A method, apparatus, medium and product for circular pipe guided wave damage imaging

Abstract

A kind of round tube waveguide damage imaging method, device, medium and product, it is related to data processing field.In the method, the healthy baseline signal matrix of round tube structure under healthy state and the current state signal matrix under the state to be measured are collected;The current state signal matrix and the healthy baseline signal matrix are subtracted point by point in time domain, to obtain difference signal matrix, and the Hilbert transform of difference signal matrix is carried out and absolute value is taken, to obtain only containing positive value amplitude envelope matrix;Three-dimensional cylindrical surface of round tube structure is virtually unfolded into two-dimensional plane, and three-dimensional cylindrical surface is discretized into two-dimensional imaging grid containing multiple pixel points;Each pixel point in two-dimensional imaging grid is traversed, and the intensity value of each pixel point is calculated based on amplitude envelope matrix;Based on the intensity value of each pixel point, generate two-dimensional heat map representing the damage position and morphology of round tube structure.Implementation of the technical solution provided in the present application improves the efficiency of round tube damage detection.

Description

Technical Field

[0001] This application relates to the field of data processing, specifically to a method, apparatus, medium, and product for imaging damage in a circular tube guided wave. Background Technology

[0002] With the rapid development of industrial production and the continuous advancement of urbanization, pipelines, as the core arteries for transporting critical media such as oil, gas, and water, are crucial to economic and social life due to their safe and stable operation. During long-term operation, pipeline structures are inevitably subjected to the combined effects of complex factors such as media erosion, environmental corrosion, and alternating stress, making the pipe walls prone to early structural damage such as microcracks and localized thinning. If these safety hazards are not detected and assessed in a timely manner, they may escalate into catastrophic accidents such as leaks or even explosions. Therefore, efficient and reliable health monitoring of pipeline structures has crucial engineering value and social significance.

[0003] Currently, traditional non-destructive testing methods (such as conventional ultrasonic thickness measurement and radiographic testing) are mostly point-by-point scanning local inspections. These methods have low inspection efficiency and high labor costs. Furthermore, for pipelines with insulation layers or buried pipelines, a large amount of resources are often required to peel off the outer coating, making it difficult to meet the needs of rapid screening and real-time online monitoring of long-distance pipelines, resulting in low inspection efficiency. Summary of the Invention

[0004] This application provides a method, apparatus, medium, and product for imaging damage in circular tube guided waves, which improves the efficiency of detecting damage in circular tubes.

[0005] A first aspect of this application provides a method for imaging damage in a circular tube guided by a waveguide. The method includes: acquiring a healthy baseline signal matrix of the circular tube structure in a healthy state and a current state signal matrix in a test state using a sensor array deployed on the circular tube structure to be monitored; performing a point-by-point time-domain subtraction between the current state signal matrix and the healthy baseline signal matrix to obtain a difference signal matrix; performing a Hilbert transform on the difference signal matrix and taking its absolute value to obtain an amplitude envelope matrix containing only positive values; virtually unfolding the three-dimensional cylindrical surface of the circular tube structure into a two-dimensional plane, and discretizing the three-dimensional cylindrical surface into a two-dimensional imaging grid containing multiple pixels, where each pixel corresponds to a physical location on the circular tube structure; traversing each pixel in the two-dimensional imaging grid and calculating the intensity value of each pixel based on the amplitude envelope matrix; and generating a two-dimensional thermogram characterizing the location and morphology of damage in the circular tube structure based on the intensity value of each pixel.

[0006] By employing the above technical solution, a sensor array was deployed to collect the health baseline signal matrix and the current state signal matrix, achieving complete acquisition of monitoring data for the circular tube structure throughout the entire time period, providing a reliable raw data foundation for subsequent damage detection. A differential signal matrix was obtained by point-by-point subtraction in the time domain, effectively filtering out the constant reflection signal generated by the inherent geometry of the circular tube, making the weak damage scattering signal stand out from the strong background noise. A Hilbert transform was performed on the differential signal matrix, and the absolute value was taken to obtain the amplitude envelope matrix, converting the high-frequency oscillating radio frequency signal into a smooth energy profile, eliminating the influence of signal phase on subsequent processing, and ensuring that signals from different propagation paths could be superimposed purely on energy. The three-dimensional cylinder was virtually unfolded into a two-dimensional plane and discretized into a two-dimensional imaging grid, realizing the mapping from complex three-dimensional space to an intuitive two-dimensional plane. Each pixel precisely corresponds to a physical position on the circular tube, laying the foundation for spatial positioning. The intensity value of each pixel was calculated iteratively, and a delay superposition algorithm was used to achieve virtual focusing at the damage location, enhancing the energy at the actual damage location. The resulting two-dimensional thermal map visually displays the precise distribution of damage along the axial and circumferential directions of the circular tube, achieving a visual representation of the damage location and an accurate presentation of its morphological characteristics. This application acquires a healthy baseline signal matrix and a current state signal matrix based on a sensor array deployed on the circular tube structure. Through differential processing and Hilbert transform, an amplitude envelope matrix is ​​obtained. The three-dimensional cylindrical surface is then virtually unfolded into a two-dimensional plane, generating a two-dimensional imaging grid containing multiple pixels. The intensity value of each pixel is then calculated based on the amplitude envelope matrix, thereby generating a two-dimensional thermal map characterizing the location and morphology of damage in the circular tube structure. This guided wave damage imaging method for circular tubes overcomes the limitations of traditional detection techniques in complex installation environments, achieving efficient localization and intuitive visualization of damage across the entire range of the circular tube. This improves detection efficiency and reliability, while also offering the advantages of non-contact and real-time monitoring, significantly enhancing the health monitoring capabilities of circular tube structures.

[0007] Optionally, the step of calculating the intensity value of each pixel based on the amplitude envelope matrix specifically includes: calculating the total flight time of the guided wave propagating from the excitation source to the pixel and being scattered by the pixel to each sensor in the sensor array; calculating the discrete-time index of each sensor based on the total flight time and the signal sampling time step; extracting and accumulating the amplitude of each sensor at the corresponding index time from the amplitude envelope matrix based on the discrete-time index, and using the sum as the intensity value of the pixel.

[0008] By employing the above technical solution, the total flight time of the guided wave from the excitation source to the pixel and then to each sensor is calculated, establishing a precise correspondence between the sound wave propagation path and time, and realizing a signal spatiotemporal mapping based on the laws of physical propagation. A discrete-time index is calculated based on the total flight time and sampling time step, completing the precise conversion from continuous physical time to discrete digital signal index, ensuring accurate extraction of the signal amplitude at the corresponding moment from the amplitude envelope matrix. By extracting and accumulating the amplitudes of each sensor at the corresponding index moment as the pixel intensity value, coherent superposition of multi-channel signals is achieved. When the pixel location is exactly at the true damage location, the scattered signals received by each sensor are synchronously aligned in time, and the accumulated amplitudes form an intensity peak; while pixels at non-damage locations, due to time mismatch, only exhibit random noise levels after accumulation. This focusing mechanism based on delay summation significantly improves the signal-to-noise ratio of damage detection, enhances the energy concentration at the true damage location, and suppresses spurious responses in non-damage areas, ensuring the accuracy and reliability of the imaging results.

