the group velocity of the wave of the selected mode.

[0030] As one implementation method, the process of filtering valid paths in the valid path filtering module is as follows:

[0031] The corresponding scattering signal is obtained by subtracting the damage signal from the healthy signal on each sensing path.

[0032] The damage factor of each sensing path is obtained by comparing the energy of the scattered signal on each sensing path with the energy of the healthy signal.

[0033] Effective paths are selected by comparing the damage factor of each sensing path with a preset loss factor threshold.

[0034] A third aspect of the present invention provides a computer-readable storage medium.

[0035] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described above.

[0036] A fourth aspect of the present invention provides an electronic device.

[0037] A pipeline damage localization device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described above.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] (1) This invention introduces time of flight into the traditional definition of probability distribution, constructs a ring-shaped damage probability distribution characterized by time variation, and combines it with an elliptical probability distribution constructed with correlation coefficient as the feature. The two probability distributions of different shapes are multiplied to construct a new probability distribution. This solves the problem of large edge damage localization error in traditional probability imaging methods and the symmetry problem of the TOF method. At the same time, the scattered energy is introduced as a weight parameter into the probability solution, which can make the results more focused. It effectively achieves the purpose of weakening the influence of artifacts at irrelevant damage locations and suppressing prior knowledge and calculation errors, thereby improving the damage detection accuracy and robustness of pipeline structures and greatly enhancing the engineering application capability of this method.

[0040] (2) Since the positioning algorithm only uses the direct wave packets received by the receiving sensor, this invention expands the pipeline into a plane and extends it to the left and right by half a plane along the pipeline expansion direction. The size of the left and right virtual planes is half that of the actual expanded plane. In this way, all direct paths can be represented in the plane without adding extra paths, thus reducing the amount of computation. At the same time, the use of direct wave packets greatly reduces the difficulty of signal feature extraction and ensures the accuracy of feature extraction.

[0041] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0042] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0043] Figure 1 This is a schematic diagram of the pipeline damage localization process based on ultrasonic guided waves according to an embodiment of the present invention;

[0044] Figure 2(a) is a phase velocity diagram of the pipe under the current material and size parameters of an embodiment of the present invention;

[0045] Figure 2(b) is a velocity diagram of the pipe group under the current material and size parameters of an embodiment of the present invention;

[0046] Figure 2(c) is a phase velocity diagram of the flat plate structure under the same parameters in an embodiment of the present invention;

[0047] Figure 2(d) is a group velocity diagram of the flat plate structure under the same parameters in an embodiment of the present invention;

[0048] Figure 3 This is a half-plane extended view of the pipe according to an embodiment of the present invention;

[0049] Figure 4 This is a component of the ultrasonic guided wave-based pipeline inspection system according to an embodiment of the present invention.

[0050] Figure 5(a) is a spatial distribution diagram of the elliptical probability distribution according to an embodiment of the present invention;

[0051] Figure 5(b) is a spatial distribution diagram of the ring probability distribution according to an embodiment of the present invention;

[0052] Figure 6 This is a flowchart of the pipeline structure damage detection based on ultrasonic guided waves according to an embodiment of the present invention.

[0053] Figure 7(a) is a localization result diagram of the current algorithm in this embodiment of the invention;

[0054] Figure 7(b) shows the imaging results of a single-damage experiment according to an embodiment of the present invention;

[0055] Figure 7(c) shows the imaging results of the multi-damage experiment in an embodiment of the present invention;

[0056] Figure 7(d) shows the imaging results of multi-damage imaging in an embodiment of the present invention after thresholding.

[0057] Figure 8 This is a comparison of direct wave packet signals from different paths according to embodiments of the present invention;

[0058] Figure 9(a) is an image of the extended plane according to an embodiment of the present invention;

[0059] Figure 9(b) is an imaging result diagram after probability superposition according to an embodiment of the present invention;

[0060] Figure 9(c) is an image of the imaging result after thresholding processing according to an embodiment of the present invention;

[0061] Figure 10 These are all the test points in this embodiment of the invention. Detailed Implementation

[0062] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0063] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0064] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0065] Example 1

[0066] according to Figure 1 and Figure 6 This embodiment provides a pipeline damage localization method based on ultrasonic guided wave multi-feature fusion, which includes:

[0067] Step 1: Obtain the ultrasonic guided wave response signals under undamaged and damaged states, i.e., obtain the healthy signal and the damaged signal.

