A Phased Array Noise Reduction and Visualization Detection Sensing Method for High-Temperature Pipelines

Abstract

This invention provides a phased array noise reduction and visualization detection sensing method for high-temperature pipelines. By acquiring acoustic signals and infrared thermal imaging data from the surface of the high-temperature pipeline, the method identifies the silencing groove region and its two edges within the acoustic signal, and extracts the radial distance of each region from the pipeline's central axis. Based on the uniformity of acoustic energy distribution within the silencing groove region and the difference in acoustic energy on both sides of the edge, the acoustic deformation coefficient of the pipeline surface is calculated. Combining the distance between the peak acoustic energy points at the edge and the radial distance, the thermal expansion deformation coefficient is estimated. The two are then fused to obtain a comprehensive deformation coefficient, which is used to compensate for the detection data, generating a thermal map of the wall thickness of the high-temperature irregularly shaped pipeline, thus achieving precise defect location. This method utilizes the integrated wedge structure features as an analytical benchmark, fusing acoustic and thermal field information to effectively suppress high-temperature noise interference, compensate for thermo-acoustic coupling deformation, and improve the accuracy and reliability of imaging for irregularly shaped pipelines.

Description

Technical Field

[0001] This invention relates to the field of high-temperature pipeline detection technology, specifically to a phased array noise reduction and visualization detection sensing method for high-temperature pipelines. Background Technology

[0002] In industries such as petrochemicals, power, and nuclear energy, high-temperature pipelines, as critical pressure-bearing equipment, operate under high temperature, high pressure, and complex corrosive environments for extended periods. They are highly susceptible to defects such as creep, cracks, and wall thinning, seriously threatening the safe operation of the system. Therefore, real-time and accurate non-destructive testing and health monitoring of high-temperature pipelines are of great significance. Phased array ultrasonic testing technology, due to its advantages of controllable beam, intuitive imaging, and high testing efficiency, has been widely used in pipeline defect detection. However, in high-temperature environments, traditional ultrasonic testing faces problems such as coupling failure, severe noise interference, and drastic signal attenuation, leading to a decrease in the signal-to-noise ratio and reduced imaging accuracy. To improve testing reliability, existing technologies often employ an integrated wedge structure with silencing grooves and damping blocks to effectively suppress lateral vibrations and clutter interference, thereby improving beam directivity and signal-to-noise ratio.

[0003] However, under high-temperature conditions, pipe materials undergo thermal expansion, leading to non-uniform changes in pipe geometry (such as diameter and wall thickness distribution). Simultaneously, thermal deformation at the wedge-pipe contact surface can introduce sound beam deflection and propagation path distortion, thus affecting defect location accuracy and wall thickness measurement accuracy. Furthermore, irregularly shaped pipes (such as bends and reducers) have complex structures and uneven sound field distribution, further exacerbating imaging distortion and the risk of misjudgment. Existing detection methods mostly process ultrasonic signals and temperature field information independently, lacking a comprehensive compensation mechanism for the thermo-acoustic coupling deformation effect, making it difficult to achieve high-precision thermal imaging of pipe wall thickness and accurate defect location under high-temperature conditions. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art, and provides a high-temperature pipeline phased array noise reduction and visualization detection sensing method.

[0005] To achieve the above objectives, the present invention provides a phased array noise reduction and visualization detection sensing method for high-temperature pipelines, the method comprising: Acquire acoustic signals and infrared thermal imaging data from the surface of a high-temperature pipeline, acquire all silencing groove regions in the acoustic signals and the two edges of each silencing groove region, and acquire the radial distance of each silencing groove region from the central axis of the pipeline. Based on the uniformity of sound wave energy distribution in all the silencing groove regions and the difference in sound wave energy between the two adjacent sides of each silencing groove edge in each silencing groove region, the acoustic deformation coefficient of the pipe surface is obtained. The thermal expansion deformation coefficient of the pipe surface is obtained based on the distance between the peak points of sound wave energy on the edges of the two sound-absorbing grooves in each sound-absorbing groove region and the radial distance of each sound-absorbing groove region. Based on the acoustic deformation coefficient and the thermal expansion deformation coefficient, the comprehensive deformation coefficient of the pipe surface is obtained; based on the comprehensive deformation coefficient of the pipe surface, the thermal map of the high-temperature irregular pipe wall thickness and the defect location results are obtained.

[0006] Optionally, the method for obtaining the silencing groove region includes: Obtain all acoustic boundaries in the sound wave signal, and obtain the average area of ​​the region enclosed by all the acoustic boundaries. The region enclosed by the acoustic boundaries with an area smaller than the average area is taken as the suspected anechoic groove region. The degree of similarity of the acoustic boundary to each of the suspected anechoic groove regions is used as the first confidence level of each of the suspected anechoic groove regions. Based on the distribution regularity of each suspected anechoic groove region and its adjacent anechoic groove regions, a second confidence level is obtained for each suspected anechoic groove region. The confidence level of the anechoic groove in each suspected anechoic groove region is obtained based on the first confidence level and the second confidence level, and both the first confidence level and the second confidence level are positively correlated with the confidence level of the anechoic groove; all suspected anechoic groove regions whose confidence level of the anechoic groove is greater than a preset threshold are taken as anechoic groove regions.

[0007] Optionally, the method for obtaining the class moment degree includes: Starting from any boundary point on each acoustic boundary, obtain the 8-chain code corresponding to the acoustic boundary; in the 8-chain code corresponding to each acoustic boundary, sort the occurrence frequency of each chain code value in descending order, take the first two chain code values ​​as the first group of chain code values, and take the third and fourth chain code values ​​as the second group of chain code values. On the acoustic boundary, based on the difference between the chain code values ​​of each group of chain code values ​​and the numerical difference of a preset positive integer, as well as the difference in the number of boundary points corresponding to the chain code values, a confidence coefficient is obtained for each group of chain code values ​​corresponding to the chain code values ​​of a set of opposite sides of a rectangle; the numerical difference and the number difference are both negatively correlated with the confidence coefficient. Multiply the confidence coefficients of the two sets of chain code values ​​to obtain the class moment degree corresponding to the acoustic boundary.

[0008] Optionally, the method for obtaining the second confidence level includes: Take any of the suspected silencing groove areas as the target area; within the neighborhood of the target area, take the two suspected silencing groove areas on the left and right as the horizontal reference areas, and take the two suspected silencing groove areas on the top and bottom as the vertical reference areas; The first distance and the second distance between the centroids of the two horizontal reference regions and the target region are obtained respectively. The difference between the first distance and the second distance is negatively correlated and mapped to obtain the horizontal uniform distribution parameters. The third and fourth distances between the centroids of the two vertical reference regions and the target region are obtained respectively. The differences between the third and fourth distances are negatively correlated and mapped to obtain the vertical uniform distribution parameters. The second confidence level of each target region is obtained by combining the horizontal uniform distribution parameters and the vertical uniform distribution parameters.

[0009] Optionally, the method for obtaining the acoustic deformation coefficient includes: Based on the difference in sound wave energy distribution on both sides of the central axis of each anechoic groove region, a first acoustic deformation parameter is obtained for each anechoic groove region; the central axis is parallel to the edge of the anechoic groove, and the distribution difference is positively correlated with the first acoustic deformation parameter; Based on the attenuation difference of the sound wave energy on both sides of each anechoic groove edge, the acoustic deformation parameter of each anechoic groove edge is obtained; based on the difference of the acoustic deformation parameters of the two anechoic groove edges in the anechoic groove region, the second acoustic deformation parameter of each anechoic groove region is obtained. Based on the first acoustic deformation parameter and the second acoustic deformation parameter, acoustic sub-parameters of each of the silencing groove regions are obtained, and both the first acoustic deformation parameter and the second acoustic deformation parameter are positively correlated with the acoustic sub-parameters; By combining the acoustic sub-parameters of all the aforementioned silencing groove regions, the acoustic deformation coefficient of the pipe surface is obtained.

[0010] Optionally, the method for obtaining the acoustic deformation parameters includes: In the acoustic signal, all outermost acoustic points and all innermost acoustic points on the edge of each silencing groove are obtained respectively. All acoustic points adjacent to the outermost acoustic points are taken as external acoustic points of the region, and all acoustic points adjacent to the innermost acoustic points are taken as internal acoustic points of the region. Neither the external acoustic points nor the internal acoustic points are on the edge of the silencing groove. The acoustic deformation parameters of each edge of the anechoic groove are obtained based on the energy difference between the total energy of the acoustic points outside the region of each edge of the anechoic groove and the total energy of the acoustic points inside the region.

[0011] Optionally, the method for obtaining the coefficient of thermal expansion deformation includes: In each of the silencing groove regions, the thermal expansion torsion parameters of each silencing groove region are obtained based on the distance between all the innermost acoustic points on the edges of the two silencing grooves. The radial distance of each of the silencing groove regions is negatively correlated and then used as a weight to weight the corresponding thermal expansion torsion parameter, thereby obtaining the thermal expansion deformation sub-parameter of each of the silencing groove regions. By combining the thermal expansion deformation sub-parameters of all the aforementioned silencing groove regions, the thermal expansion deformation coefficient of the pipe surface is obtained.

[0012] Optionally, the method for obtaining the thermal expansion distortion parameter includes: In each of the silencing groove regions, all the innermost acoustic points on the edge of each silencing groove are sequentially sorted. Based on the distance between the innermost acoustic points with the same sorting number on the edges of two silencing grooves, the average width and width range of the corresponding silencing groove are obtained. The relative flatness of each silencing groove region is obtained based on the average width of the groove corresponding to each silencing groove region and the degree of deviation of the average width of the grooves corresponding to all silencing groove regions; the width range is used as the torsion coefficient of the silencing groove region. The thermal expansion torsion parameters of each of the silencing groove regions are obtained based on the relative flatness and the torsion coefficient, and both the relative flatness and the torsion coefficient are positively correlated with the thermal expansion torsion parameters.

[0013] Optionally, the method for obtaining the comprehensive deformation coefficient includes: The product of the acoustic deformation coefficient and the thermal expansion deformation coefficient is taken as the comprehensive deformation coefficient of the pipe surface.

[0014] Optionally, the method for obtaining the defect location result includes: When the comprehensive deformation coefficient is greater than the preset threshold, the defect location result is determined to be that there is a defect; otherwise, the defect location result is determined to be that there is no defect. When it is determined that there is a defect, the specific location of the defect is determined according to the maximum value of the comprehensive deformation coefficient, and a thermal map of the wall thickness of the high-temperature irregular pipe is generated.