[0009] Optionally, calculating the total flight time of the guided wave from the excitation source to the pixel and from the pixel to each sensor in the sensor array specifically includes: converting the position of the pixel in the two-dimensional imaging grid into spatial coordinates in a three-dimensional Cartesian coordinate system through a coordinate mapping relationship. The spatial coordinates include an abscissa and a ordinate, wherein the abscissa is the product of the inner diameter of the tube and the cosine of the circumferential angle of the pixel, and the ordinate is the product of the inner diameter of the tube and the sine of the circumferential angle of the pixel; calculating the first segment of flight time of the guided wave from the excitation source position along the axial direction to the pixel, the first segment of flight time being equal to the ratio of the axial coordinate of the pixel to the velocity of the guided wave group; calculating the second segment of flight time of the guided wave after being scattered from the pixel to the sensor, the second segment of flight time being equal to the ratio of the three-dimensional Euclidean distance between the pixel and the sensor to the velocity of the guided wave group; and adding the first segment of flight time and the second segment of flight time to obtain the total flight time.

[0010] By employing the above technical solution, the coordinate mapping relationship transforms two-dimensional grid pixels into three-dimensional Cartesian coordinates. The horizontal and vertical coordinates are determined by trigonometric function values ​​of the inner diameter of the circular pipe and the circumferential angle, respectively, achieving a precise conversion from cylindrical coordinates to rectangular coordinates and ensuring the accuracy of spatial position calculations. When calculating the first flight time, the ratio of the pixel's axial coordinate to the waveguide group velocity yields the propagation time of the excitation wave along the pipe's axis, accurately reflecting the axial propagation characteristics of the guided wave within the pipe wall. When calculating the second flight time, the spatial distance from the pixel to the sensor is precisely calculated using the three-dimensional Euclidean distance formula, and then divided by the waveguide group velocity to obtain the propagation time of the scattered wave, fully considering the actual propagation path of the scattered wave in three-dimensional space. Adding the two flight times yields the total flight time, completely describing the entire process of the guided wave from excitation, propagation, scattering to reception, providing a precise time reference for accurately extracting the amplitude at the corresponding moment from the time-domain signal. This segmented calculation method not only has clear physical meaning but also high computational accuracy, effectively avoiding positioning errors caused by simplified models.

[0011] Optionally, discretizing the three-dimensional cylindrical surface into a two-dimensional imaging mesh containing multiple pixels specifically includes: establishing a two-dimensional rectangular coordinate system with the axial direction of the circular tube structure as the horizontal axis and the circumferential angle of the circular tube structure as the vertical axis, wherein the value range of the horizontal axis is from the start position to the end position of the tube, and the value range of the vertical axis is from -π to π; discretizing the axial distance into a preset number of first nodes and the circumferential angle into a preset number of second nodes to construct an initial imaging mesh, and locally refining the initial imaging mesh to obtain the two-dimensional imaging network.

[0012] By adopting the above technical solution, a two-dimensional rectangular coordinate system is established with the axial direction of the circular pipe as the abscissa and the circumferential angle as the ordinate. This enables the intuitive unfolding of the three-dimensional cylindrical structure into a two-dimensional plane, transforming the complex spatial damage localization problem into a planar imaging problem. The abscissa range covers the pipe from its starting position to its ending position, and the ordinate range covers the entire circumference from -π to π, ensuring full coverage imaging of the entire circular pipe surface without missing any potential damage areas. The axial distance is discretized into a preset number of first nodes, and the circumferential angle is discretized into a preset number of second nodes. By reasonably setting the node density, a balance between imaging resolution and computational efficiency is achieved. After constructing the initial imaging mesh, local refinement is performed to obtain the final two-dimensional imaging network. This adaptive mesh strategy ensures overall coverage while refining key areas, improving the imaging accuracy of damaged areas and avoiding the computational burden caused by a globally high-density mesh. Through this hierarchical mesh design, the optimal allocation of computational resources is achieved, significantly improving the practicality of the imaging system.

[0013] Optionally, the step of locally refining the initial imaging grid to obtain the two-dimensional imaging network specifically includes: performing global imaging with the initial imaging grid to obtain an initial pixel intensity value matrix; identifying candidate regions of interest in the initial pixel intensity value matrix whose intensity values ​​are greater than a preset initial screening threshold; and increasing the density of the first node and the density of the second node to preset multiples of the initial density within the axial coordinate range and circumferential angle range corresponding to the candidate regions of interest, respectively, to generate the two-dimensional imaging network.

[0014] By employing the aforementioned technical solution, a rapid coarse scan of the entire circular pipe structure is achieved through global imaging based on the initial imaging grid to obtain the initial pixel intensity value matrix, providing preliminary location information for identifying potential damage areas. By setting a preset initial screening threshold to identify candidate areas of interest, regions with abnormal intensity values ​​are effectively filtered out. These regions have a high probability of damage presence, avoiding indiscriminate high-density calculations across the entire imaging area. Within the candidate areas of interest, the density of the first and second nodes is increased to a preset multiple of the initial density, achieving local grid densification. This significantly improves the spatial sampling density of the damage area, resulting in more refined damage morphology features. This adaptive densification strategy fully utilizes the spatial sparsity of the damage signal, concentrating computational resources on areas truly requiring high-resolution imaging. Through this two-stage imaging strategy, both the detection sensitivity and location accuracy for minor damage are ensured, while significantly reducing the overall computational load. This method meets the needs of real-time online monitoring and is particularly suitable for efficient detection of long-distance pipelines.

[0015] Optionally, before performing time-domain pointwise subtraction between the current state signal matrix and the healthy baseline signal matrix, the method further includes: performing DC component removal processing on the healthy baseline signal matrix and the current state signal matrix; performing detrending processing on the DC-removed healthy baseline signal matrix and the DC-removed current state signal matrix, fitting the trend term of the signal using the least squares method, and removing the trend term; determining an effective signal time window based on the effective propagation distance and group velocity of the guided wave in the circular tube structure, and performing time-domain truncation on the healthy baseline signal matrix and the current state signal matrix based on the effective signal time window.