[0068] In the specific implementation process, before acquiring the ultrasonic guided wave response signals under both undamaged and damaged states, the following steps are also included:

[0069] Based on numerical simulation data of waveguide propagation mode and dispersion characteristics in the pipeline structure, the sensor network and excitation frequency are determined.

[0070] Numerical simulations of a flat plate structure and a pipe structure with identical material parameters (elastic modulus, shear modulus, Poisson's ratio, density, etc.) were performed using the open-source Dispersion Calculator and the Pcdisp software package, respectively, and the dispersion curves of the guided waves were plotted. Taking the two structures with a thickness of 2 mm as examples, the dispersion curves are shown in Figures 2(a)-2(d). Comparing the dispersion curves of the two structures, it is clear that when the excitation frequency is greater than 50 kHz, the axial mode in the pipe is close to the A0 mode of the plate structure. At this time, the pipe can be unfolded along the axial direction as a flat plate for analysis.

[0071] Pipeline planar development and expansion, as follows Figure 3 As shown. The original plane at the middle position is denoted as Ω0, and the half-plane extending to the left is denoted as Ω. -1 Let the half-plane extending to the right be denoted as Ω1. Then the x-coordinate of the virtual sensor in the left half-plane is... The horizontal coordinate of the virtual sensor in the right half-plane is The expression for the path length between the excitation sensor and the receiving sensor is:

[0072]

[0073] In the formula, It is the first j The x-coordinate of each receiving sensor, Let be the x-coordinate of the i-th excitation sensor, R be the outer diameter of the pipe, L be the horizontal distance between the excitation sensor and the receiving sensor, and r be the order of the extended plane.

[0074] Since the analysis focuses on the direct signal wave packet, and the guided wave propagation in the pipe is helical and directional, calculations combining path length and guided wave propagation speed are necessary to facilitate the separation of the shortest-distance direct wave packet from the opposite-direction wave packet in the signal. Specifically, during the determination of the sensor network and excitation frequency, given the preset time and amplitude conditions between the two peaks of the two wave packets in the response signal, the corresponding excitation frequency and sensor array distance are calculated. Specifically, the distance difference between the guided wave arriving from the left and from the right along the same sensing path is calculated, and then, combined with the wave velocity, the excitation frequency and sensor array distance corresponding to the condition that the two wave packets in the response signal can be clearly separated are obtained.

[0075] When the pipe thickness is small, ultrasonic guided waves similar to those on a planar structure can be excited. When the transducer excites the guided waves, it generates a certain diffusion angle. Therefore, the propagation process of the guided waves in the pipe starts from the excitation source and spreads outwards. Thus, ultrasonic guided waves excited from one location will have a time difference when they reach the receiving sensor due to their different propagation directions. Waves propagating to the left are called left-handed spiral waves, and those propagating to the right are called right-handed spiral waves. This paper calculates characteristic parameters based on the direct wave packets, which requires the receiving sensor to separate the left-handed and right-handed direct wave packets. Simultaneously, by calculating the arrival times of the left-handed and right-handed direct signals, and centering on the maximum value of the envelope, the time difference between the two peak values ​​is calculated to see if it is greater than the total duration of the excitation signal. Furthermore, the envelope of the signal between the two peaks is compared to see if it is less than 10% of the maximum peak value. If the time and amplitude conditions are satisfied, the two wave packets are considered to be clearly separated.

[0076] like Figure 8 As shown, taking A1 excitation as an example, the distances the wave travels from the left and right to R2 are 506.4mm and 750.7mm respectively. The distance difference is large, so the wave packets can be clearly separated. However, the distances to R4 are 554.6mm and 640.3mm respectively, which will cause aliasing of the direct wave packets. Therefore, it is necessary to select an appropriate frequency as the excitation frequency.

[0077] In this embodiment, the final selected distance between the excitation sensor array and the receiving sensor array is 500mm, and the excitation frequency is 150kHz.

[0078] By preparing test specimens, attaching sensors, building a pipeline ultrasonic testing system, designing a sensor network, and acquiring ultrasonic guided wave response signals under both non-destructive and damaged conditions, a pipeline ultrasonic testing system can be constructed.