[0015] This invention provides a phased array noise reduction and visualization detection sensing method for high-temperature pipelines. By fusing acoustic signals and infrared thermal imaging data, and fully utilizing the structural features of the integrated anechoic groove and damping block wedge in high-temperature detection, the method achieves coordinated perception of the acoustic propagation environment and thermal deformation state of high-temperature pipelines, significantly improving the accuracy and reliability of detection under complex operating conditions. Specifically, by identifying the anechoic groove region and its edge features in the acoustic signal, and combining the analysis of acoustic energy distribution uniformity and edge energy differences, the acoustic deformation coefficient reflecting the degree of ultrasonic propagation distortion is effectively extracted, realizing dynamic evaluation of the acoustic field stability of the detection system itself and visual characterization of noise interference. Simultaneously, by utilizing the peak spacing and radial position relationship of acoustic energy at the edge of the anechoic groove, the local thermal expansion deformation of the pipeline is inverted to obtain the thermal expansion deformation coefficient, truly reflecting the geometric deformation trend of the pipeline at high temperatures. Furthermore, by fusing the acoustic and thermal expansion deformation coefficients, a comprehensive deformation coefficient is obtained, providing dual compensation for wall thickness measurement and defect location, significantly reducing imaging distortion and location errors caused by high-temperature thermal drift and structural irregularities. Finally, a thermal map of the wall thickness of high-temperature irregular-shaped pipes is generated based on the comprehensive deformation coefficient, and intelligent defect location is achieved, giving the detection results good visualization, quantification, and interpretability. This method requires no additional calibration equipment, utilizes existing wedge structure features as the analysis benchmark, and has the advantages of simple implementation, fast response, and strong adaptability. It is particularly suitable for online, high-precision non-destructive testing and health monitoring of high-temperature, complex-structured pipes, significantly improving the early identification capability of potential safety hazards in industrial sites. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an embodiment of the high-temperature pipeline phased array noise reduction and visualization detection sensing method of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] Unless otherwise specifically stated, the technical or scientific terms used in the embodiments of this invention should be understood in their ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains. The terms "comprising" or "including," as used in the embodiments of this invention, do not limit the shapes, numbers, steps, actions, operations, components, elements, and / or groups thereof mentioned, nor do they exclude the appearance or addition of one or more other different shapes, numbers, steps, actions, operations, components, elements, and / or groups thereof, or the inclusion of these.

[0020] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale, and techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail; however, where appropriate, the illustrated techniques, methods, and apparatus should be considered part of the specification. In all the examples shown and discussed herein, any other specific example may have different values. It should be noted that similar symbols and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0021] In the description of the embodiments of the present invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In the embodiments of the present invention, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in the embodiments of the present invention, as well as the features of different embodiments or examples.

[0022] Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0023] Reference Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the high-temperature pipeline phased array noise reduction visualization detection sensing method of the present invention, which presents an embodiment of the high-temperature pipeline phased array noise reduction visualization detection sensing method of the present invention.

[0024] In one embodiment, the high-temperature pipeline phased array noise reduction and visualization detection sensing method includes: Step S100: Acquire acoustic wave signals and infrared thermal imaging data of the surface of the high-temperature pipeline, acquire all silencing groove regions in the acoustic wave signals and the two edges of the silencing grooves in each silencing groove region, and acquire the radial distance of each silencing groove region from the central axis of the pipeline.

[0025] The acoustic signal from the high-temperature pipe surface can be an ultrasonic echo signal received by a phased array ultrasonic probe coupled to the pipe surface via an integrated wedge. This echo signal characterizes the internal structural state and sound field propagation characteristics of the pipe, serving as the fundamental data source for extracting the acoustic deformation coefficient. In this embodiment, the acoustic signal from the high-temperature pipe surface can be emitted by a phased array transducer, focused, and transmitted to the pipe surface via an integrated wedge consisting of an anechoic groove and a damping block. Reflected or transmitted echoes are received to form time-domain or frequency-domain signals. Infrared thermal imaging data can be temperature distribution images of the high-temperature pipe surface acquired by an infrared thermal imager. This helps to understand the trend of thermal deformation distribution and is fused with the acoustic signal to achieve thermo-acoustic collaborative sensing. For example, the infrared thermal imaging data can be based on the correspondence between the object's thermal radiation intensity and temperature, capturing the pipe surface thermal field information through an infrared detector array. The anechoic groove region can be the response area of ​​a specific geometric groove structure in the integrated wedge used to suppress lateral vibration and clutter interference in the acoustic signal. This region serves as an intrinsic reference benchmark for sound field stability and is used to assess the degree of ultrasonic propagation distortion. Furthermore, the anechoic groove region can be located by identifying continuous intervals in the acoustic signal where energy attenuates significantly or has specific time delay characteristics.

[0026] The edges of the anechoic groove can be the two boundary locations where the anechoic groove region exhibits abrupt energy changes in the acoustic signal, providing a benchmark for acoustic energy distribution comparison and used to calculate the energy difference and peak spacing on both sides of the edge. In this embodiment, the edges of the anechoic groove can be identified by detecting the maximum energy gradient point in the acoustic signal or by using an edge detection algorithm. Furthermore, the edges of the anechoic groove can be used in conjunction with the acoustic energy peak points for inverting thermal expansion deformation; together with the anechoic groove region, they constitute an acoustic deformation analysis unit. The pipe central axis can be the symmetrical centerline of the high-temperature pipe geometry, serving as a reference line for radial distance calculation, used to establish the relationship between local position and overall geometry. The radial distance can be the perpendicular distance from the center point of the anechoic groove region to the pipe central axis, relating the local thermal expansion deformation to the overall geometric state of the pipe, supporting the modeling of the thermal expansion deformation coefficient. For example, the radial distance can be calculated by measuring the pipe outer diameter combined with the geometric relationship of the wedge installation position, or by inverting the focusing depth of the phased array acoustic beam.

[0027] Acquiring acoustic signals and infrared thermal imaging data from the surface of a high-temperature pipeline can be achieved by simultaneously acquiring phased array ultrasonic echo signals and temperature field images output by an infrared thermal imager. Furthermore, this operation can be implemented through a hardware-triggered synchronization mechanism to ensure time alignment of acoustic and infrared data, or through software timestamp matching of asynchronously acquired acoustic and infrared data streams, thereby establishing a foundation for thermo-acoustic multimodal data and supporting subsequent collaborative sensing analysis. Acquiring all anechoic groove regions and the two edges of each anechoic groove region in the acoustic signal can be achieved by identifying intervals with characteristic attenuation modes in the acoustic signal and extracting their start and end boundaries. Furthermore, this operation can be achieved by locating the preset anechoic groove response waveform in the time-domain signal based on template matching, or by using wavelet transform to detect energy abrupt change points to define edge positions, thereby establishing a structured acoustic reference unit for deformation analysis. Obtaining the radial distance of each anechoic groove region from the pipeline's central axis can be achieved by combining pipeline geometric parameters and wedge installation positions to calculate the radial coordinates corresponding to each anechoic groove region. Furthermore, this operation can be achieved by calculating the radial distance using the known pipe outer diameter and wedge contact angle trigonometric function, or by using phased array focusing depth information to infer the radial position, thereby establishing a spatial relationship between local measurement points and global geometry.

[0028] Step S200: Based on the uniformity of sound wave energy distribution in all silencing groove areas and the difference in sound wave energy between adjacent sides of each silencing groove edge in each silencing groove area, obtain the acoustic deformation coefficient of the pipe surface.

[0029] The uniformity of sound energy distribution can be an indicator of the spatial or temporal dispersion of sound energy within the anechoic groove region, reflecting the stability of the sound field after thermal disturbance, and is a core input parameter of the acoustic deformation coefficient. In this embodiment, the uniformity of sound energy distribution can be quantified by calculating the standard deviation, coefficient of variation, or entropy value of the sound pressure amplitude within the region. The sound energy difference can be the relative deviation of sound energy in adjacent areas on both sides of the anechoic groove edge, characterizing the asymmetry of the sound beam deflection caused by thermal deformation, and enhancing sensitivity to noise interference. For example, the sound energy difference can be calculated by taking the difference or ratio of the average energy values ​​within a fixed window on the left and right sides of the edge. The acoustic deformation coefficient can be a dimensionless parameter that quantifies the degree of distortion of the ultrasonic propagation path due to the thermo-acoustic coupling effect, dynamically evaluating the sound field stability of the detection system, and realizing a visual characterization of noise interference. Furthermore, the acoustic deformation coefficient can be calculated by integrating the uniformity of sound energy distribution and the energy difference on both sides of the edge using a normalized weighted function.

[0030] Based on the uniformity of acoustic energy distribution across all anechoic groove regions and the difference in acoustic energy between adjacent sides of each anechoic groove edge within each region, the acoustic deformation coefficient of the pipe surface is obtained. This can be achieved by quantifying the energy dispersion and edge asymmetry within the region and fusing them to generate the acoustic deformation coefficient. Furthermore, this operation can use the coefficient of variation to characterize uniformity and combine it with the edge energy ratio to construct a composite index, or use principal component analysis to extract the energy distribution feature vector and map it to the deformation coefficient space. This allows for dynamic assessment of the degree of sound field distortion and visualization of noise interference.

[0031] Step S300: Obtain the thermal expansion deformation coefficient of the pipe surface based on the distance between the peak points of sound wave energy on the edges of the two silencing grooves in each silencing groove region and the radial distance of each silencing groove region.

[0032] The peak points of acoustic energy can be sampling points or spatial locations corresponding to the local maximum acoustic energy at the edge of the anechoic groove, and the variation in their spacing directly reflects the geometric displacement caused by local material expansion. In this embodiment, the peak points of acoustic energy can be identified within the edge neighborhood using a local maximum detection algorithm. The thermal expansion deformation coefficient can be a physical parameter characterizing the degree of geometric deformation caused by local thermal expansion of the pipeline, truly reflecting the non-uniform expansion trend of the pipeline at high temperatures, and providing a geometric compensation basis for wall thickness measurement. Furthermore, the thermal expansion deformation coefficient can be obtained by inversion based on the functional relationship between the spacing of the peak points of acoustic energy at the edge of the anechoic groove and the corresponding radial distance.

[0033] Based on the distance between the peak sound wave energy points on the edges of the two silencing grooves in each silencing groove region, and the radial distance of each silencing groove region, the thermal expansion deformation coefficient of the pipe surface can be obtained. This can be achieved by establishing a mapping relationship between the peak spacing and the radial distance, and inverting the local thermal expansion. Furthermore, this operation can be used to construct an empirical formula to convert the spacing change rate into expansion strain, or to train a regression model to predict the local expansion coefficient from the spacing-radial pair. This can accurately reflect the geometric deformation trend of the pipe at high temperatures, providing a basis for compensation.

[0034] Step S400: Obtain the comprehensive deformation coefficient of the pipe surface based on the acoustic deformation coefficient and the thermal expansion deformation coefficient. Based on the comprehensive deformation coefficient of the pipe surface, obtain the thermal map of the high-temperature irregular pipe wall thickness and the defect location results.

[0035] The comprehensive deformation coefficient can be a joint compensation parameter formed by fusing the acoustic deformation coefficient and the thermal expansion deformation coefficient, providing dual physical compensation for wall thickness measurement and defect location, reducing errors caused by thermal drift and structural irregularities. In this embodiment, the comprehensive deformation coefficient can integrate the two types of deformation information through weighted fusion, tensor superposition, or machine learning mapping. Furthermore, the comprehensive deformation coefficient can directly drive the generation of thermal maps of the wall thickness of high-temperature irregular pipes and the correction of defect location results. The thermal map of the wall thickness of high-temperature irregular pipes can be a color-mapped image of the wall thickness values ​​distributed along the surface of the high-temperature irregular pipe, realizing the visualization and quantitative expression of the wall thickness state, facilitating rapid identification of thinning areas. For example, the thermal map of the wall thickness of high-temperature irregular pipes can be generated based on the original data of phased array ultrasonic thickness measurement, combined with the comprehensive deformation coefficient for spatial correction. The defect location result can be the coordinate position and morphological description of defects inside or on the surface of the pipe in three-dimensional space, providing high-precision defect spatial information and supporting intelligent diagnosis and risk assessment. In this embodiment, after the defect location result is initially located by the phased array imaging algorithm, the sound beam path and propagation time are corrected using the comprehensive deformation coefficient.