[0016] By employing the above technical solutions, the DC component removal process eliminates systematic errors caused by sensor zero-point drift and DC bias in the acquisition system, ensuring that the healthy baseline signal matrix and the current state signal matrix have the same reference zero point, laying the foundation for the accuracy of subsequent differential operations. The de-trend processing uses the least squares method to fit and remove linear trend terms from the signal, effectively eliminating baseline drift caused by factors such as slow changes in ambient temperature and aging of electronic components, improving signal stability and consistency. The effective signal time window is determined based on the effective propagation distance and group velocity, precisely defining the time period containing useful scattering information and avoiding interference from invalid data in subsequent processing. Time-domain truncation based on the effective signal time window removes tail noise and irrelevant signal segments generated by multiple reflections, not only purifying the data quality but also significantly reducing the dimensionality of the data matrix. This series of preprocessing steps systematically improves signal quality, reduces computational complexity, and ensures that the differential signal matrix accurately reflects the weak scattering characteristics caused by damage, creating favorable conditions for high-precision damage imaging.

[0017] Optionally, after generating a two-dimensional heatmap characterizing the location and morphology of damage to the circular tube structure based on the intensity value of each pixel, the method further includes: converting the two-dimensional heatmap into a binary image based on a preset intensity threshold, wherein pixels with intensity values ​​higher than the preset intensity threshold are identified as damaged areas, and pixels with intensity values ​​lower than or equal to the preset intensity threshold are identified as healthy areas; using an image connected component analysis algorithm to identify and mark damaged areas in the binary image; calculating the geometric center and area of ​​each damaged area, using the physical location corresponding to the geometric center as the damage localization result, converting the area into actual physical size, and using the actual physical size as a quantitative indicator of damage size.

[0018] By employing the above technical solution, a two-dimensional heatmap is converted into a binary image based on a preset intensity threshold, realizing the transformation from continuous intensity distribution to discrete damage assessment. This clearly distinguishes between damaged and healthy areas, simplifying the subsequent quantitative analysis process. By appropriately setting the intensity threshold, low-intensity background noise is effectively filtered out while retaining the high-intensity response generated by actual damage, improving the accuracy of damage identification. Using an image connected component analysis algorithm to identify and label independent damage regions can distinguish multiple discrete damage locations, avoiding erroneous merging of adjacent damages and providing a basis for independent assessment of each damage. Calculating the geometric center of the damage region as the damage localization result provides precise coordinate information of the damage on the circular pipe, facilitating rapid location and repair by maintenance personnel. Converting the area of ​​the damage region into actual physical dimensions as a quantitative indicator of damage size achieves accurate mapping from image pixels to physical dimensions, providing a quantitative basis for damage severity assessment and maintenance decisions. This automated damage localization and quantification method significantly improves the practicality of the detection results, enabling the monitoring system to directly output actionable maintenance recommendations.

[0019] In a second aspect, embodiments of this application provide a circular tube waveguide damage imaging device, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions, and the one or more processors call the computer instructions to cause the circular tube waveguide damage imaging device to perform the method described in the first aspect and any possible implementation thereof.

[0020] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a circular tube waveguide damage imaging device, cause the circular tube waveguide damage imaging device to perform the method described in the first aspect and any possible implementation thereof.

[0021] Fourthly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a circular tube waveguide damage imaging device, cause the circular tube waveguide damage imaging device to perform the method described in the first aspect and any possible implementation thereof.

[0022] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages: 1. Efficient and accurate damage localization: By using differential processing and Hilbert transform based on the healthy baseline signal matrix and the current state signal matrix, combined with the virtual unfolding of the three-dimensional cylindrical surface of the circular tube and the discretization of the two-dimensional imaging grid, this application realizes two-dimensional visualization imaging of damage, which can intuitively characterize the location and morphology of damage to the circular tube structure, thus improving the efficiency and accuracy of damage localization.

[0023] 2. Enhanced local detail detection capability: By performing a full-domain scan of the initial imaging grid and local refinement processing of candidate regions, this application can improve the resolution of the imaging grid in key areas of interest, thereby capturing more detailed damage information and enhancing the detection capability for complex damage morphologies.

[0024] 3. Quantitative and intelligent damage assessment: After generating a two-dimensional heat map, this application uses binary image processing and image connected component analysis algorithms to automatically identify the damaged area, calculate the geometric center and physical size of the damage, realize quantitative damage assessment, reduce manual intervention, and improve the intelligence and automation level of the monitoring process. Attached Figure Description

[0025] Figure 1 This is a schematic flowchart of a circular tube guided wave damage imaging method disclosed in an embodiment of this application; Figure 2 This is another schematic flowchart of a circular tube guided wave damage imaging method disclosed in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a device provided in an embodiment of this application.

[0026] Explanation of reference numerals in the attached drawings: 301, Central Processing Unit; 302, Read-Only Memory; 303, Random Access Memory; 304, Bus; 305, Input / Output Interface; 306, Input Section; 307, Output Section; 308, Storage Section; 309, Communication Section; 310, Driver; 311, Removable Media. Detailed Implementation

[0027] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0028] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0029] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple system devices refer to two or more system devices, and multiple screen terminals refer to two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0030] This application provides a method for imaging damage in a circular tube guided wave, referring to... Figure 1 , Figure 1 This is a schematic flowchart of a circular tube waveguide damage imaging method provided in an embodiment of this application. The method is applied to a server, which can execute a circular tube waveguide damage imaging program. The method includes steps S101 to S105, as follows: Step S101: Collect the health baseline signal matrix of the circular tube structure in a healthy state and the current state signal matrix in a test state by using a sensor array deployed on the circular tube structure to be monitored.

[0031] In step S101, the circular tube structure refers to a pipe or tubular structural component whose health status is to be monitored. The sensor array refers to a collection of multiple sensors arranged in a predetermined spatial layout on the surface of the circular tube structure for exciting and receiving guided wave signals. The healthy state refers to the initial intact state of the circular tube structure before use or when it is confirmed to be undamaged. The health baseline signal matrix refers to the set of time-domain signal data collected by the sensor array in the healthy state, which can fully characterize the propagation characteristics of guided waves in a defect-free structure. The state under test refers to the current state of the circular tube structure during service or after a period of time, when damage detection is required. The current state signal matrix refers to the set of time-domain signal data collected by the sensor array in the state under test, containing information on scattered waves that may be caused by damage.

[0032] Specifically, the server first controls the data acquisition system connected to the sensor array. In the healthy state of the circular tube structure, the server designates one or more sensors in the sensor array as excitation sources to emit preset guided wave signals, such as ultrasonic guided waves of a specific frequency and period. The guided waves propagate along the circular tube structure, and all sensors in the sensor array act as receivers, synchronously acquiring response voltage signals within a preset time period. The server discretizes the analog signals acquired by each sensor into digital time series using an analog-to-digital converter and combines the time series of all sensors into a matrix, which is the health baseline signal matrix. The rows of this matrix correspond to different sensors, and the columns correspond to different sampling time points. In subsequent monitoring cycles, the server, in the state under test, repeats the above process using the exact same excitation signal and acquisition parameters to acquire and generate the current state signal matrix.