[0079] The constructed ultrasonic guided wave testing system, such as Figure 4 As shown. The acoustic guided wave detection system consists of the pipe under test, a piezoelectric transducer, a signal generator, a power amplifier, and a data acquisition card. The pipe under test is 2000 mm long, 2 mm thick, and has an outer diameter of 204 mm. The 16 sensors are divided into two groups: one group serves as the excitation sensors, named A1-A8, and the other group serves as the receiving sensors, named S1-S8. The area between the two groups of sensors is the test area.

[0080] The ultrasonic guided wave response signal under healthy conditions was acquired. The excitation signal was a 5-cycle sinusoidal modulation signal. Since the ultrasonic guided wave system uses a polling excitation method, the excitation Lamb signal was polled to 16 sensors on the non-destructive plate, and the response signal was received to construct a healthy signal matrix. The size is 128×8000.

[0081] Ultrasonic guided wave response signals under damaged conditions were acquired. Damage at different locations was simulated using cylindrical damped soil of fixed size. The ultrasonic guided wave response signals were collected and excited in a polling manner using the same excitation method, and a healthy signal matrix was constructed. The size is 128×8000.

[0082] The acquired signals are preprocessed to select the effective signal segments. Since the detection system uses a polling excitation method, some unnecessary path signals, such as A1-A2, are generated. Therefore, the acquired signals first need to be classified to remove AA and RR signals. The size of the effective signal matrix is ​​64×8000. The acquired signal data and sensor coordinates are used as inputs to the algorithm for subsequent calculations.

[0083] Step 2: Based on the pipe plane and velocity matching wave packets, select the effective signal segments of the healthy and damaged signals, and then filter out the effective paths; whereby, the pipe plane packet consists of the actual plane of the pipe extending along the axial direction and half of the virtual plane of the pipe extending to both sides, such as Figure 3 As shown.

[0084] In the specific implementation process, when selecting the effective signal segments for health and damage signals, the length of the sensing path is considered. D The group velocity of the wave in the selected mode determines the length of the effective signal.

[0085]

[0086]

[0087]

[0088] t m Indicates the peak position of the received wave packet. t 0 represents the peak position of the crosstalk signal. v g This represents the group velocity of the Lamb wave in this mode at the current frequency. k The number of peaks in the excitation signal is represented by λ, where λ is the wavelength of a single sinusoidal signal. M This represents the margin of the signal length. Different paths correspond to [...]. t a , t b ] are different. Among them, [ t a , t b ] represents the time of the valid signal segment, where t a Indicates the start time of the valid signal segment. t b Indicates the end time of the valid signal segment.

[0089] The process of filtering valid paths is as follows:

[0090] The corresponding scattering signal is obtained by subtracting the damage signal from the healthy signal on each sensing path.

[0091] The damage factor of each sensing path is obtained by comparing the energy of the scattered signal on each sensing path with the energy of the healthy signal.

[0092] Effective paths are selected by comparing the damage factor of each sensing path with a preset loss factor threshold.

[0093] During the screening of effective paths, the energy of the scattered signal on each sensing path is compared with the energy of the healthy signal to obtain the energy damage factor E of each sensing path. DI The energy loss factor threshold is set to E. TH If E DI Greater than E TH If the path is considered to have strong damage scattering and is significantly affected by damage, it will be retained in subsequent calculations. The energy of the scattered signal from the nth path can be obtained using the same method; the energy of the health signal can be obtained from this energy. .

[0094]

[0095]

[0096]

[0097] Among them, X n (t) represents the acquired signal. When calculating energy, healthy signals, damaged signals, and scattered signals can all be used as X. n Substitute (t) into the formula .

[0098] Step 3: Construct an elliptical path probability distribution function based on the correlation between health and damage signals on each effective path; construct a circular path probability distribution function based on the flight time of the signal on each effective path.

[0099] In this embodiment, the approximate location of the damage is determined by an elliptical probability distribution, and then the precise location is optimized by a circular probability distribution to remove artifacts.

[0100] The process of constructing the probability distribution function of the elliptical path is as follows:

[0101] When the sensor array is initially installed on the pipeline, its structure is assumed to be intact. The signal detected at this time is the health signal, which also serves as the reference signal in subsequent calculations. The damage signal is a new signal obtained after re-inspection of the structure. By comparing the two sets of signals, it is determined whether the amplitude, phase, and other characteristics of the signals have changed, thereby determining whether damage has occurred in the structure.