[0036] Based on the acoustic deformation coefficient and the thermal expansion deformation coefficient, the comprehensive deformation coefficient of the pipe surface is obtained. This can be achieved by fusing the two types of deformation coefficients to form a unified compensation parameter. Furthermore, this operation can employ a weighted average method, with the weights dynamically adjusted by the temperature field gradient, or by constructing a neural network to map the two types of coefficients into a comprehensive deformation output, thus providing a dual correction basis that takes into account both acoustic field distortion and geometric deformation. Based on the comprehensive deformation coefficient of the pipe surface, the wall thickness thermogram and defect location results of high-temperature irregular pipes are obtained. This can be achieved by applying the comprehensive deformation coefficient to the original ultrasonic thickness measurement and imaging data for spatial correction. Further, this operation can be achieved by embedding the comprehensive deformation coefficient to correct the acoustic path during total focusing imaging (TFM) reconstruction, or by performing pixel-level geometric correction on the C-scan wall thickness map to generate a thermogram, thereby generating a high-precision, interpretable wall thickness distribution map and defect location information.

[0037] Taking the online monitoring of the main steam pipeline of a nuclear power plant as an example, the high-temperature pipeline phased array noise reduction visualization detection sensing method in this embodiment can deploy a phased array probe and an infrared thermal imager in the high-temperature main steam bend section of the nuclear power plant to simultaneously acquire acoustic signals and thermal imaging data; the system automatically identifies multiple silencing groove regions and their edges in the acoustic signal, calculates the uniformity of acoustic energy in each region and the difference in edge energy, and generates an acoustic deformation coefficient; at the same time, it measures the peak spacing and corresponding radial distance of the edge of the silencing groove to invert the local thermal expansion deformation coefficient; after the two are integrated into a comprehensive deformation coefficient, it is used to correct the wall thickness measurement value and crack location coordinates of the bend section, and finally outputs a color-coded wall thickness thermal map and a three-dimensional defect location mark, which can achieve high-precision health assessment without stopping the machine for calibration.

[0038] In one embodiment, the method for obtaining the silencing groove region includes: Obtain all acoustic boundaries in the sound wave signal, and obtain the average area of ​​the region enclosed by all acoustic boundaries. The region enclosed by acoustic boundaries with an area smaller than the average area is taken as the suspected anechoic groove region. In this context, all acoustic boundaries in the acoustic signal can be continuous contour lines that define closed regions due to significant changes in energy or gradient. These contours can be used to enclose potential anechoic chamber response regions, serving as the basic units for subsequent screening and confidence assessment. In an exemplary embodiment, all acoustic boundaries in the acoustic signal can be connected to form closed contours using edge detection algorithms (such as Canny and Sobel) or energy abrupt change points. Furthermore, obtaining all acoustic boundaries in the acoustic signal can be achieved by image processing of the acoustic signal followed by the application of contour extraction algorithms to identify closed boundaries. For example, this operation can be achieved by converting the time-frequency domain acoustic signal to a grayscale image and then using a contour extraction function to extract boundaries, or by detecting energy abrupt change points in the original A-scan signal using a sliding window and connecting them into closed paths, thereby establishing a set of geometric units suitable for region analysis. The average area of ​​the region enclosed by the acoustic boundaries can be the statistical average of the areas of the closed regions enclosed by all acoustic boundaries, which can be used as an area screening threshold to distinguish typical anechoic chamber response regions from large clutter or noise clusters. In one specific embodiment, the average area of ​​the region enclosed by the acoustic boundaries can be obtained by calculating the arithmetic mean of the number of pixels or sampling points inside each closed acoustic boundary. Furthermore, obtaining the average area of ​​the region enclosed by all acoustic boundaries can be achieved by calculating the average of the number of pixels or sampling points contained within each closed boundary. For example, this operation can directly count the number of foreground points within the binary mask using pixel counting, or calculate the area of ​​the closed region by integrating the boundary coordinates using Green's formula, thereby providing a statistical benchmark for area filtering.

[0039] A suspected anechoic groove region can be a candidate region enclosed by acoustic boundaries with an area smaller than the average area. This can be used to narrow down the anechoic groove identification range and exclude large-area non-target interference regions. In an exemplary embodiment, the suspected anechoic groove region can include, but is not limited to, one or more of isolated small-area regions, densely clustered regions, and edge-attached regions. Further, using regions enclosed by acoustic boundaries with an area smaller than the average area as suspected anechoic groove regions can be achieved by comparing the area of ​​each region with the average and retaining the smaller one as a candidate. For example, this operation can set an upper limit of the area to 0.8 times the average value to enhance conservatism, or combine area and energy intensity dual thresholds for joint screening, thereby initially filtering large non-target regions and focusing on the typical anechoic groove scale range.

[0040] The degree of similarity of the acoustic boundary corresponding to each suspected anechoic groove region is used as the first confidence level of each suspected anechoic groove region. The similarity of the acoustic boundary corresponding to the suspected anechoic groove region can be a measure of the similarity between the shape of the acoustic boundary and the geometry of an ideal rectangle. This can be used as the first confidence level to reflect whether the region conforms to the inherent regular geometric characteristics of an anechoic groove. In one specific embodiment, the similarity of the acoustic boundary corresponding to the suspected anechoic groove region can be quantified by calculating the aspect ratio, number of corner points, Hausdorff distance, or rectangle fitting residual of the boundary profile. Further, using the similarity of the acoustic boundary corresponding to each suspected anechoic groove region as the first confidence level for each region can quantify the geometric similarity between the boundary and the rectangle, and map it to a score within the [0, 1] interval. For example, this operation can use the overlap rate of the minimum bounding rectangle as the similarity index, or calculate the similarity between the profile and the spectrum of a standard rectangle using Fourier descriptors, thereby introducing shape priors and improving the accuracy of single-region discrimination. The first confidence level can be a reliability score assigned to the suspected anechoic groove region based on the similarity of the acoustic boundary, which can be used to characterize the geometric rationality of a single region and for preliminary confidence ranking.

[0041] Based on the distribution pattern of each suspected anechoic chamber area and its neighboring anechoic chamber areas, the second confidence level of each suspected anechoic chamber area is obtained. The distribution regularity of adjacent anechoic slot regions can be the orderliness of the suspected anechoic slot region and its spatially adjacent regions in terms of position, spacing, or arrangement direction. This can be used to reflect the inherent periodic arrangement characteristics of anechoic slots in the integrated wedge block for group behavior verification. In an exemplary embodiment, the distribution regularity of adjacent anechoic slot regions can be obtained by calculating the standard deviation of the center-to-center distance between adjacent regions, angular consistency, or periodic spectral analysis. Further, based on the distribution regularity of each suspected anechoic slot region and its adjacent anechoic slot regions, a second confidence level is obtained for each suspected anechoic slot region. This can be achieved by analyzing the spatial arrangement pattern of multiple suspected anechoic slots within a local area to assess its orderliness. For example, this operation can calculate the standard deviation of the center-to-center distance between adjacent regions; the smaller the deviation, the higher the regularity. Alternatively, RANSAC fitting of straight lines or arcs can be used to assess the degree of collinearity / circularity of the point set, thereby enhancing the robustness of group identification using structural context information. The second confidence level can be a contextual consistency score assigned to the suspected anechoic slot region based on the distribution regularity, which can be used to measure whether the region conforms to the prior knowledge of the overall structural arrangement and suppress isolated artifacts.

[0042] The confidence level of the anechoic groove in each suspected anechoic groove region is obtained based on the first confidence level and the second confidence level. Both the first confidence level and the second confidence level are positively correlated with the confidence level of the anechoic groove. All suspected anechoic groove regions with a confidence level of the anechoic groove greater than a preset threshold are taken as anechoic groove regions.

[0043] The confidence score of the anechoic groove can be a comprehensive credibility index formed by fusing a first confidence score and a second confidence score. This index can be used as the final criterion for determining whether a suspected area is confirmed as a valid anechoic groove area. In one specific embodiment, the anechoic groove confidence score can be fused using methods such as weighted summation, logistic regression, or fuzzy inference. Furthermore, the anechoic groove confidence score is constructed based on both the first and second confidence scores; its output directly determines the final set of anechoic groove areas. The preset threshold can be a fixed or adaptive numerical limit used to determine whether the anechoic groove confidence score meets the standard. This threshold can be used to implement binary decision-making and filter out highly reliable anechoic groove areas. Furthermore, obtaining the anechoic groove confidence score for each suspected anechoic groove area based on the first and second confidence scores can be achieved by fusing the two confidence scores to generate a unified score. For example, this operation can employ a weighted average: w1 × first confidence level + w2 × second confidence level, with the weights dynamically adjusted by the temperature field. Alternatively, a lightweight classifier (such as SVM) can be trained to predict the comprehensive confidence level using the two confidence levels as input, thereby achieving multi-dimensional evidence fusion and improving recognition reliability. All suspected anechoic groove regions with a confidence level greater than a preset threshold can be considered as anechoic groove regions. This can be achieved by thresholding the comprehensive confidence level and outputting the final set of valid regions. For example, this operation can employ an adaptive threshold, dynamically set based on the quantiles of the confidence level distribution, or introduce a hysteresis threshold to avoid boundary fluctuations, thereby achieving highly robust automatic identification of anechoic grooves and providing a reliable benchmark for subsequent deformation coefficient calculation.

[0044] For example, in the scenario of inspecting the outlet bend of a high-temperature cracking furnace in a petrochemical plant, the high-temperature pipeline phased array noise reduction visualization detection sensing method of this embodiment can be as follows: deploy a phased array probe on a bend of different diameters operating at 600℃, collect acoustic signals, and automatically extract all acoustic boundaries; the system calculates the average area of ​​the region enclosed by each boundary to 500 pixels², and filters out 32 suspected anechoic groove regions with an area smaller than this value; assign a high first confidence level to each region whose boundary overlaps with the rectangle by more than 0.7; at the same time, analyze the distribution of these regions on the wedge contact surface, and find that 24 of them are arranged linearly with equal spacing, and assign a high second confidence level; after fusion, the anechoic groove confidence level is obtained, a threshold of 0.65 is set, and finally 22 effective anechoic groove regions are confirmed; these regions are used for subsequent acoustic and thermal expansion deformation coefficient calculations to support the generation of high-precision wall thickness thermal maps.

[0045] In one embodiment, the method for obtaining the class moment degree includes: Starting from any boundary point on each acoustic boundary, obtain the corresponding 8-chain code of the acoustic boundary; in the 8-chain code corresponding to each acoustic boundary, sort the occurrence frequency of each chain code value in descending order, take the first two chain code values ​​as the first group of chain code values, and take the third and fourth chain code values ​​as the second group of chain code values. On the acoustic boundary, based on the difference between the chain code values ​​of each group of chain code values ​​and the numerical difference of a preset positive integer, as well as the difference in the number of boundary points corresponding to the chain code values, a confidence coefficient is obtained for each group of chain code values ​​corresponding to the chain code values ​​of a set of opposite sides of a rectangle; both the numerical difference and the number difference are negatively correlated with the confidence coefficient. Multiply the confidence coefficients of the two sets of chain code values ​​to obtain the class moment degree of the corresponding acoustic boundary.

[0046] In this embodiment, any boundary point on the acoustic boundary can be any starting position among the discrete sampling points constituting the acoustic boundary. This can be used as the starting point for generating the 8-chain code sequence, ensuring the continuity of the boundary orientation description. The 8-chain code corresponding to the acoustic boundary can be a digital sequence encoded along the acoustic boundary in eight neighborhood directions. Each code value (0–7) represents the direction of movement from the current point to the next boundary point, which can be used to transform the geometric contour into a statistically analyzable direction sequence to quantify the boundary orientation features. In this embodiment, the 8-chain code corresponding to the acoustic boundary can be obtained by traversing the boundary clockwise or counterclockwise from a selected starting point, recording the 8-directional displacement encoding relative to the previous point at each step. The frequency of occurrence of each chain code value can be a statistical count of the number of times each direction code value (0–7) appears in the 8-chain code sequence. This can be used to reflect the main orientation distribution of the boundary and is the basis for identifying the opposite side orientation of the rectangle. In an exemplary embodiment, the frequency of occurrence of each chain code value can be obtained by traversing the 8-chain code sequence and counting the values ​​from 0 to 7. The top two chain code values ​​can be the two most frequent 8-chain code direction values, which can be used as the first set of candidate opposite side directions, corresponding to a set of parallel sides of the rectangle. The third and fourth most frequent chain code values ​​can be the third and fourth most frequent 8-chain code direction values, which can be used as the second set of candidate opposite side directions, corresponding to another set of parallel sides of the rectangle.