[0033] Step S102: Subtract the current state signal matrix from the healthy baseline signal matrix point by point in the time domain to obtain the difference signal matrix. Perform a Hilbert transform on the difference signal matrix and take its absolute value to obtain the amplitude envelope matrix containing only positive values.

[0034] In step S102, the differential signal matrix is ​​the matrix obtained by subtracting the current state signal matrix from the healthy baseline signal matrix point by point in the time domain. This matrix mainly contains the scattered signal components caused by structural state changes, especially damage. The Hilbert transform is a mathematical signal processing method used to generate analytic signals, thus facilitating the calculation of the instantaneous amplitude and phase of the signal. The amplitude envelope matrix is ​​the matrix obtained by performing a Hilbert transform on each row of the time-domain signal in the differential signal matrix and taking its magnitude. Each element of this matrix is ​​a positive value, representing the instantaneous energy or amplitude of the differential signal at each moment.

[0035] Specifically, the server reads the health baseline signal matrix and the current state signal matrix from the storage device. Since both matrices were obtained under the same acquisition conditions, they have the same dimensions. The server performs matrix subtraction, subtracting the corresponding element from the health baseline signal matrix from each element in the current state signal matrix to generate a differential signal matrix. This operation effectively suppresses stable signal components unrelated to damage, such as direct waves and end-face reflected waves from the excitation signal, thus highlighting the weak scattered signals generated by the damage. Subsequently, the server applies the Hilbert transform algorithm to each row of the differential signal matrix, i.e., the differential signal time series of each sensor, to obtain a complex time series. Next, the server calculates the absolute value, i.e., the amplitude, at each point in this complex time series, thereby obtaining the envelope of that row of signals. After processing all rows, an amplitude envelope matrix containing only positive real values ​​is generated.

[0036] In one possible implementation, before performing point-by-point subtraction in the time domain between the current state signal matrix and the healthy baseline signal matrix, the method further includes: performing DC component removal processing on the healthy baseline signal matrix and the current state signal matrix; performing detrending processing on the DC-removed healthy baseline signal matrix and the DC-removed current state signal matrix, fitting the trend terms of the signals using the least squares method, and removing the trend terms; determining the effective signal time window based on the effective propagation distance and group velocity of the guided wave in the circular tube structure, and performing time-domain truncation on the healthy baseline signal matrix and the current state signal matrix based on the effective signal time window.

[0037] Specifically, the server first receives the raw health baseline signal matrix and the current status signal matrix from the data acquisition system. These two matrices contain voltage signal data collected by all sensor channels in the sensor array at different times.

[0038] In the DC component removal stage, the server processes the signal from each sensor channel separately. The server calculates the arithmetic mean of each channel's signal over the entire time series; this average value is the DC bias component for that channel. Subsequently, the server subtracts the corresponding DC bias value from the value of each sample point in the original signal. For example, if the original signal sequence of a sensor channel is [2.1, 2.3, 1.9, 2.2, 2.0] volts, and the calculated average is 2.1 volts, then the DC-removed signal becomes [0, 0.2, -0.2, 0.1, -0.1] volts. This processing eliminates systematic biases caused by sensor zero-point drift or data acquisition card bias voltage.

[0039] In the detrending stage, the server uses the least squares method to fit a linear trend to the DC-free signal. The server constructs a first-order polynomial model of the time series and solves for the slope and intercept parameters of the trend line by minimizing the sum of squared residuals. After calculating the trend term, the server subtracts the corresponding trend value point-by-point from the DC-free signal. This step primarily eliminates signal baseline drift caused by slow changes in ambient temperature or aging of electronic components, ensuring the accuracy of subsequent differential calculations.

[0040] In the effective signal time window determination phase, the server calculates the effective propagation time based on the physical parameters of the circular pipe structure. The server first reads the pipe length (e.g., 0.4 meters) and waveguide group velocity (e.g., 3200 m / s) from the system configuration, and calculates the theoretical round-trip time for the guided wave to propagate from the excitation source to the end of the pipe and back. The server typically sets the effective time window to 1.5 to 2 times the theoretical round-trip time to ensure capture of the main scattered signal while avoiding signal aliasing caused by multiple reflections. For example, for a 0.4-meter-long pipe with a theoretical round-trip time of 0.25 milliseconds, the server might set the effective window to 0.4 milliseconds.

[0041] During the time-domain truncation implementation phase, the server calculates the number of sampling points to be retained based on the determined effective time window and system sampling rate. If the sampling rate is 10MHz, a 0.4ms window corresponds to 4000 sampling points. The server performs the same truncation operation on each channel of the healthy baseline signal matrix and the current state signal matrix, retaining only the first 4000 data points. This uniform truncation ensures dimensionality matching between the two matrices during subsequent difference operations, while significantly reducing data processing volume and improving server computational efficiency.

[0042] Through this series of preprocessing steps, the server effectively cleans up the raw signal, removing various systematic noises and invalid data, laying a reliable data foundation for subsequent differential operations and damage imaging.

[0043] Step S103: Virtually unfold the three-dimensional cylindrical surface of the circular tube structure into a two-dimensional plane, and discretize the three-dimensional cylindrical surface into a two-dimensional imaging mesh containing multiple pixels, where each pixel corresponds to a physical position on the circular tube structure.

[0044] In step S103, the three-dimensional cylindrical surface refers to the geometric surface shape of the circular tube structure in three-dimensional space. The two-dimensional imaging mesh refers to the virtual mesh structure formed by unfolding the three-dimensional cylindrical surface into a two-dimensional rectangular plane along the axial and circumferential directions, and then discretizing the plane. A pixel is the smallest unit that constitutes the two-dimensional imaging mesh, and each pixel uniquely corresponds to a tiny physical region on the three-dimensional cylindrical surface of the circular tube structure through a coordinate mapping relationship.

[0045] Specifically, to facilitate calculation and visualization, the server establishes a virtual two-dimensional Cartesian coordinate system. The horizontal axis of this coordinate system represents the axial position of the circular tube structure, and the vertical axis represents the circumferential angle of the tube structure, for example, from 0 to 360 degrees or -π to π. Based on the monitoring area and the required imaging accuracy, the server divides both the axial and circumferential coordinates into a predetermined number of discrete nodes. The array of intersections formed by these discrete nodes constitutes the two-dimensional imaging grid. The server creates a data structure in memory to store this grid, where each pixel has a unique index and corresponding axial and circumferential coordinates, preparing for subsequent intensity value calculations.

[0046] In one possible implementation, the three-dimensional cylindrical surface is discretized into a two-dimensional imaging mesh containing multiple pixels, specifically including steps S1031-S1035, as follows: Step S1031: Establish a two-dimensional rectangular coordinate system with the axial direction of the circular pipe structure as the abscissa and the circumferential angle of the circular pipe structure as the ordinate. The range of the abscissa is from the starting position to the ending position of the pipe, and the range of the ordinate is from -π to π.