[0102] The damage factor C is obtained based on the correlation between health signals and damage signals. DI As shown in the formula below:

[0103]

[0104] in, This represents the average of health signals. This is the mean of the scattered signal.

[0105] C DI It is a number between 0 and 1. The larger the value, the closer the path is to the location of the damage, and the greater the impact of the damage; the smaller the value, the less impact the path is on the damage.

[0106] In a sensor network with N sensing paths, what is the probability of damage at location (x, y) in the monitored area? P C (x,y) can be represented as:

[0107]

[0108] In the formula Let Rn(x, y) represent the damage probability of the damage factor of the nth path at (x, y), and let Rn(x, y) represent the spatial distribution function.

[0109] The monitoring area for each path is set as an ellipse, with the excitation sensor and the receiving sensor serving as the foci of the ellipse. The size of the area is controlled by the parameter β, as shown in Figure 5(a). The closer the location is to the path, the higher the probability; the probability of being outside the area is 0. Rn(x, y) can be expressed as:

[0110]

[0111]

[0112] In the formula, β is a constant used to adjust the size of the ellipse. For point The relative distance between the executor and the receiver on the nth path. For point To the executor of the nth path (coordinates are...) The distance; For point To the nth path receiver (coordinates are) The distance; Let be the distance from the actuator to the sensor along the nth path.

[0113] Specifically, the process of constructing the probability distribution function of the circular path is as follows:

[0114] For the i Find a path to obtain any reference point in the grid. The damage probability distribution at the location is as follows:

[0115]

[0116] in, Let be the probability distribution function. The weighting factor is used to amplify the degree of damage to the damaged path, and the energy value of the peak value of the scattered signal from the i-th path is selected. As a weighted signal, it is used to distinguish the magnitude of damage to different paths.

[0117] As shown in Figure 5(b), the flight time is incorporated into the damage probability distribution function, and the expression is:

[0118]

[0119] In the formula, This represents the excitation and reception of the sensor from point (x, y) to the nth path, along with the time and reception. t tof-nThis represents the time for the nth path. This represents the defined time-of-flight error.

[0120]

[0121]

[0122] In the formula, This represents the starting position of the valid signal segment of the nth path. The peak position of the effective scattered signal. This represents the peak value of the excitation signal.

[0123] Step 4: Calculate the damage distribution probability on the pipe plane based on the product of the elliptical path probability distribution function and the circular path probability distribution function. Then, superimpose the damage distribution probabilities of the virtual plane and the actual plane at the same location points to finally determine the pipe damage distribution. Figure 7(a) is the localization result of the current algorithm in this embodiment; Figure 7(b) is the imaging result of the single-damage experiment in this embodiment; Figure 7(c) is the imaging result of the multi-damage experiment in this embodiment; Figure 7(d) is the imaging result of the multi-damage imaging in this embodiment after thresholding.

[0124] By calculating the probability distributions of two different shapes, the two probabilities of damage existing on the plane can be obtained. P c and P a The probability that damage exists at point (x, y). P yes P c and P a The product of the two probabilities. Since the path matching process extends half a plane to the left and right of the tube's unfolded plane, it is necessary to superimpose the probabilities of points at the same positions in the virtual plane and the actual plane. The superimposed probability is denoted as... I (x, y), calculated using the formula shown below, produces imaging results with a large highlighted area, requiring further 90% thresholding. The main purpose of 90% thresholding is to filter out artifacts in the imaging results, making the location of the damage clearer. Figure 9(c) shows a comparison of the damage imaging effects before and after thresholding.

[0125]

[0126]

[0127] in, This represents the probability after the regions are superimposed. This represents the probability that a point on the 0th-order plane is damaged. This represents the probability that damage exists at the midpoint of the right half of the extended plane. This represents the probability that a point in the left half of the extended plane is damaged. The coordinates of the virtual point corresponding to the point (x, y) in the right half of the plane are (x+πR, y), and the coordinates of the virtual point corresponding to the point (x-πR, y) in the left half of the plane are (x-πR, y).

[0128] Repeat the above steps, conduct multiple experiments, calculate the damage error index, and evaluate the algorithm performance.

[0129] Figure 9(a) shows the image of the extended plane, Figure 9(b) shows the image after probability superposition, and Figure 9(c) shows the image after thresholding.