[0047] The first set of chain code values ​​can be a set of the top two chain code values, used to characterize the dominant direction of a group of opposite sides of the rectangle. This set can be used in confidence coefficient calculations to assess whether the group constitutes a valid pair of opposite sides. The second set of chain code values ​​can be a set of the third and fourth chain code values, used to characterize the dominant direction of another group of opposite sides of the rectangle. This set can be used in conjunction with the first set to verify the overall rectangular structure's rationality. The chain code value difference can be the modulo-8 difference between two chain code values ​​within the same group, used to measure whether the direction of this group is close to antiparallelism (ideally, opposite sides of a rectangle should differ by 4). Furthermore, the chain code value difference, together with a preset positive integer difference, can jointly determine the rationality of the direction. The preset positive integer difference can be the absolute deviation between the chain code value difference and the ideal opposite side difference (usually 4), used to quantify the degree of deviation between the actual direction and the theoretical antiparallel direction. In a specific embodiment, the preset positive integer difference can include, but is not limited to, horizontal opposite side deviation, vertical opposite side deviation, and diagonal opposite side deviation.

[0048] The difference in the number of boundary points corresponding to chain code values ​​can be the absolute difference in the number of boundary points corresponding to two chain code values ​​in the same set. This can be used to reflect the balance of opposite side lengths of a rectangle; the smaller the difference, the more it conforms to the characteristics of a rectangle. The confidence coefficient of a set of chain code values ​​corresponding to opposite sides of a rectangle can be a credibility score measuring whether a set of chain code values ​​constitutes a valid pair of opposite sides of a rectangle. It can be used to quantify the geometric rationality of a single set of opposite sides, providing a component for the degree of similarity. In this embodiment, the confidence coefficient of a set of chain code values ​​corresponding to opposite sides of a rectangle can be calculated using a negative correlation function (such as exponential decay or reciprocal) based on the deviation of the chain code value difference from a preset positive integer (such as 4) and the difference in the number of corresponding boundary points. The confidence coefficients of two sets of chain code values ​​can be the set of the confidence coefficients of the first and second sets, which can be used to characterize the reliability of the two sets of opposite sides respectively, jointly determining the overall rectangle similarity.

[0049] Starting from any boundary point on each acoustic boundary, the corresponding 8-chain code is obtained. This can be achieved by traversing the acoustic boundary in a fixed direction and recording the 8-neighborhood movement direction at each step to form a coding sequence. Furthermore, this operation can be implemented by extracting the 8-chain code from the binary mask boundary using the Freeman chain code algorithm, or by directly tracing pixel-level boundaries to generate chain codes after edge refinement of the original acoustic image. This transforms continuous geometric contours into discrete direction sequences, facilitating statistical analysis. Within the 8-chain codes corresponding to each acoustic boundary, the frequency of each chain code value is sorted in descending order. This can be done by counting the occurrences of each code value from 0 to 7 and arranging them from highest to lowest frequency. For example, this operation can be achieved by using a hash table for fast counting and then calling a sorting function, or by constructing a frequency histogram and outputting it in descending index order. This allows identification of the dominant boundary direction, providing a basis for grouping. The top two chain code values ​​are used as the first group of chain code values; this can be achieved by taking the two most frequent and second most frequent chain code values ​​to form the first group. In one specific embodiment, this operation can be achieved by sorting the code values ​​by the two largest if the frequencies are the same, or by combining spatial continuity verification to ensure that the two directions do indeed alternate, thereby initially determining the directions that may constitute a set of opposite sides of a rectangle.

[0050] The third and fourth most frequent chaincode values ​​are used as the second set of chaincode values. Alternatively, the chaincode values ​​with the third and fourth highest frequencies can be used to form the second set. Furthermore, this operation can be achieved by repeating the third as the fourth if there are only three significant directions, or by excluding candidates whose angle with the first set of directions is less than 30° to avoid redundancy. This can determine another set of potential opposite side directions and form a complete rectangle hypothesis. Based on the difference between the chaincode values ​​of each set of chaincode values ​​and the numerical difference of the preset positive integer, as well as the difference in the number of boundary points corresponding to the chaincode values, the confidence coefficient of each set of chaincode values ​​is obtained. This can be achieved by calculating the deviation between the direction difference and the ideal value (4), and the difference in the number of points corresponding to the two directions, and substituting them into the negative correlation function to generate the confidence coefficient. In an exemplary embodiment, this operation can be achieved by using a confidence coefficient equal to an exponential function (the parameter is the weighted sum of the direction deviation and the number difference), or by using a piecewise linear function (the confidence is 1 when the deviation is less than the threshold, otherwise it decreases linearly). This can quantify whether a single set of opposite sides satisfies the rectangular geometric constraints. Multiplying the confidence coefficients of the two sets of chain code values ​​yields the class moment degree of the corresponding acoustic boundary. This can be achieved by multiplying the confidence coefficients of the first and second sets to obtain the final class moment degree. For example, this operation can be implemented by using a geometric mean instead of the product to mitigate the influence of extreme values, or by introducing a minimum confidence threshold (if either set is below the threshold, the class moment degree is zero). This allows for joint verification of the two sets of opposite edges, ensuring the consistency of the overall rectangular structure.

[0051] Taking the online detection of high-temperature steam main pipeline as an example, the high-temperature pipeline phased array noise reduction visualization detection sensing method in this embodiment can collect sound wave signals in a straight pipe section operating at 550℃ and extract the acoustic boundary of a suspected anechoic groove area. The system generates an 8-chain code sequence from the boundary starting point. The chain code 2 (upward) appears 120 times, 6 (downward) 118 times, 0 (rightward) 95 times, and 4 (leftward) 93 times. The other directions all appear less than 10 times. The first two {2, 6} are the first group, and the third and fourth {0, 4} are the second group. The chain code difference of the first group is 4, which is consistent with the preset value. The point difference is 2, and the confidence coefficient is 0.98. The chain code difference of the second group is also 4, the point difference is 4, and the confidence coefficient is 0.96. The class moment degree = 0.98 × 0.96 = 0.941, which is much higher than the threshold of 0.7. It is confirmed as a high-reliability anechoic groove area and used for subsequent deformation coefficient calculation.

[0052] In one embodiment, the method for obtaining the second confidence level includes: Take any suspected silencing groove area as the target area; within the neighborhood of the target area, take the two suspected silencing groove areas on the left and right as the horizontal reference areas, and take the two suspected silencing groove areas on the top and bottom as the vertical reference areas. The first and second distances between the centroids of the two horizontal reference regions and the target region are obtained respectively. The differences between the first and second distances are negatively correlated and mapped to obtain the horizontal uniform distribution parameters. The third and fourth distances between the centroids of the two vertical reference regions and the target region are obtained respectively. The differences between the third and fourth distances are negatively correlated and mapped to obtain the vertical uniform distribution parameters. The second confidence level of each target region is obtained by combining the horizontal uniform distribution parameters and the vertical uniform distribution parameters.

[0053] The target region can be a currently evaluated suspected anechoic trough region, serving as the central unit for calculating the second confidence level. The target region can be used as a benchmark for spatial distribution regularity analysis, establishing geometric relationships with neighboring regions. In an exemplary embodiment, the target regions can be sorted by area from largest to smallest and processed sequentially, or all regions can be processed in parallel to improve computational efficiency. The neighborhood can be a local spatial range defined around the target region, used to filter adjacent suspected anechoic trough regions participating in the regularity assessment. The neighborhood can be used to limit the selection range of the reference region, ensuring that the selected regions have structural correlation in physical arrangement. Further, the neighborhood can be a search window defined with the centroid of the target region as the center, according to a preset radius or grid step size. The two suspected anechoic trough regions on the left and right can be the two nearest suspected anechoic trough regions located on either side of the target region in the horizontal direction and within the neighborhood. The two suspected anechoic trough regions on the left and right can be used as reference objects for horizontal distribution uniformity, used to calculate the first distance and the second distance.

[0054] The two suspected anechoic groove regions, one above the other, can be the two nearest suspected anechoic groove regions located vertically on either side of the target area and within their neighborhood. These two suspected anechoic groove regions can be used as reference objects for vertical uniformity calculation, specifically for determining the third and fourth distances. The horizontal reference region can be a horizontal reference set jointly formed by the two suspected anechoic groove regions (left and right). The horizontal reference region can provide spatial symmetry information in the horizontal dimension, supporting the calculation of horizontal uniform distribution parameters. The vertical reference region can be a vertical reference set jointly formed by the two suspected anechoic groove regions (upper and lower). The vertical reference region can provide spatial symmetry information in the vertical dimension, supporting the calculation of vertical uniform distribution parameters.

[0055] The centroid corresponding to the target region can be the centroid coordinates of the geometric contour of the target region. The centroid can be used as a reference point for distance calculation, representing the region's position in space. In one specific embodiment, the centroid can be obtained by averaging the region's pixel coordinates or by using the image moment method. The first distance between the centroid of the horizontal reference region and the target region can be the Euclidean distance between the centroid of the left horizontal reference region and the centroid of the target region. This first distance can reflect the left-side layout offset and is used to assess horizontal symmetry. The second distance between the centroid of the horizontal reference region and the target region can be the Euclidean distance between the centroid of the right horizontal reference region and the centroid of the target region. This second distance can reflect the right-side layout offset and is compared with the first distance to determine uniformity.

[0056] The third distance between the corresponding centroids of the vertical reference area and the target area can be the Euclidean distance between the centroids of the upper vertical reference area and the target area. This third distance can reflect the upper arrangement offset and is used to assess vertical symmetry. The fourth distance between the corresponding centroids of the vertical reference area and the target area can be the Euclidean distance between the centroids of the lower vertical reference area and the target area. This fourth distance can reflect the lower arrangement offset and is used to compare with the third distance to determine uniformity. The horizontal uniform distribution parameter can be a horizontal arrangement uniformity index obtained by negatively correlated mapping based on the difference between the first and second distances. The horizontal uniform distribution parameter can be used to quantify whether the target area conforms to periodic arrangement characteristics in the horizontal direction; a higher value indicates greater uniformity. Furthermore, the horizontal uniform distribution parameter can be generated by substituting |first distance - second distance| into a monotonically decreasing function (such as exponential decay, reciprocal, or 1 - normalized difference).

[0057] The vertical uniform distribution parameter can be a vertical orientation uniformity index obtained by negatively correlated mapping of the difference between the third and fourth distances. The vertical uniform distribution parameter can be used to quantify whether the target area conforms to periodic arrangement characteristics in the vertical direction; a higher value indicates greater uniformity. For example, the vertical uniform distribution parameter can be generated by substituting |third distance - fourth distance| into the same negative correlation mapping function as the horizontal parameter. The second confidence score can be a comprehensive spatial regularity score formed by fusing the horizontal and vertical uniform distribution parameters. The second confidence score can be used to characterize the contextual consistency of the target area within the overall grid structure, thereby improving the robustness of anechoic groove identification. In a specific embodiment, the second confidence score can integrate the two directional parameters through weighted average, geometric average, or maximum-minimum combination methods. Furthermore, the second confidence score can be constructed based on both horizontal and vertical uniform distribution parameters, and its output is used for anechoic groove confidence calculation.