[0047] In step S1031, the two-dimensional rectangular coordinate system refers to a coordinate system composed of two mutually perpendicular number axes intersecting at the origin, used to determine the position of points on a plane. The horizontal axis in this step represents the length direction of the circular pipe structure along its centerline, i.e., its axial position. The vertical axis in this step represents the direction of rotation of the circular pipe structure around its centerline, i.e., its circumferential angle. The pipe's starting and ending positions refer to the axial coordinates of the two endpoints of the pipe section being inspected.

[0048] Specifically, the server establishes a logical two-dimensional Cartesian coordinate system. The server defines the horizontal axis of this coordinate system as representing the axial direction of the circular pipe, and sets the range of values ​​for the horizontal axis based on the actual length of the pipe to be detected, for example, from the starting position of 0 meters to the ending position of 50 meters. The server defines the vertical axis of this coordinate system as representing the circumferential angle of the circular pipe, and sets the range of values ​​for the vertical axis to be from -π to π radians, a range that completely covers the circumference of the entire circular pipe.

[0049] Step S1032: Discretize the axial distance into a preset number of first nodes and the circumferential angle into a preset number of second nodes to construct the initial imaging grid.

[0050] In step S1032, the first node refers to a series of discrete points selected on the axial coordinate axis. The second node refers to a series of discrete points selected on the circumferential angular coordinate axis. The initial imaging grid refers to a preliminary, low-resolution pixel grid covering the entire detection area, composed of the first node set and the second node set.

[0051] Specifically, the server discretizes the coordinate system established in the previous step based on preset initial resolution parameters. The server reads a preset number, such as 1000, as the number of first nodes and divides the axial coordinate range into 1000 equal segments, setting a first node at each segment point. Similarly, the server reads another preset number, such as 360, as the number of second nodes and divides the circumferential angle range from -π to π into 360 equal segments, setting a second node at each segment point. The combination of all the first and second nodes constitutes an initial imaging grid of 1000 rows and 360 columns.

[0052] Step S1033: Perform global imaging using the initial imaging grid to obtain the initial pixel intensity value matrix.

[0053] In step S1033, global imaging refers to performing a complete intensity value calculation process for each pixel in the initial imaging grid. The initial pixel intensity value matrix is ​​a two-dimensional array used to store the calculated intensity value corresponding to each pixel in the initial imaging grid.

[0054] Specifically, the server performs global imaging calculations on the initial imaging grid. The server iterates through every pixel in the initial imaging grid, performs a complete intensity value calculation process for each pixel, and calculates the intensity value of that point. The server stores all the calculated intensity values ​​in a 1000-row, 360-column matrix, which is the initial pixel intensity value matrix. The value and position of each element in the matrix correspond one-to-one with the pixel in the initial imaging grid.

[0055] Step S1034: Identify candidate regions of interest in the initial pixel intensity value matrix whose intensity values ​​are greater than the preset initial screening threshold.

[0056] In step S1034, the preset initial screening threshold is a pre-set intensity value standard used to initially determine whether a pixel might represent a defect. The candidate region of interest refers to the local area composed of pixels whose intensity values ​​exceed the preset initial screening threshold in the initial imaging.

[0057] Specifically, the server analyzes the initial pixel intensity value matrix to identify potential defective regions. The server loads a preset initial screening threshold and then iterates through each intensity value in the initial pixel intensity value matrix. When an intensity value is found to be greater than the preset initial screening threshold, the server marks that pixel as a candidate point. After the iteration is complete, the server performs cluster analysis on all marked candidate points, merging spatially adjacent candidate points into one or more rectangular regions; these regions are the candidate regions of interest. Each candidate region of interest is defined by an axial coordinate range and a circumferential angle range.

[0058] Step S1035: Within the axial coordinate range and circumferential angle range corresponding to the candidate region of interest, increase the density of the first node and the density of the second node to a preset multiple of the initial density to generate a two-dimensional imaging network.

[0059] In step S1035, the density of the first node represents the number of nodes distributed within a unit axial length. The density of the second node represents the number of nodes distributed within a unit circumferential angle. The preset multiplier is a value greater than 1, used to define the proportion by which the node density needs to be increased within the candidate region of interest. The two-dimensional imaging network is the final generated pixel grid for fine imaging, which has low resolution in most areas but significantly improved resolution in the candidate region of interest.

[0060] Specifically, the server refines the mesh within the identified candidate regions of interest to generate the final 2D imaging network. The server reads a preset density enhancement factor, such as 10. For each candidate region of interest, the server re-discretizes the nodes within its corresponding axial coordinate range and circumferential angle range. The new first node density will be 10 times the initial density, and the new second node density will also be 10 times the initial density. This generates a sub-mesh much denser than the initial mesh within the candidate regions of interest. The final generated 2D imaging network is a non-uniform mesh composed of the initial sparse mesh portion of all non-candidate regions of interest and the refined dense mesh portion of all candidate regions of interest. Step S104: Traverse each pixel in the 2D imaging mesh and calculate the intensity value of each pixel based on the magnitude envelope matrix.

[0061] In step S104, the intensity value refers to a scalar value calculated for each pixel in the two-dimensional imaging grid. The magnitude of this value reflects the possibility of damage at the physical location corresponding to that pixel.

[0062] Specifically, the server initiates a traversal process, sequentially processing each pixel in the 2D imaging grid. For the currently processed pixel, the server first calculates its 3D spatial coordinates on the circular tube structure using coordinate transformation formulas based on its axial and circumferential coordinates in the 2D imaging grid. Then, based on the known waveguide group velocity, the server calculates the total flight time of the guided wave propagating from the excitation source location to the pixel's 3D spatial position, and then scattering and propagating from that position to each receiving sensor in the sensor array. This process calculates a unique total flight time for each receiving sensor. The server uses the signal sampling time step to convert these calculated total flight times into discrete time indices, i.e., column indices in the amplitude envelope matrix. Subsequently, the server extracts the amplitude of each sensor signal at the corresponding time from the amplitude envelope matrix based on these indices. Finally, the server sums all the extracted amplitudes and assigns this sum to the current pixel as its intensity value. The server repeats this process until the intensity values ​​of all pixels in the 2D imaging grid have been calculated.

[0063] In one possible implementation, the intensity value of each pixel is calculated based on the amplitude envelope matrix, specifically including steps S1041-S1045, as follows: Step S1041: Convert the position of the pixel in the two-dimensional imaging grid into spatial coordinates in the three-dimensional Cartesian coordinate system through coordinate mapping relationship. The spatial coordinates include the horizontal coordinate and the vertical coordinate. The horizontal coordinate is the product of the inner diameter of the tube and the cosine of the circumferential angle of the pixel. The vertical coordinate is the product of the inner diameter of the tube and the sine of the circumferential angle of the pixel.