[0130] The peak points corresponding to the probability matrix x and y The coordinates are used as the x and y coordinates of the damage location, denoted as . D ( x , y ).

[0131] Set the damage location S ( x , y )and D ( x , y The Euclidean distance between them is used as the positioning error. e .right N Mean error of damage in group experiments e MAE Calculations are performed to determine the accuracy of the algorithm, while also considering damage errors. e STD The mean squared error was calculated to determine the stability of the algorithm.

[0132]

[0133]

[0134] Due to the symmetry of the pipeline structure, test points were selected within a rectangular area defined by vertices A4, A5, R4, and R5. The distance between A4 and A5 is 80mm, and this length was divided into four equal parts, with a point taken every 20mm. The distance between A4 and R4 is 500mm, and this length was divided into ten equal parts, with a point taken every 50mm, resulting in 45 sampling points within this area. Nine points along the axial direction were grouped together, numbered 1 to 9 from A4 to R4, and each group was named a row. Along the direction from A4 to A5, each group was named row 1 to row 5. All test points are as follows: Figure 10 As shown.

[0135] Statistical analysis of the detection errors at 45 sets of points showed that the average positioning error of the proposed method was 7.01 mm with a standard deviation of 3.72 mm, while the positioning error of the RAPID method was 25.4 mm with a standard deviation of 24.5 mm. The comparison reveals that the original RAPID method exhibits significant positioning errors near the sensor location and can only achieve relatively accurate positioning in the center of the area. This is because the sensing path only exists in the axial direction and cannot effectively represent the circumferential direction. Furthermore, locations with numerous path intersections are easier to detect, while positioning errors are larger in areas with sparse paths. The proposed algorithm, by incorporating time-of-flight, can effectively achieve positioning across the entire monitoring area.

[0136] Example 2

[0137] This embodiment provides a pipeline damage localization system based on ultrasonic guided wave multi-feature fusion, which includes:

[0138] (1) Response signal acquisition module, which is used to acquire ultrasonic guided wave response signals under non-destructive and damaged states, i.e., to obtain healthy signals and damaged signals.

[0139] The response signal acquisition module further includes, before acquiring the ultrasonic guided wave response signals under both undamaged and damaged states, the following:

[0140] Based on numerical simulation data of waveguide propagation mode and dispersion characteristics in the pipeline structure, the sensor network and excitation frequency are determined.

[0141] (2) Effective path filtering module, which is used to select effective signal segments of healthy signals and damaged signals based on the pipeline plane and velocity matching wave packet, and then filter out effective paths; wherein, the pipeline plane packet consists of the actual plane of the pipeline along the axial direction and half of the virtual plane of the pipeline extending to both sides.

[0142] In the effective path filtering module, during the selection of effective signal segments for healthy and damaged signals, the length of the effective signal is determined based on the sensing path length and the group velocity of the wave of the selected mode.

[0143] In the effective path filtering module, the process of filtering effective paths is as follows:

[0144] The corresponding scattering signal is obtained by subtracting the damage signal from the healthy signal on each sensing path.

[0145] The damage factor of each sensing path is obtained by comparing the energy of the scattered signal on each sensing path with the energy of the healthy signal.

[0146] Effective paths are selected by comparing the damage factor of each sensing path with a preset loss factor threshold.

[0147] (3) Distribution function construction module, which is used to construct elliptical path probability distribution function based on the correlation between health signal and damage signal on each effective path; and to construct ring path probability distribution function based on the flight time of the signal on each effective path.

[0148] (4) Damage distribution determination module, which is used to calculate the damage distribution probability on the pipeline plane based on the product of the elliptical path probability distribution function and the ring path probability distribution function, and then superimpose the damage distribution probabilities of the virtual plane and the actual plane at the same location point to finally determine the pipeline damage distribution.

[0149] It should be noted that each module in this embodiment corresponds one-to-one with each step in Embodiment 1, and their specific implementation processes are the same, so they will not be repeated here.

[0150] Example 3

[0151] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described above.

[0152] Example 4

[0153] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described above.