[0058] Taking any suspected anechoic groove region as the target region, this can be achieved by traversing all suspected anechoic groove regions and setting them sequentially as the current analysis center. Furthermore, this operation can be implemented by sorting the regions by area from largest to smallest and processing them sequentially, or by processing all regions in parallel to improve computational efficiency, thus enabling individualized regularity evaluation of each candidate region. Within the neighborhood of the target region, the two suspected anechoic groove regions on the left and right are used as horizontal reference regions. This can be achieved by searching for the two nearest suspected regions in the horizontal neighborhood of the target region and designating them as the left and right references, respectively. Further, this operation can be implemented by sorting by x-coordinate and taking the nearest left and right regions, or by symmetrically padding with zeros or skipping the evaluation in that direction if a region exists only on one side, thus establishing the reference system required for horizontal symmetry analysis.

[0059] Within the neighborhood of the target area, two suspected anechoic groove areas (upper and lower) are designated as vertical reference areas. This can be achieved by searching for the two nearest suspected areas in the vertical neighborhood of the target area and designating them as the upper and lower references, respectively. For example, this operation can be implemented by sorting by y-coordinate and selecting the two nearest adjacent areas (upper and lower), or by defining the vertical direction along the local tangential / normal direction of the pipe in curved surface scenarios such as curved pipes, thus establishing the reference system required for vertical symmetry analysis. The first and second distances between the centroids of the two horizontal reference areas and the corresponding centroids of the target area are obtained. This can be achieved by calculating the distance between the centroid of the left reference area and the centroid of the target area as the first distance, and the distance on the right as the second distance. Furthermore, this operation can be implemented by using Manhattan distance instead of Euclidean distance to reduce computational complexity, or by calculating the arc length distance in polar coordinates to adapt to the curvature of the curved pipe, thereby obtaining the spatial offsets on both sides in the horizontal direction for uniformity assessment.

[0060] By negatively correlated mapping of the differences between the first and second distances to obtain horizontally uniform distribution parameters, the absolute difference between the two distances can be calculated and mapped to a value in the interval [0, 1] using a monotonically decreasing function. In an exemplary embodiment, this operation can be performed using an exponential decay mapping in the form exp(-α·|d1-d2|), where α is the decay coefficient and |d1-d2| is the absolute difference between the two distances, thus achieving a uniformity measure where "the closer the distance, the higher the parameter". The third and fourth distances between the centroids of the two vertical reference regions and the target region are obtained respectively. This can be achieved by calculating the distance between the centroid of the upper reference region and the centroid of the target region as the third distance, and the distance below as the fourth distance. Furthermore, this operation can be implemented by defining a local vertical direction in conjunction with the pipe axis to adapt to irregular structures, or by assigning a default distance value to the case of a missing reference region, thereby obtaining the spatial offset on both sides of the vertical direction for uniformity judgment.

[0061] By negatively mapping the differences between the third and fourth distances, a vertically uniformly distributed parameter can be obtained. This can be achieved by calculating the absolute difference between the two distances and mapping them using the same negative correlation function as the horizontal parameter. For example, this operation can be achieved by sharing the mapping function parameters in the horizontal direction to ensure scale uniformity, or by dynamically adjusting the mapping sensitivity based on the temperature gradient, thereby quantifying the uniformity in the vertical direction and maintaining consistency in the evaluation dimensions. Combining the horizontal and vertical uniformly distributed parameters, a second confidence score for each target region can be obtained. This can be achieved by fusing the uniformity parameters in both directions to generate a single confidence score value. Further, this operation can be performed using an arithmetic mean: (horizontal parameter + vertical parameter) / 2, or by a weighted fusion: w_h·horizontal parameter + w_v·vertical parameter. The contribution of the two directional parameters can be adjusted using weight coefficients (w_h, w_v). The weights can be adaptively adjusted according to the pipe orientation, making it suitable for anisotropic structures, thus comprehensively reflecting the structural rationality of the target region in the two-dimensional grid.

[0062] Taking the online detection of high-temperature steam main pipeline in a nuclear power plant as an example, the high-temperature pipeline phased array noise reduction visualization detection sensing method in this embodiment can deploy a phased array probe on the main steam curved pipeline with a diameter of 800mm and an operating temperature of 550℃. The system identifies 36 suspected silencing groove regions. For the target region numbered 15, a 50mm neighborhood is defined with its centroid as the center, and two horizontal reference regions (14, 16) and two vertical reference regions (8, 22) are found. The first distance is calculated to be 24.1mm, the second distance to be 23.8mm, with a difference of 0.3mm. After mapping with exp(-0.1×0.3), the horizontal uniform distribution parameter is 0.97; the third distance is 25.2mm, the fourth distance to be 26.0mm, with a difference of 0.8mm. The vertical parameter is mapped to be 0.92; the average of the two results in a second confidence level of 0.945. This high confidence level indicates that region 15 conforms to the regular grid arrangement of the silencing groove in the wedge and can be reliably included in the subsequent deformation coefficient calculation process.

[0063] In one embodiment, the method for obtaining the acoustic deformation coefficient includes: Based on the difference in sound wave energy distribution on both sides of the central axis of each anechoic groove region, the first acoustic deformation parameter of each anechoic groove region is obtained; the central axis is parallel to the edge of the anechoic groove, and the distribution difference is positively correlated with the first acoustic deformation parameter; The central axis of the anechoic groove region can be a central reference line parallel and equidistant from the edges of the two anechoic grooves within the anechoic groove region. This line can serve as a benchmark for evaluating the symmetry of the left-right distribution of sound wave energy and is used to extract the first acoustic deformation parameter. In this embodiment, the central axis of the anechoic groove region, together with the difference in sound wave energy distribution on both sides of the central axis, forms the basis for calculating the first acoustic deformation parameter. The difference in sound wave energy distribution on both sides of the central axis can be the degree of asymmetry in sound wave energy in the adjacent areas to the left and right of the central axis within the anechoic groove region. This can be used to reflect the overall offset of the sound beam or the asymmetry of the propagation path caused by thermal deformation. In an exemplary embodiment, the difference in sound wave energy distribution on both sides of the central axis can be quantified by calculating the difference, ratio, or relative entropy of the mean sound pressure amplitude within the left and right windows. The first acoustic deformation parameter can be a local sound field offset metric parameter constructed based on the difference in sound wave energy distribution on both sides of the central axis within the anechoic groove region. This parameter can be used to quantify the overall deflection of the sound beam caused by thermal deformation of the wedge-pipe contact surface. Furthermore, the first acoustic deformation parameter can be obtained by converting the distribution difference into a dimensionless index through normalization or nonlinear mapping. In one specific embodiment, the first acoustic deformation parameter can be fused with the second acoustic deformation parameter to generate acoustic sub-parameters.

[0064] Based on the difference in sound wave energy distribution on both sides of the central axis of each anechoic groove region, the first acoustic deformation parameter of each anechoic groove region is obtained. This can be achieved by determining the central axis within the anechoic groove region, calculating the statistical characteristics of sound wave energy in its left and right adjacent regions, quantifying its asymmetry, and mapping it to the first acoustic deformation parameter. Furthermore, this operation can be achieved by using a sliding window to calculate the ratio of the average energy within a fixed width to the left and right of the central axis, or by using the Fourier spectrum centroid shift to characterize the difference in energy distribution between the left and right sides, thereby enabling a quantitative characterization of the overall sound beam shift or propagation asymmetry.

[0065] Based on the attenuation difference of the sound wave energy on both sides of the edge of each silencing groove, the acoustic deformation parameter of each silencing groove edge is obtained; based on the difference of the acoustic deformation parameters of the two silencing groove edges in the silencing groove region, the second acoustic deformation parameter of each silencing groove region is obtained. The attenuation difference of the acoustic wave energy on both sides of the edge of each anechoic groove can be the difference in the attenuation rate of the acoustic wave energy in the adjacent regions on the inner and outer sides of a single anechoic groove edge as the propagation distance changes. This can be used to characterize the scattering enhancement effect caused by local interface coupling or material thermal disturbance. In a specific embodiment, the attenuation difference of the acoustic wave energy on both sides of the edge of each anechoic groove can be obtained by fitting the slope of the local energy attenuation curve on both sides of the edge and calculating its difference. The acoustic deformation parameter can be a local disturbance index derived from the attenuation difference of the acoustic wave energy on both sides of a single anechoic groove edge. This can be used to characterize the degree of local distortion of the sound field near the edge due to thermal disturbance or structural discontinuity. Furthermore, the acoustic deformation parameter can be obtained by independently calculating the attenuation difference for each edge and mapping it to a scalar parameter. The second acoustic deformation parameter can be a measure of the difference between the corresponding acoustic deformation parameters of two anechoic groove edges within the same anechoic groove region. This can be used to capture the asymmetry of transverse sound field distortion caused by irregular structures or non-uniform thermal expansion. In one exemplary embodiment, the second acoustic deformation parameter can be obtained by calculating the absolute difference, relative deviation, or covariance of the acoustic deformation parameters at the two edges. Furthermore, the second acoustic deformation parameter can, together with the first acoustic deformation parameter, determine the acoustic sub-parameters.

[0066] Based on the attenuation difference of the sound wave energy on both sides of each anechoic groove edge, the acoustic deformation parameter of each anechoic groove edge is obtained. This can be achieved by setting a proximity analysis window on both sides of each anechoic groove edge, fitting the attenuation curve of sound energy with propagation depth, calculating the slope difference, and converting it into an acoustic deformation parameter. Furthermore, this operation can be achieved by exponentially fitting the time-domain signal envelope to extract the difference in attenuation constants, or by analyzing the difference in attenuation rate of high-frequency components in the frequency domain as a parameter input, thereby quantifying the scattering and attenuation anomalies in the local sound field caused by thermal disturbances or interface changes. Based on the difference in the acoustic deformation parameters of the two anechoic groove edges in the anechoic groove region, a second acoustic deformation parameter for each anechoic groove region is obtained. This can be achieved by comparing the acoustic deformation parameters corresponding to the two edges within the same anechoic groove region, calculating their numerical differences, and standardizing them. Furthermore, this operation can be achieved by calculating the absolute difference between the two parameters and dividing by the mean to obtain the relative deviation, or by constructing a two-dimensional vector and calculating the cosine of their included angle as a similarity measure, thereby reflecting the asymmetry of lateral sound field distortion and enhancing sensitivity to irregular structures or local thermal gradients.

[0067] Based on the first acoustic deformation parameter and the second acoustic deformation parameter, the acoustic sub-parameters of each anechoic groove region are obtained. Both the first acoustic deformation parameter and the second acoustic deformation parameter are positively correlated with the acoustic sub-parameters. The acoustic sub-parameters can be a comprehensive acoustic distortion index for a single anechoic groove region formed by fusing the first and second acoustic deformation parameters. This index can be used to achieve a multi-dimensional comprehensive measurement of the degree of distortion in the local acoustic environment. In a specific embodiment, the acoustic sub-parameters can be obtained by integrating the two types of parameters through weighted summation, product fusion, or logistic regression. Obtaining the acoustic sub-parameters for each anechoic groove region based on the first and second acoustic deformation parameters can be achieved by fusing the first and second acoustic deformation parameters according to a preset rule to generate a single comprehensive index. Furthermore, this operation can employ linear weighting: α × first parameter + β × second parameter, with the weights dynamically adjusted by the temperature gradient, or by using a small neural network to map the two parameters into a non-linearly fused acoustic sub-parameter, thereby achieving a multi-dimensional unified representation of local acoustic distortion.

[0068] By combining the acoustic sub-parameters of all silencing groove regions, the acoustic deformation coefficient of the pipe surface is obtained.