[0064] In step S1041, a three-dimensional Cartesian coordinate system refers to a coordinate system that determines the position of a point in space using three mutually perpendicular coordinate axes. Spatial coordinates refer to a set of values ​​used to uniquely determine the position of a point in a three-dimensional Cartesian coordinate system; this set of values ​​typically consists of an abscissa, a ordinate, and an axial coordinate. The abscissa and ordinate refer to the two coordinate values ​​that determine the position of a point on a two-dimensional plane perpendicular to the axis of the circular tube. The inner diameter of the circular tube refers to the diameter inside the circular tube structure. The circumferential angle of a pixel refers to the position of a pixel on the circumference of the circular tube structure, usually expressed in angles or radians.

[0065] Specifically, when the server begins calculating the intensity value of a pixel, it first performs a coordinate transformation task. The server reads the circumferential angle and axial coordinates of the pixel to be processed from the two-dimensional imaging grid. Simultaneously, the server obtains the geometric parameters of the circular tube structure, i.e., the inner diameter of the tube, from a preset configuration file. Based on the coordinate mapping relationship defined in step S1041, the server performs the following calculations: multiplying the inner diameter of the tube by the cosine of the pixel's circumferential angle yields the abscissa; multiplying the inner diameter of the tube by the sine of the pixel's circumferential angle yields the ordinate. The server combines the calculated abscissa and ordinate with the axial coordinate read from the two-dimensional imaging grid to form a three-dimensional vector, which represents the spatial coordinates of the pixel in a three-dimensional Cartesian coordinate system.

[0066] Step S1042: Calculate the first flight time of the guided wave propagating from the excitation source position along the axial direction to the pixel. The first flight time is equal to the ratio of the axial coordinate of the pixel to the velocity of the guided wave group.

[0067] In step S1042, the first flight time is the time required for the guide wave energy to propagate axially from the excitation source location to the circular tube cross-section with the same axial coordinate as the pixel to be calculated. The waveguide group velocity is the effective speed at which the guide wave energy propagates in the circular tube structure, and it is the fundamental physical quantity for all flight time calculations.

[0068] Specifically, the server calculates the first flight time based on the pixel's axial coordinates obtained in the previous step, combined with the known axial position of the excitation source and the pre-determined or set waveguide group velocity. According to step S1042, this calculation is simplified. The server directly treats the pixel's axial coordinate value as the distance the guided wave propagates along the axial direction, and then divides this distance by the waveguide group velocity. The quotient is the first flight time. This time represents the time required for the guided wave packet to propagate from the excitation source along the pipe wall to the cross-section where the pixel is located.

[0069] Step S1043: Calculate the second flight time of the guided wave after it is scattered from the pixel and propagates to the sensor. The second flight time is equal to the ratio of the three-dimensional Euclidean distance between the pixel and the sensor to the velocity of the guided wave group. Add the first flight time and the second flight time to obtain the total flight time.

[0070] In step S1043, the second flight time is the time required for the guided wave to propagate from the pixel location to the sensor location as a scattered wave. The three-dimensional Euclidean distance refers to the straight-line distance between the pixel and the sensor in three-dimensional space. The total flight time refers to the complete time from the emission of the guided wave from the excitation source to its reception by the specific sensor after being scattered by the pixel location.

[0071] Specifically, the server initiates a traversal process, processing each sensor in the sensor array sequentially. For the currently processed sensor, the server first retrieves the precise three-dimensional spatial coordinates of the sensor from storage; these coordinates are pre-calibrated and stored. Using the pixel's three-dimensional spatial coordinates calculated in step S1041 and the current sensor's three-dimensional spatial coordinates, the server applies the distance formula between two points in three-dimensional space to calculate the three-dimensional Euclidean distance between the pixel and the sensor. Subsequently, the server divides the calculated three-dimensional Euclidean distance by the waveguide group velocity, and the result is the second flight time. Finally, the server adds the first flight time calculated in step S1042 to the currently calculated second flight time, and the result is the total flight time, which uniquely corresponds to the current pixel and the current sensor.

[0072] Step S1044: Calculate the discrete time index of each sensor based on the total flight time and the signal sampling time step.

[0073] In step S1044, the signal sampling time step refers to the time interval between two adjacent sampling points when discretizing a continuous analog signal during data acquisition. The discrete-time index refers to converting continuous time values ​​into an integer position number in the digital signal sequence, which is used for precise addressing within the matrix storing the signal data.

[0074] Specifically, after calculating the total flight time, the server needs to convert this continuous time quantity into an index that can be used in the discrete signal matrix. The server reads the signal sampling time step set during signal acquisition from the system configuration. Then, the server divides the total flight time obtained in step S1043 by the signal sampling time step. Since the result of the division operation may be a floating-point number, while the matrix index must be an integer, the server performs a rounding operation on the result, such as rounding down or rounding to the nearest integer value. This integer value is the discrete-time index used to locate the signal amplitude.

[0075] Step S1045: Based on the discrete time index, extract and accumulate the amplitude values ​​of each sensor at the corresponding index time from the amplitude envelope matrix, and use the sum as the intensity value of the pixel.

[0076] In step S1045, the summation refers to the sum obtained by adding up multiple amplitude data corresponding to all sensors extracted from the amplitude envelope matrix for the same pixel point. This sum is ultimately used as the intensity value of the pixel point.

[0077] Specifically, within the loop iterating through the sensors, the server uses the discrete-time index calculated in the previous step. Using the current sensor number as the row number and the discrete-time index as the column number, the server locates a unique element in the amplitude envelope matrix. The server extracts the value of this element, which is the signal amplitude at the corresponding moment. The server maintains an accumulator variable with an initial value of zero for the currently processed pixel. The server adds the extracted amplitude to this accumulator variable. The server continues to iterate through the next sensor, repeating steps S1043 to S1045, calculating the new amplitude and adding it to the same accumulator variable. When all sensors in the sensor array have been processed, the final value in this accumulator variable is the sum. The server assigns this sum to the current pixel as its final intensity value.

[0078] Step S105: Based on the intensity value of each pixel, generate a two-dimensional thermogram representing the location and morphology of damage to the circular tube structure.

[0079] In step S105, the two-dimensional heat map is a data visualization image that uses color spectrum to represent the intensity value of each pixel on the two-dimensional imaging grid, thereby intuitively showing the location, shape and severity of damage on the unfolded plane of the circular tube structure.

[0080] Specifically, after calculating the intensity values ​​of all pixels, the server obtains an intensity value matrix of the same size as the two-dimensional imaging grid. The server calls a graphics generation function to create a blank image corresponding to the size of the two-dimensional imaging grid. Then, the server iterates through the intensity value matrix, mapping the intensity value of each pixel to a preset color gradient. For example, the lowest intensity value is mapped to blue, the highest intensity value to red, and the intermediate intensity values ​​are mapped proportionally to a transitional color from blue to red. The server fills the corresponding pixel position in the image with the mapped color. The final generated color image is a two-dimensional heatmap. The brightly colored "hot spots" in the image clearly indicate the location and approximate shape of damage on the circular pipe structure. The server can output this image to a display device or save it as a file for technicians to analyze.