[0154] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0155] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for locating pipeline damage based on multi-feature fusion of ultrasonic guided waves, characterized in that, include: The ultrasonic guided wave response signals under both undamaged and damaged states are obtained, i.e., healthy and damaged signals are obtained. Based on the pipeline plane and velocity matching wave packet, the effective signal segments of the health signal and the damage signal are selected, and then the effective path is screened out; wherein, the pipeline plane packet consists of the actual plane of the pipeline along the axial direction and half of the virtual plane of the pipeline extending to both sides; Based on the correlation between health signals and damage signals of each effective path, an elliptical path probability distribution function is constructed. The process of constructing the probability distribution function of the elliptical path is as follows: When the sensor array is initially installed on the pipeline, its structure is considered to be intact. The signal detected at this time is the health signal, which also serves as the reference signal in subsequent calculations. The damage signal is a new signal obtained by re-detecting the structure. By comparing the two sets of signals, it is determined whether the amplitude and phase characteristics of the signal have changed, thereby determining whether the structure has been damaged. The damage factor C is obtained based on the correlation between health signals and damage signals. DI As shown in the formula below: in, This represents the average of health signals. To obtain the mean of the scattered signal, the damaged signal along each sensing path is subtracted from the healthy signal to obtain the corresponding scattered signal; C DI It is a number between 0 and 1; In a sensor network with N sensing paths, what is the probability of damage at location (x, y) in the monitored area? P C (x,y) can be represented as: In the formula Let Rn(x, y) represent the damage probability of the damage factor of the nth path at (x, y), and let Rn(x, y) represent the spatial distribution function. The monitoring area for each path is set as an ellipse, with the excitation sensor and the receiving sensor serving as the foci of the ellipse. The size of the area is controlled by the parameter β, with higher probability for positions closer to the path and zero probability for positions outside the area; Rn(x, y) can be represented as: In the formula, β is a constant used to adjust the size of the ellipse; For point The relative distance between the executor and the receiver on the nth path; For point The coordinates of the executor on the nth path are The distance; For point The coordinates of the receiver on the nth path are The distance; Let be the distance from the actuator to the sensor along the nth path; Based on the flight time of the signal on each effective path, construct the probability distribution function of the circular path; The process of constructing the probability distribution function of the circular path is as follows: For the th i Find a path to obtain any reference point in the grid. The damage probability distribution at the location is as follows: in, It is the probability distribution function; The weighting factor is used to amplify the degree of damage to the damaged path, and the energy value of the peak value of the scattered signal from the i-th path is selected. As a weighted signal, it is used to distinguish the magnitude of damage to different paths; Incorporating flight time into the damage probability distribution function, the expression is: In the formula, This represents the excitation and reception of the sensor from point (x, y) to the nth path, and the time and reception. t tof-n This represents the time for the nth path. This represents the defined time-of-flight error; In the formula, This represents the starting position of the valid signal segment of the nth path. The peak position of the effective scattered signal. Indicates the peak value of the excitation signal; The damage distribution probability on the pipeline plane is calculated by multiplying the elliptical path probability distribution function and the circular path probability distribution function. Then, the damage distribution probabilities of the virtual plane and the actual plane are superimposed at the same location points to finally determine the pipeline damage distribution.

2. The pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described in claim 1, characterized in that, Before acquiring the ultrasonic guided wave response signals under both undamaged and damaged states, the following steps are also included: Based on numerical simulation data of waveguide propagation mode and dispersion characteristics in the pipeline structure, the sensor network and excitation frequency are determined.

3. The pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described in claim 2, characterized in that, In determining the sensor network and excitation frequency, when the preset time and amplitude conditions of the signal between the two peaks of the two wave packets in the response signal are met, the current corresponding excitation frequency and sensor array distance are calculated.

4. The pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described in claim 1, characterized in that, In the process of selecting the effective signal segment for health and damage signals, the length of the effective signal is determined based on the sensing path length and the group velocity of the wave of the selected mode.

5. The pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described in claim 1, characterized in that, The process of filtering valid paths is as follows: The corresponding scattering signal is obtained by subtracting the damage signal from the healthy signal on each sensing path. The damage factor of each sensing path is obtained by comparing the energy of the scattered signal on each sensing path with the energy of the healthy signal. Effective paths are selected by comparing the damage factor of each sensing path with the preset loss factor threshold.