[0069] The acoustic sub-parameters of all anechoic groove regions can be a set of acoustic sub-parameters corresponding to all anechoic grooves covering the entire detection area. This set can provide spatially distributed prior information on acoustic field distortion, supporting the construction of the global acoustic deformation coefficient. The acoustic deformation coefficient of the pipe surface can be a global acoustic distortion characterization parameter generated by comprehensively analyzing the acoustic sub-parameters of all anechoic groove regions. This parameter can provide a high-dimensional, interpretable basis for acoustic distortion compensation for wall thickness measurement and defect location. In an exemplary embodiment, the acoustic deformation coefficient of the pipe surface can be generated through statistical aggregation (such as mean, weighted average, principal component projection) or spatial interpolation methods. Furthermore, compared to the simplified model based directly on energy uniformity and edge differences in the previous scheme, this embodiment introduces central axis analysis and edge attenuation differences to construct a more refined multi-level acoustic deformation characterization system, improving the ability to resolve complex thermo-acoustic coupling distortions. By comprehensively analyzing the acoustic sub-parameters of all anechoic groove regions, the acoustic deformation coefficient of the pipe surface can be obtained by spatial aggregation or statistical summarization of the acoustic sub-parameters of the entire region to form the global acoustic deformation coefficient. Furthermore, this operation can be achieved by calculating a weighted average of all acoustic sub-parameters, with the weights determined by the regional signal-to-noise ratio, or by constructing a continuous acoustic deformation field through Kriging interpolation and taking the mean or maximum value of the entire field as the coefficients. This can generate spatially representative acoustic distortion priors to support subsequent compensation.

[0070] Taking the online detection of high-temperature variable-diameter pipelines in petrochemical plants as an example, the high-temperature pipeline phased array noise reduction visualization detection sensing method in this embodiment can be to deploy phased array probes in the high-temperature variable-diameter section. After the system identifies multiple silencing groove regions, it calculates the energy distribution difference on both sides of the central axis of each region to obtain the first acoustic deformation parameter; at the same time, it analyzes the attenuation characteristics of the adjacent regions of each edge to generate acoustic deformation parameters, and compares the difference between the parameters of the two edges in the same region to obtain the second acoustic deformation parameter; after the two are fused into acoustic sub-parameters, the acoustic deformation coefficient is generated by combining the sub-parameters of the whole field; this coefficient accurately reflects the complex sound field distortion caused by the geometric change and thermal gradient superposition at the variable-diameter section, which is significantly better than the traditional single energy uniformity assessment method, thus providing a more accurate acoustic compensation basis for subsequent wall thickness thermal map correction.

[0071] In one embodiment, the method for obtaining acoustic deformation parameters includes: In the acoustic signal, all outermost acoustic points and all innermost acoustic points on the edge of each silencing groove are obtained respectively. All acoustic points adjacent to the outermost acoustic points are taken as the outer acoustic points of the region, and all acoustic points adjacent to the innermost acoustic points are taken as the inner acoustic points of the region. Neither the outer acoustic points nor the inner acoustic points of the region are on the edge of the silencing groove. Based on the energy difference between the total energy of the acoustic points on the outer side of each anechoic groove edge and the total energy of the acoustic points on the inner side of the edge, the acoustic deformation parameters of each anechoic groove edge are obtained.

[0072] The outermost acoustic points can be the set of acoustic response sampling points located on the edge of each anechoic groove in the acoustic signal, radially furthest from the pipe's central axis (i.e., closest to the pipe's outer surface). These points can serve as reference points defining the outer neighborhood of the region, used to construct the energy assessment boundary of the non-edge interference zone. In an exemplary embodiment, the outermost acoustic points can be obtained by analyzing the radial coordinates or sound path depth of each sampling point within the edge of the anechoic groove, selecting the point corresponding to the maximum value. Furthermore, the outermost acoustic points can work in conjunction with the innermost acoustic points to define the extreme boundary of the anechoic groove edge in the radial direction. The innermost acoustic points can be the set of acoustic response sampling points located on the edge of each anechoic groove in the acoustic signal, radially closest to the pipe's central axis (i.e., closest to the pipe's inner wall). These points can serve as reference points defining the inner neighborhood of the region, used to construct the non-interference energy assessment region facing the pipe's interior. For example, the innermost acoustic points can be obtained by analyzing the radial coordinates or sound path depth of each sampling point within the edge of the anechoic groove, selecting the point corresponding to the minimum value.

[0073] The acoustic points on the outer edge of the region can be all acoustic sampling points directly adjacent to the outermost acoustic point and not on the edge of the anechoic groove. These points can be used to characterize the sound propagation state from the outer edge of the anechoic groove towards the outside of the pipe, avoiding interference from the edge itself. In one specific embodiment, the acoustic points on the outer edge can be obtained in the acoustic signal space or imaging grid by identifying a set of points that are radially or temporally adjacent to the outermost acoustic point but do not belong to the edge of the anechoic groove. The acoustic points on the inner edge can be all acoustic sampling points directly adjacent to the innermost acoustic point and not on the edge of the anechoic groove. These points can be used to characterize the sound propagation state from the inner edge of the anechoic groove towards the center of the pipe, ensuring that energy assessment is not affected by reflections from the edge structure. In this embodiment, the acoustic points on the inner edge can be obtained in the acoustic signal space by identifying a set of points adjacent to the innermost acoustic point but excluding the edge itself. The total energy of the acoustic points on the outer edge can be the sum of squares of the acoustic signal amplitudes or the energy integral value corresponding to all acoustic points on the outer edge, which can be used to quantify the sound field intensity on the outer edge as input for asymmetric analysis. In one specific embodiment, the total energy of the acoustic points outside the region is obtained by performing energy calculations on the sound pressure signals of the acoustic points outside the region, such as L2 norm or RMS.

[0074] The total energy of the acoustic points within the region can be the sum of the energy of the corresponding acoustic signals of all acoustic points within the region. This energy can be used to quantify the sound field intensity inside the edge and form a comparison benchmark with the outside. Furthermore, the total energy of the acoustic points within the region is obtained by applying the same calculation method to the inner point set as the energy outside the region. The energy difference is the relative or absolute deviation between the total energy of the acoustic points outside the region and the total energy of the acoustic points inside the region. This can be used to directly reflect the asymmetry of sound beam deflection and attenuation caused by thermal deformation or structural irregularities. In this embodiment, the energy difference can be expressed as a difference, ratio, normalized difference index, etc. The acoustic deformation parameter can be a scalar index based on the above energy difference mapping, characterizing the degree of local sound field distortion at the edge of a single anechoic groove. This can be used to provide high-fidelity local input for the second acoustic deformation parameter, improving sensitivity to small thermo-acoustic coupling disturbances. In an exemplary embodiment, the acoustic deformation parameter is obtained by converting the energy difference into a dimensionless parameter through a linear or nonlinear function. Compared to the generalized description based on the difference in acoustic wave energy attenuation on both sides of the edge in the previous scheme, this embodiment explicitly defines the outermost / innermost acoustic points and their adjacent non-edge point sets, strictly isolates the interference of edge self-reflection, and makes the energy difference calculation more physically meaningful and spatially localized, thereby optimizing the accuracy and robustness of acoustic deformation parameters.

[0075] In the acoustic signal, all outermost and innermost acoustic points on the edge of each anechoic groove are acquired. This can be achieved by selecting extreme points for each identified anechoic groove edge based on the radial position or acoustic path depth of its sampling points. Further, this operation can be implemented by extracting the maximum / minimum radial coordinates of the edge points based on the phased array focusing depth map, or by selecting the earliest / latest response points as the innermost / outermost points in the time domain signal using the first arrival time as a depth proxy. This establishes a precise boundary for the inner and outer neighborhood division, avoiding bias introduced by subjective window settings. All acoustic points adjacent to the outermost acoustic point are designated as outer acoustic points of the region. This can be achieved by identifying points in the acoustic data space that are spatially, temporally, or indexally adjacent to the outermost acoustic point but do not belong to the edge of the anechoic groove. For example, this operation can be achieved by excluding the remaining points of the edge points in the 8-neighborhood of the outermost point in the imaging grid, or by taking a fixed number of non-edge sampling points before and after the outermost point in the A-scan sequence, thereby constructing an outer sound field evaluation region without edge interference.

[0076] All acoustic points adjacent to the innermost acoustic point are considered as the innermost acoustic points within the region. This can be achieved by identifying the set of points adjacent to the innermost acoustic point in the acoustic data space, excluding the edge itself. In one specific embodiment, this operation can be implemented by using radial interpolation to determine the innermost neighboring positions of the innermost point and mapping them back to acoustic points, or by using pixel-level images after beamforming to identify the inner connected regions, thereby constructing a clean sound field evaluation region facing the inside of the pipe. Based on the energy difference between the total energy of the outermost acoustic points of each anechoic groove edge and the total energy of the innermost acoustic points within the region, the acoustic deformation parameter of each anechoic groove edge is obtained. This can be achieved by calculating the total energy of the inner and outer sides separately, determining their difference, and mapping it to the acoustic deformation parameter. Furthermore, this operation can be implemented by using the energy ratio after log transformation as a parameter, or by using normalized difference, thereby enabling quantitative perception of local asymmetric distortion of the sound field.

[0077] For example, in the scenario of monitoring cracks in high-temperature bends in industrial high-temperature pipeline systems, the high-temperature pipeline phased array noise reduction visualization detection sensing method of this embodiment can be as follows: In the detection of the main steam bend in the high-temperature pipeline area, after the system identifies the edge of the silencer groove, it accurately locates the outermost and innermost acoustic points of each edge; then, it extracts the acoustic points on the outer side and inner side of the adjacent but non-edge regions, and integrates the energy respectively; it is found that because the outer arc of the bend is heated more intensely, the energy on the outer side of the region is significantly lower than that on the inner side, and the energy difference is mapped to a high-value acoustic deformation parameter; this parameter accurately reflects the inward deflection effect of the sound beam caused by the thermal gradient, providing a high signal-to-noise ratio input for the subsequent calculation of the second acoustic deformation parameter, and finally reducing the defect location error and avoiding misjudging the thermally induced acoustic deflection as wall thickness reduction.

[0078] In one embodiment, the method for obtaining the coefficient of thermal expansion deformation includes: In each anechoic groove region, the thermal expansion torsion parameters of each anechoic groove region are obtained based on the distance between all the innermost acoustic points on the edges of the two anechoic grooves. The radial distance of each silencing groove region is negatively correlated and mapped as a weight. The corresponding thermal expansion and torsion parameters are then weighted to obtain the thermal expansion deformation sub-parameters of each silencing groove region. By combining the thermal expansion deformation parameters of all the silencing groove regions, the thermal expansion deformation coefficient of the pipe surface is obtained.

[0079] In this system, all the innermost acoustic points on the edges of the two anechoic grooves can be the set of acoustic response sampling points located radially closest to the central axis of the pipe within a single anechoic groove region, on the edges of the two anechoic grooves respectively. These points can serve as sensitive indicators reflecting geometric deformation caused by local thermal expansion, and their spatial relationship is directly related to changes in the lateral dimensions of the pipe. In an exemplary embodiment, the definition of the innermost acoustic points on the edges of the two anechoic grooves follows the previous scheme, emphasizing that they appear in pairs on the two edges of the same anechoic groove region for distance calculation. The distance between the innermost acoustic points can be the Euclidean distance or equivalent sound path distance between the innermost acoustic points on each edge within the same anechoic groove region in the acoustic imaging space or physical space. This distance can be used to directly characterize the degree of lateral geometric distortion caused by thermal expansion in the local area. Furthermore, the distance between the innermost acoustic points can be calculated by extracting coordinates, or converted into physical spacing by combining time delay difference with a sound velocity model.