[0081] To facilitate understanding, the following concrete example will be used to explain steps S101-S105: Taking a 5-meter-long, 200-millimeter-diameter oil pipeline as an example, this method achieves remote online monitoring of damage through a server. First, a sensor array consisting of 16 piezoelectric sensors is uniformly installed circumferentially at a 1-meter axial position along the pipeline. Before the pipeline is put into operation and confirmed to be in a healthy, undamaged state, the server controls the data acquisition system, sequentially activating each sensor as an excitation source to emit 300kHz ultrasonic guided waves. Simultaneously, all 16 sensors receive signals. After the acquisition process is complete, the server integrates all data into a health baseline signal matrix and stores it. Six months after the pipeline has been operating, the server automatically executes the exact same acquisition process in a test-ready state, generating the current status signal matrix. Assume that a pit has formed due to corrosion at a position 3.5 meters axially and 270 degrees circumferentially along the pipeline.

[0082] After receiving the two signal matrices, the server immediately performs point-by-point subtraction in the time domain to obtain the differential signal matrix that mainly reflects the scattering signal of the corrosion pits. Then, it performs a Hilbert transform on this matrix and takes its absolute value to generate the amplitude envelope matrix. At the same time, the server has virtually unfolded and discretized the monitoring area of ​​the pipeline, namely the cylindrical surface with an axial length of 1 to 5 meters and a circumferential angle of 0 to 360 degrees, into a high-resolution two-dimensional imaging grid.

[0083] The server then begins the core imaging calculations. It iterates through every pixel on the two-dimensional imaging grid. When calculating the pixel at coordinates (3.5 meters, 270 degrees), the server calculates the theoretical flight time of the guided wave propagating from each excitation source to that point and then scattering back to all 16 receiving sensors. Since this pixel is precisely the location of the actual damage, the moments corresponding to these theoretical flight times are the moments when energy peaks appear in the amplitude envelope matrix due to damage scattering. Therefore, when the server extracts and sums the amplitude values ​​at these moments from the amplitude envelope matrix, it obtains a very large sum, i.e., a high intensity value. Conversely, for other pixels far from the damage, the values ​​in the amplitude envelope matrix corresponding to their theoretical flight times will be very small, and the summed intensity value will also be correspondingly low.

[0084] After calculating the intensity values ​​of all pixels, the server generates a two-dimensional heatmap based on these values. Most areas of this map display cool tones, such as dark blue, representing low intensity values. However, a bright red or yellow spot appears at a location on the image corresponding to a point 3.5 meters axially and 270 degrees circumferentially. By observing this two-dimensional heatmap, maintenance personnel can accurately determine the precise location and approximate size of corrosion damage without touching the pipes, thus providing a reliable basis for subsequent maintenance decisions.

[0085] Please refer to Figure 2 In one possible implementation, after generating a two-dimensional heat map characterizing the location and morphology of damage to the circular tube structure based on the intensity value of each pixel, the method further includes steps S201-S203, as follows: Step S201: Based on a preset intensity threshold, the two-dimensional heat map is converted into a binary image, wherein pixels with intensity values ​​higher than the preset intensity threshold are identified as damaged areas, and pixels with intensity values ​​lower than or equal to the preset intensity threshold are identified as healthy areas.

[0086] In step S201, the preset intensity threshold refers to a pre-defined critical intensity value used to distinguish between damaged and non-damaged pixels. A binarized image is an image where each pixel has only two possible values; in this step, these two values ​​represent damaged and healthy regions, respectively. A damaged region is a set of pixels with intensity values ​​higher than the preset intensity threshold. A healthy region is a set of pixels with intensity values ​​lower than or equal to the preset intensity threshold.

[0087] Specifically, based on preset decision criteria, the server converts a two-dimensional heatmap containing continuous color changes into a binary image containing only two states. The server loads a preset intensity threshold, typically set based on normalized intensity values, such as 0.8. The server creates a new blank image of the same size as the two-dimensional heatmap as the carrier for the binary image. Then, the server iterates through each pixel of the two-dimensional heatmap, reading its intensity value. If the pixel's intensity value is higher than the preset intensity threshold, the server sets the pixel value to 1 at the corresponding position in the binary image, indicating that the point belongs to a damaged area. If the pixel's intensity value is lower than or equal to the preset intensity threshold, the server sets the pixel value to 0 at the corresponding position in the binary image, indicating that the point belongs to a healthy area. After processing all pixels, the server obtains a complete binary image that clearly distinguishes the contours of potential damage.

[0088] Step S202: Use an image connected component analysis algorithm to identify and mark the damaged regions in the binarized image.

[0089] In step S202, the image connectivity analysis algorithm refers to a class of algorithms used to find and mark all interconnected clusters of pixels with the same pixel value in a digital image.

[0090] Specifically, the server analyzes the binarized image generated in the previous step to identify and separate individual damage regions. The server uses an image connectivity analysis algorithm to process the binarized image. This algorithm scans all pixels in the image with a value of 1. When an unlabeled pixel is found, the algorithm uses that point as a starting point and searches for all adjacent pixels with the same value of 1 to find the entire connected pixel block. All pixels belonging to the same connected block are assigned the same unique label, such as "Damage Region 1". The algorithm repeats this process until all pixels with a value of 1 in the binarized image are classified into a labeled damage region. Through this step, even if multiple unconnected damages exist in the image, they can be accurately identified and independently labeled.

[0091] Step S203: Calculate the geometric center and area of ​​each damaged region, take the physical location corresponding to the geometric center as the damage location result, convert the area into the actual physical size, and take the actual physical size as the quantitative indicator of damage size.

[0092] In step S203, the geometric center refers to the arithmetic mean of the coordinates of all pixels within a region, representing the center of the region. The area refers to the total number of pixels constituting a region. The damage localization result refers to the calculated physical location of the damage on the circular tube structure. The actual physical size refers to the converted size of the damaged area in the real world. The damage size quantification index is a numerical value with physical units used to describe the severity or extent of the damage.

[0093] Specifically, the server calculates the geometric parameters for each identified and marked damage region to achieve precise damage localization and quantification. For each marked damage region, the server first calculates its geometric center. The server iterates through all pixels within the damage region, accumulating the axial coordinates and circumferential angles of all pixels, and then divides each by the total number of pixels in the region to obtain the average axial coordinates and average circumferential angle of the geometric center. The physical location corresponding to this coordinate point is recorded as the damage localization result. Simultaneously, the server calculates the area of ​​the damage region by counting the total number of pixels within it. Based on the resolution parameters used when generating the 2D imaging network, the server calculates the actual physical area represented by a single pixel, for example, axial resolution multiplied by circumferential arc length resolution. Finally, the server multiplies the total number of pixels by the actual physical area of ​​a single pixel; the result is the actual physical size of the damage, which serves as a quantitative indicator of damage severity. The server repeats this calculation process for all identified damage regions.