6. A pipeline damage localization system based on ultrasonic guided wave multi-feature fusion, characterized in that, include: The response signal acquisition module is used to acquire ultrasonic guided wave response signals under both undamaged and damaged states, i.e., to obtain healthy and damaged signals. The effective path filtering module is used to select effective signal segments of healthy and damaged signals based on the pipe plane and velocity matching wave packets, thereby filtering out effective paths; wherein, the pipe plane packet consists of the actual plane of the pipe extending along the axial direction and half of the virtual pipe plane extending to both sides; The distribution function construction module is used to construct an elliptical path probability distribution function based on the correlation between health signals and damage signals of each effective path. The process of constructing the elliptical path probability distribution function is as follows: when the sensor array is initially installed on the pipeline, its structure is considered to be intact, and the signal detected at this time is the health signal, which also serves as the reference signal in subsequent calculations. The damage signal is a new signal obtained by re-detecting the structure. By comparing the two sets of signals, it is determined whether the amplitude and phase characteristics of the signals have changed, thereby determining whether damage has occurred in the structure. The damage factor C is obtained based on the correlation between health signals and damage signals. DI As shown in the formula below: in, This represents the average of health signals. To obtain the mean of the scattered signal, the damaged signal along each sensing path is subtracted from the healthy signal to obtain the corresponding scattered signal; C DI It is a number between 0 and 1; In a sensor network with N sensing paths, what is the probability of damage at location (x, y) in the monitored area? P C (x,y) can be represented as: In the formula Let Rn(x, y) represent the damage probability of the damage factor of the nth path at (x, y), and let Rn(x, y) represent the spatial distribution function. The monitoring area for each path is set as an ellipse, with the excitation sensor and the receiving sensor serving as the foci of the ellipse. The size of the area is controlled by the parameter β, with higher probability for positions closer to the path and zero probability for positions outside the area; Rn(x, y) can be represented as: In the formula, β is a constant used to adjust the size of the ellipse; For point The relative distance between the executor and the receiver on the nth path; For point The coordinates of the executor on the nth path are The distance; For point The coordinates of the receiver on the nth path are The distance; Let be the distance from the actuator to the sensor along the nth path; Based on the flight time of the signal on each effective path, a circular path probability distribution function is constructed; the process of constructing the circular path probability distribution function is as follows: for the i Find a path to obtain any reference point in the grid. The damage probability distribution at the location is as follows: in, It is the probability distribution function; The weighting factor is used to amplify the degree of damage to the damaged path, and the energy value of the peak value of the scattered signal from the i-th path is selected. As a weighted signal, it is used to distinguish the magnitude of damage to different paths; Incorporating flight time into the damage probability distribution function, the expression is: In the formula, This represents the excitation and reception of the sensor from point (x, y) to the nth path, and the time and reception. t tof-n This represents the time for the nth path. This represents the defined time-of-flight error; In the formula, This represents the starting position of the valid signal segment of the nth path. The peak position of the effective scattered signal. Indicates the peak value of the excitation signal; The damage distribution determination module is used to calculate the damage distribution probability on the pipeline plane based on the product of the elliptical path probability distribution function and the circular path probability distribution function. Then, the damage distribution probabilities of the virtual plane and the actual plane are superimposed at the same location points to finally determine the pipeline damage distribution.

7. The pipeline damage localization system based on ultrasonic guided wave multi-feature fusion as described in claim 6, characterized in that, The response signal acquisition module further includes, before acquiring the ultrasonic guided wave response signals under both undamaged and damaged states: Based on numerical simulation data of waveguide propagation mode and dispersion characteristics in the pipeline structure, the sensor network and excitation frequency are determined.

8. The pipeline damage localization system based on ultrasonic guided wave multi-feature fusion as described in claim 6, characterized in that, In the effective path filtering module, during the selection of effective signal segments for healthy and damaged signals, the length of the effective signal is determined based on the sensing path length and the group velocity of the wave of the selected mode. or In the effective path filtering module, the process of filtering effective paths is as follows: The corresponding scattering signal is obtained by subtracting the damage signal from the healthy signal on each sensing path. The damage factor of each sensing path is obtained by comparing the energy of the scattered signal on each sensing path with the energy of the healthy signal. Effective paths are selected by comparing the damage factor of each sensing path with the preset loss factor threshold.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described in any one of claims 1-5.

10. A pipeline damage location device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the pipeline damage localization method based on ultrasonic guided wave multi-feature fusion as described in any one of claims 1-5.

Images