[0080] The thermal expansion distortion parameter of each anechoic groove region can be a dimensionless index reflecting local thermally induced geometric distortion, constructed based on the distance between the innermost acoustic points. This index can be used to quantify the non-uniform expansion effect of local pipe regions at high temperatures, serving as a basic unit for thermal deformation modeling. In one specific embodiment, the thermal expansion distortion parameter of each anechoic groove region can be calculated by comparing the distance value with a baseline distance at room temperature to determine the relative rate of change or deviation. The radial distance of each anechoic groove region can be the perpendicular distance from the center of the anechoic groove region to the central axis of the pipe, which can be used to evaluate the region's position on the pipe cross-section and its reliability weight in response to thermal expansion. Furthermore, the radial distance of each anechoic groove region follows the previous definition, explicitly used here for negative correlation mapping to generate weighting factors. The weights after negative correlation mapping can be weighting coefficients obtained by converting the radial distance through a monotonically decreasing function, which can be used to assign higher weights to the inner regions, balancing the contradiction between thermal expansion sensitivity and signal-to-noise ratio. In an exemplary embodiment, the weights after negative correlation mapping can be achieved through reciprocal or exponential decay methods.

[0081] The thermal expansion deformation sub-parameter of each anechoic groove region can be a local thermal deformation characterization quantity obtained by weighting the thermal expansion torsion parameter with a negative correlation weight based on radial distance. This can be used to fuse geometric distortion information with signal reliability, improving the robustness of local thermal deformation estimation. Furthermore, the thermal expansion deformation sub-parameter of each anechoic groove region can be obtained by multiplying the thermal expansion torsion parameter by the weight after negative correlation mapping. The thermal expansion deformation sub-parameters of all anechoic groove regions can be a set of thermal expansion deformation sub-parameters corresponding to all anechoic grooves covering the detection area. This can be used to provide spatially distributed thermal expansion priors, supporting the construction of global thermal expansion deformation coefficients.

[0082] The thermal expansion deformation coefficient of the pipe surface can be a global thermal deformation characterization parameter formed by the comprehensive aggregation of all thermal expansion deformation sub-parameters. This parameter can provide a highly reliable thermodynamic compensation basis for wall thickness measurement and defect location. In one specific embodiment, the thermal expansion deformation coefficient of the pipe surface can be generated as a single or field-form coefficient through statistical averaging, weighted summation, or spatial interpolation. Furthermore, compared to the simple inversion based solely on peak point spacing and radial distance in previous schemes, this embodiment introduces the distance of the innermost acoustic points as a more stable geometric feature and adds a radial distance negative correlation weighting mechanism. This makes the thermal expansion deformation coefficient possess both deformation sensitivity and signal reliability, significantly optimizing the compensation accuracy under high-temperature irregular pipe conditions. In each anechoic groove region, the thermal expansion torsion parameter of each anechoic groove region is obtained based on the distance between all innermost acoustic points on the edges of the two anechoic grooves. This can be achieved by extracting the innermost acoustic points of each of the two edges of each anechoic groove region, calculating the distance between them, and converting this distance into a relative deformation as the thermal expansion torsion parameter. Furthermore, this operation can be achieved by using Euclidean distance to calculate the distance between two points in physical space (requiring knowledge of the sound speed and imaging ratio), or by estimating the equivalent distance based on the time delay difference multiplied by the sound speed (suitable for online scenarios where the geometry is not fully calibrated), thereby enabling direct and sensitive quantification of local thermally induced geometric distortions.

[0083] The radial distance of each silencing groove region is negatively correlated and mapped as a weight. This weight is then applied to the corresponding thermal expansion and torsion parameter to obtain the thermal expansion deformation sub-parameter for each silencing groove region. This can be achieved by applying a negative correlation function to the radial distance of each region to generate the weight, multiplying it by the thermal expansion and torsion parameter, and outputting the weighted sub-parameter. Further, this operation can avoid division by zero by using w = 1 / (r + ε), where r is the radial distance of the silencing groove region from the central axis of the pipe (the larger the distance, the smaller the weight), and ε is a small constant (e.g., 0.001) to avoid division by zero when r = 0, thus balancing numerical stability. Alternatively, a piecewise function can be used (w = 1 when r < r0, otherwise w = exp(-β(r...)). (r0) where r0 is the radial distance threshold, distinguishing between near-field and far-field regions; β is the attenuation rate coefficient, controlling the attenuation rate of the weights in the far-field region (the larger β is, the faster the weights attenuate with increasing distance). This allows the introduction of signal quality priors in thermal deformation estimation, suppressing interference from low signal-to-noise ratio regions on the outside. The thermal expansion deformation coefficient of the pipe surface is obtained by synthesizing the thermal expansion deformation sub-parameters of all anechoic groove regions. This can be achieved by aggregating the thermal expansion deformation sub-parameters of all regions to form a global thermal expansion deformation coefficient. Furthermore, this operation can be achieved by calculating a weighted average (the weights can further consider regional signal-to-noise ratio or temperature gradient), or by constructing a thermal expansion deformation field and extracting the dominant mode as coefficients through principal component analysis, thereby generating a spatially representative and reliable weighted thermal deformation prior.

[0084] For example, in the scenario of online monitoring of high-temperature main pipeline bends, the high-temperature pipeline phased array noise reduction visualization detection sensing method of this embodiment can be to deploy a phased array probe at the 90° bend of the high-temperature main pipeline. After the system identifies multiple silencing groove regions, it extracts the innermost acoustic points of the two edges of each region and calculates their spacing. It is found that the spacing on the outer arc side increases significantly, indicating thermal expansion and stretching. At the same time, because the radial distance of the outer arc region is large, its negative correlation weight is small, while the inner arc has a small deformation but a high weight. After weighting, the thermal expansion deformation sub-parameters of each region more evenly reflect the real thermal expansion trend. Finally, the aggregated thermal expansion deformation coefficient accurately captures the asymmetric expansion characteristics of the bend, effectively compensates for the underestimation of the wall thickness caused by thermal drift, and avoids misjudgment.

[0085] In one embodiment, the method for obtaining the thermal expansion distortion parameter includes: In each silencing groove region, all the innermost acoustic points on the edge of each silencing groove are sorted sequentially. Based on the distance between the innermost acoustic points with the same sorting number on the edges of two silencing grooves, the average width and width range of the corresponding silencing groove are obtained. The relative flatness of each silencing groove region is obtained based on the average width of the groove corresponding to each silencing groove region and its deviation from the average width of the groove corresponding to all silencing groove regions; the width range is used as the torsion coefficient of the silencing groove region. The thermal expansion torsion parameters of each silencing groove region are obtained based on the relative flatness and torsion coefficient. Both the relative flatness and the torsion coefficient are positively correlated with the thermal expansion torsion parameters.

[0086] In this embodiment, all innermost acoustic points on the edge of each anechoic groove can be the set of all acoustic response sampling points whose radial position is closest to the central axis of the pipe on the edge of a single anechoic groove. These can be used as sensitive reference points reflecting local thermal expansion geometric changes. In this embodiment, all innermost acoustic points on the edge of each anechoic groove need to be sequentially sorted to establish point-to-point correspondence. Furthermore, all innermost acoustic points on the edge of each anechoic groove can be sorted according to the arrival time of the sound waves, which is suitable for time-domain signals, or sorted according to the circumferential coordinates of the pixels after imaging, which is suitable for TFM or SAFT reconstructed images. Innermost acoustic points with the same sorting number can be a pair of points with the same number after the innermost acoustic points of the two edges within the same anechoic groove region are sorted in spatial or temporal order. This can be used to establish a geometric correspondence across the edges. In an exemplary embodiment, innermost acoustic points with the same sorting number are sorted along the length of the anechoic groove to ensure that the point pairs correspond geometrically.

[0087] The distance between the innermost acoustic points with the same sequence number can be the spacing between each pair of innermost acoustic points with the same sequence number in the imaging space or physical space, and can be used to characterize the instantaneous width of the anechoic groove at a local location. The average groove width can be the statistical average of the distances between all pairs of innermost acoustic points with the same sequence number within the same anechoic groove area, and can be used to reflect the overall lateral dimensional change trend of the area due to thermal expansion. In a specific embodiment, the average groove width is obtained by taking the arithmetic mean or median of the distances between all point pairs. The width range can be the difference between the maximum and minimum values ​​of the distances between all pairs of point pairs with the same sequence number within the same anechoic groove area, and can be used to quantify the degree of non-uniform deformation along the length of the anechoic groove. The average groove width corresponding to all anechoic groove areas can be the set of the average groove widths of each anechoic groove area within the entire detection area, and can be used to provide a global reference benchmark. The deviation of the average groove width corresponding to each anechoic groove area can be the deviation of the average groove width of a single area from the global average value, which can be the absolute difference or standardized residual, and can be used to measure the degree of abnormality of the expansion or contraction of the area relative to the overall pipeline. Furthermore, the deviation of the average width of the corresponding groove in each silencing groove area is obtained by calculating the difference between the individual value and the global mean, or by dividing by the standard deviation to obtain the Z-score.

[0088] Relative flatness can be a dimensionless parameter characterizing local relative expansion, constructed based on the average width of the groove and its global deviation. It can be used to reflect the flattening or bulging trend of a local area of ​​the pipe relative to the whole at high temperatures. In one specific embodiment, the relative flatness is converted into a positively correlated index by normalizing or nonlinearly mapping the deviation. The torsion coefficient can be a direct measure of the width range as an index to quantify local non-uniform deformation. It can be used to capture local acoustic path distortion caused by pipe curvature, support constraints, or temperature gradients. In this embodiment, the torsion coefficient is explicitly defined for the first time, distinguishing it from the traditional single width index and emphasizing sensitivity to non-uniformity. The thermal expansion torsion parameter can be a composite parameter that integrates relative flatness and torsion coefficient to comprehensively characterize local thermally induced geometric distortion. It can be used to simultaneously reflect global scale changes and local non-uniform distortion. In an exemplary embodiment, the thermal expansion torsion parameter combines the two through a weighted sum, product, or other fusion function, and both are positively correlated with the output. Compared to the previous simplified model based solely on the distance between two points, this embodiment introduces multi-point pair statistical features and separates the two physical dimensions of "flatness" and "twist," significantly improving the ability to distinguish complex thermo-mechanical coupled deformations.

[0089] In each anechoic groove region, all innermost acoustic points on the edge of each anechoic groove are sequentially sorted, either along the length of the groove. Furthermore, this sequential sorting can be achieved by ordering by sound wave arrival time or by the circumferential coordinates of the pixels after imaging, thus establishing a geometric correspondence and ensuring that subsequent point-to-point distance calculations have physical meaning. Based on the distance between all innermost acoustic points with the same sorting number on the edges of two anechoic grooves, the average width and width range of the corresponding anechoic groove are obtained. This can be achieved by calculating the average distance of all point pairs with the same number, using the difference between the maximum and minimum values ​​as the width range. Furthermore, based on the distance between the innermost acoustic points with the same sorting number on the edges of the two silencing grooves, the average width and width range of the corresponding silencing groove can be obtained. This can be achieved by using the median instead of the mean to suppress interference from outliers, or by using a sliding window local range to identify abrupt change regions. In this way, overall expansion information and local non-uniformity features can be extracted simultaneously.