[0094] The following describes a circular tube guided wave damage imaging device according to an embodiment of the present invention from the perspective of hardware processing. Please refer to [link to relevant documentation]. Figure 3 This is a schematic diagram of the structure of a circular tube guided wave damage imaging device in an embodiment of this application.

[0095] It should be noted that, Figure 3 The structure of the circular tube guided wave damage imaging device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0096] like Figure 3 As shown, a circular tube guided wave damage imaging device includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage section 308 into a random access memory (RAM) 303, such as performing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for device operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0097] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0098] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the various functions defined in the present invention.

[0099] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0100] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0101] Specifically, the circular tube waveguide damage imaging device of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the circular tube waveguide damage imaging method provided in the above embodiment.

[0102] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the circular tube waveguide damage imaging device described in the above embodiments; or it may exist independently and not assembled into the circular tube waveguide damage imaging device. The storage medium carries one or more computer programs, which, when executed by a processor of the circular tube waveguide damage imaging device, cause the circular tube waveguide damage imaging device to implement the circular tube waveguide damage imaging method based on IoT data encryption transmission provided in the above embodiments.

Claims

1. A method for imaging damage in a circular tube guided wave, characterized in that, The method includes: The health baseline signal matrix of the circular tube structure in a healthy state and the current state signal matrix in the test state are collected by a sensor array deployed on the circular tube structure to be monitored. The current state signal matrix and the healthy baseline signal matrix are subtracted point by point in the time domain to obtain a difference signal matrix. The difference signal matrix is ​​then subjected to a Hilbert transform and its absolute value is taken to obtain an amplitude envelope matrix containing only positive values. The three-dimensional cylindrical surface of the circular tube structure is virtually unfolded into a two-dimensional plane, and the three-dimensional cylindrical surface is discretized into a two-dimensional imaging grid containing multiple pixels, where each pixel corresponds to a physical position on the circular tube structure. Traverse each pixel in the two-dimensional imaging grid and calculate the intensity value of each pixel based on the amplitude envelope matrix; Based on the intensity value of each pixel, a two-dimensional thermogram is generated to characterize the location and morphology of damage to the circular tube structure.

2. The method according to claim 1, characterized in that, The calculation of the intensity value of each pixel based on the amplitude envelope matrix specifically includes: Calculate the total flight time of the guided wave from the excitation source to the pixel and from the pixel to each sensor in the sensor array; Based on the total flight time and signal sampling time step, the discrete time index of each sensor is calculated; Based on the discrete-time index, the amplitude values ​​of each sensor at the corresponding index time are extracted and accumulated from the amplitude envelope matrix, and the sum is used as the intensity value of the pixel.

3. The method according to claim 2, characterized in that, The calculation of the total flight time of the guided wave propagating from the excitation source to the pixel and being scattered by the pixel to each sensor in the sensor array specifically includes: The position of the pixel in the two-dimensional imaging grid is converted into spatial coordinates in the three-dimensional Cartesian coordinate system through coordinate mapping relationship. The spatial coordinates include abscissa and ordinate. The abscissa is the product of the inner diameter of the tube and the cosine of the circumferential angle of the pixel. The ordinate is the product of the inner diameter of the tube and the sine of the circumferential angle of the pixel. Calculate the first flight time of the guided wave propagating axially from the excitation source location to the pixel point, where the first flight time is equal to the ratio of the axial coordinate of the pixel point to the guided wave group velocity; The second flight time of the guided wave after scattering from the pixel to the sensor is calculated. The second flight time is equal to the ratio of the three-dimensional Euclidean distance between the pixel and the sensor to the velocity of the guided wave group. The first flight time and the second flight time are added together to obtain the total flight time.

4. The method according to claim 1, characterized in that, Discretizing the three-dimensional cylindrical surface into a two-dimensional imaging mesh containing multiple pixels specifically includes: A two-dimensional rectangular coordinate system is established with the axial direction of the circular pipe structure as the horizontal axis and the circumferential angle of the circular pipe structure as the vertical axis. The value range of the horizontal axis is from the starting position to the ending position of the pipe, and the value range of the vertical axis is from -π to π. The axial distance is discretized into a preset number of first nodes, and the circumferential angle is discretized into a preset number of second nodes to construct an initial imaging grid. The initial imaging grid is then locally refined to obtain the two-dimensional imaging network.

5. The method according to claim 4, characterized in that, The step of locally refining the initial imaging grid to obtain the two-dimensional imaging network specifically includes: Global imaging is performed using the initial imaging grid to obtain an initial pixel intensity value matrix; In the initial pixel intensity value matrix, candidate regions of interest with intensity values ​​greater than a preset initial screening threshold are identified; Within the axial coordinate range and circumferential angle range corresponding to the candidate region of interest, the density of the first node and the density of the second node are increased to a preset multiple of the initial density to generate the two-dimensional imaging network.

6. The method according to claim 1, characterized in that, Before performing the point-by-point time-domain subtraction between the current state signal matrix and the health baseline signal matrix, the method further includes: The DC component is removed from the health baseline signal matrix and the current state signal matrix; The healthy baseline signal matrix and the current state signal matrix after DC removal are detrended. The trend term of the signal is fitted by the least squares method and then removed. Based on the effective propagation distance and group velocity of the guided wave in the circular tube structure, an effective signal time window is determined, and the healthy baseline signal matrix and the current state signal matrix are truncated in the time domain based on the effective signal time window.

7. The method according to claim 1, characterized in that, After generating a two-dimensional thermal map characterizing the location and morphology of damage to the circular tube structure based on the intensity value of each pixel, the method further includes: Based on a preset intensity threshold, the two-dimensional heat map is converted into a binary image, wherein pixels with an intensity value higher than the preset intensity threshold are identified as damaged areas, and pixels with an intensity value lower than or equal to the preset intensity threshold are identified as healthy areas. An image connected component analysis algorithm is used to identify and label damaged regions in the binarized image; Calculate the geometric center and area of ​​each damaged region, take the physical location corresponding to the geometric center as the damage localization result, convert the area into actual physical size, and take the actual physical size as the damage size quantification index.

8. A circular tube guided wave damage imaging device, characterized in that, The circular tube waveguide damage imaging device includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the circular tube waveguide damage imaging device to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the circular tube waveguide damage imaging device, the circular tube waveguide damage imaging device performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the circular tube waveguide damage imaging device, the circular tube waveguide damage imaging device performs the method as described in any one of claims 1-7.

Images