[0090] Based on the average width of each silencing groove region and its deviation from the average width of all corresponding silencing groove regions, the relative flatness of each silencing groove region can be obtained. This can be achieved by calculating the deviation between the average width of an individual groove and the global mean, and mapping it to the relative flatness. Furthermore, the relative flatness of each silencing groove region can be obtained by dividing the relative flatness by the difference between the individual width and the global mean, or by using the Sigmoid function to compress the deviation to the [0, 1] interval as the flatness output, thereby quantifying the degree of expansion anomaly of the local region relative to the overall pipeline. The width range can be used as the distortion coefficient of the silencing groove region, which can be directly assigned: the distortion coefficient equals the width range. Further, the width range can be used as the distortion coefficient of the silencing groove region by performing a logarithmic transformation on the range to compress the dynamic range, or by combining the local signal-to-noise ratio with confidence weighting of the range, thereby achieving direct quantification of local non-uniform deformation. The thermal expansion torsion parameters of each anechoic groove region can be obtained by fusing the relative flatness and torsion coefficient into a single parameter according to a positive correlation rule. Furthermore, the thermal expansion torsion parameters of each anechoic groove region can be obtained by multiplying the relative flatness by the positive coefficient and the relative flatness by the positive coefficient and the torsion coefficient, or by multiplying the relative flatness by (1 plus the positive coefficient and the torsion coefficient). This allows for the construction of a composite thermal deformation characterization that considers both overall expansion and local torsion.

[0091] For example, in the scenario of online monitoring of high-temperature main steam bends in nuclear power plants, the high-temperature pipeline phased array noise reduction visualization detection sensing method of this embodiment can be to deploy phased array probes in the high-temperature zone of the 90° bend. After the system identifies multiple silencing groove areas, the innermost acoustic points of each edge are sorted circumferentially, and the distance between points with the same number is calculated. It is found that the average width of the outer arc side groove is significantly larger than that of the inner arc, and the local width difference of the outer arc is large, indicating the existence of non-uniform expansion. The relative flatness shows that the outer arc area is obviously bulging, and the torsion coefficient reveals the deformation abrupt change near the local support point. The thermal expansion torsion parameter generated by the fusion of the two accurately characterizes the thermally induced composite deformation characteristics of the bend. After radial weighting, this parameter participates in the construction of the thermal expansion deformation coefficient, effectively compensating for the wall thickness misjudgment caused by ignoring non-uniformity in traditional methods, and improving the defect location accuracy.

[0092] In one embodiment, the method for obtaining the comprehensive deformation coefficient includes: The product of the acoustic deformation coefficient and the thermal expansion deformation coefficient is used as the comprehensive deformation coefficient of the pipe surface.

[0093] The product can be the mathematical result of multiplying the acoustic deformation coefficient and the thermal expansion deformation coefficient, and can be used as a specific fusion method to construct the comprehensive deformation coefficient, reflecting the nonlinearity of the thermo-acoustic coupling effect. In an exemplary embodiment, the operating principle of the product can be explained in context: at the same spatial location, the acoustic deformation coefficient reflects the degree of sound beam deflection and noise interference caused by thermal disturbance at the wedge-pipe contact surface, while the thermal expansion deformation coefficient characterizes the local geometric deformation of the pipe body caused by the temperature gradient. The multiplication of the two implies the physical logic that the sensitivity of sound field distortion to geometric deformation is amplified with the degree of thermal expansion. Furthermore, the product can be used to achieve nonlinear modeling of the combined effect of sound field distortion and geometric deformation at high temperatures. Using the product of the acoustic deformation coefficient and the thermal expansion deformation coefficient as the comprehensive deformation coefficient of the pipe surface can be achieved by performing point-by-point multiplication of the acoustic deformation coefficient and the thermal expansion deformation coefficient corresponding to the same spatial location to generate the comprehensive deformation coefficient at that location. Furthermore, the product of the acoustic deformation coefficient and the thermal expansion deformation coefficient can be used as the comprehensive deformation coefficient of the pipe surface. This can be achieved by multiplying the two deformation coefficient fields element by element at the pixel level or sampling point level, or by inputting the two coefficients into a nonlinear product model through a lookup table or function mapping to output the comprehensive coefficient. This enables the nonlinear fusion modeling of the thermo-acoustic coupling effect, which more realistically reflects the combined effects of sound field distortion and geometric deformation at high temperatures.

[0094] Taking the online monitoring of high-temperature bends in petrochemical plants as an example, the high-temperature pipeline phased array noise reduction visualization detection sensing method in this embodiment can deploy a phased array probe on a high-temperature variable-diameter bend in a petrochemical unit. The system simultaneously acquires acoustic signals and infrared images, identifies the acoustic isolation structure area, and calculates the acoustic deformation coefficient (reflecting the sound beam deflection caused by thermal disturbance at the wedge contact surface) and the thermal expansion deformation coefficient (reflecting local diameter expansion). At the point of maximum bend curvature, the thermal expansion coefficient is higher. Even if the acoustic distortion is slight, its product still significantly amplifies the comprehensive deformation coefficient, thereby applying stronger compensation in the wall thickness inversion. The final generated heat map accurately presents the actual wall thickness reduction in this area due to thermal expansion, avoiding the problem of underestimating errors in the traditional linear superposition method.

[0095] In one embodiment, the method for obtaining the defect location result includes: When the comprehensive deformation coefficient is greater than the preset threshold, the defect location result is determined to be that there is a defect; otherwise, the defect location result is determined to be that there is no defect. When it is determined that there is a defect, the specific location of the defect is determined according to the location of the maximum value of the comprehensive deformation coefficient, and a thermal map of the wall thickness of the high-temperature irregular pipe is generated.

[0096] The preset threshold can be a critical value of the comprehensive deformation coefficient used to determine whether a defect exists. It can be used as a criterion for determining whether a defect exists, thereby achieving automated anomaly identification. In this embodiment, the preset threshold can be set through statistical analysis of historical detection data, simulation modeling, or on-site calibration, and can be a fixed value or a temperature adaptive function.

[0097] When the overall deformation coefficient exceeds a preset threshold, the defect location result is determined to be defective. This can be achieved by comparing the maximum value of the overall deformation coefficient across the entire field or a local area with the preset threshold; if it exceeds, a defect is marked. Furthermore, this determination can be achieved by taking the global maximum value of the overall deformation coefficient across the entire pipe surface for single-point judgment, or by dividing the detection into multiple sub-regions and judging whether each region has local exceedances of the threshold. This enables automatic initial defect screening based on physical fusion indicators, reducing the false alarm rate. When the overall deformation coefficient is not greater than the preset threshold, the defect location result is determined to be defect-free. This can be achieved by outputting a defect-free conclusion if the overall deformation coefficient does not exceed the preset threshold. Furthermore, this determination can be achieved by using a hysteresis comparison mechanism to prevent frequent switching of the determination result near the threshold, or by combining time series stability analysis to confirm a state consistently below the threshold. This avoids over-response to normal thermo-acoustic disturbances and improves system robustness.

[0098] The location of the maximum value of the comprehensive deformation coefficient can be the coordinate position of the maximum value in the spatial distribution of the comprehensive deformation coefficient on the pipe surface. This can be used to indicate the center of the most likely defect area for precise location. In an exemplary embodiment, the location of the maximum value of the comprehensive deformation coefficient can be determined by its spatial index or physical coordinates in the comprehensive deformation coefficient field using an extremum search algorithm (such as local maximum detection or gradient ascent). The specific location of the defect can be the geometric coordinates of the defect on the pipe surface output after location by the maximum value of the comprehensive deformation coefficient. This can be used to provide actionable defect spatial information, supporting maintenance decisions and risk assessments. In a specific embodiment, the specific location of the defect is directly mapped from the location of the maximum value of the comprehensive deformation coefficient, serving as a core component of the defect location result.

[0099] When a defect is identified, its specific location is determined based on the maximum value of the comprehensive deformation coefficient. This can be achieved by extracting the spatial coordinates corresponding to the maximum value in the comprehensive deformation coefficient field as the defect location, given the defect is identified. Furthermore, this location can be improved by using sub-pixel interpolation to enhance the accuracy of the maximum value location, or by combining a neighborhood weighted centering method to optimize the stability of extreme point location. This allows for high-confidence spatial location of defects, reducing subsequent manual verification costs. Generating a thermal map of the wall thickness of high-temperature irregular-shaped pipes can be achieved by mapping the comprehensive deformation coefficient or corrected wall thickness data to a color-coded image based on spatial location. Further, this thermal map can be generated by highlighting abnormal areas using the comprehensive deformation coefficient itself as the intensity basis, or by using the comprehensive deformation coefficient to correct the original ultrasonic thickness measurement value to generate a true wall thickness thermal map. This provides an intuitive, quantitative, and interpretable visualization of the wall thickness status.

[0100] For example, in the scenario of online inspection of high-temperature variable-diameter pipelines in petrochemical plants, the high-temperature pipeline phased array noise reduction visualization detection sensing method of this embodiment can be to deploy phased array probes on the operating high-temperature variable-diameter pipeline, and the system calculates the comprehensive deformation coefficient field in real time; when the coefficient value of a certain area exceeds a preset threshold (such as 0.35), it is determined that there is a defect there; further locate the position of the maximum value in the field, and accurately output that the defect is located at 120° circumferentially and 85mm axially from the weld; at the same time, the comprehensive deformation coefficient distribution of the entire pipeline is rendered as a color heat map, with the red area corresponding to the high deformation area, so that maintenance personnel can quickly identify the potential thinning or crack location, without the need for machine shutdown or additional calibration.

[0101] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of the present invention, and the present invention is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A phased array noise reduction and visualization detection sensing method for high-temperature pipelines, characterized in that, The method includes: Acquire acoustic signals and infrared thermal imaging data from the surface of a high-temperature pipeline, acquire all silencing groove regions in the acoustic signals and the two edges of each silencing groove region, and acquire the radial distance of each silencing groove region from the central axis of the pipeline. Based on the uniformity of sound wave energy distribution in all the silencing groove regions and the difference in sound wave energy between the two adjacent sides of each silencing groove edge in each silencing groove region, the acoustic deformation coefficient of the pipe surface is obtained. The thermal expansion deformation coefficient of the pipe surface is obtained based on the distance between the peak points of sound wave energy on the edges of the two sound-absorbing grooves in each sound-absorbing groove region and the radial distance of each sound-absorbing groove region. Based on the acoustic deformation coefficient and the thermal expansion deformation coefficient, the comprehensive deformation coefficient of the pipe surface is obtained; based on the comprehensive deformation coefficient of the pipe surface, the thermal map of the high-temperature irregular pipe wall thickness and the defect location results are obtained.

2. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 1, characterized in that, The method for obtaining the silencing groove region includes: Obtain all acoustic boundaries in the sound wave signal, and obtain the average area of ​​the region enclosed by all the acoustic boundaries. The region enclosed by the acoustic boundaries with an area smaller than the average area is taken as the suspected anechoic groove region. The degree of similarity of the acoustic boundary to each of the suspected anechoic groove regions is used as the first confidence level of each of the suspected anechoic groove regions. Based on the distribution regularity of each suspected anechoic groove region and its adjacent anechoic groove regions, a second confidence level is obtained for each suspected anechoic groove region. The confidence level of the anechoic groove in each suspected anechoic groove region is obtained based on the first confidence level and the second confidence level, and both the first confidence level and the second confidence level are positively correlated with the confidence level of the anechoic groove; all suspected anechoic groove regions whose confidence level of the anechoic groove is greater than a preset threshold are taken as anechoic groove regions.

3. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 2, characterized in that, The method for obtaining the class moment degree includes: Starting from any boundary point on each acoustic boundary, obtain the 8-chain code corresponding to the acoustic boundary; in the 8-chain code corresponding to each acoustic boundary, sort the occurrence frequency of each chain code value in descending order, take the first two chain code values ​​as the first group of chain code values, and take the third and fourth chain code values ​​as the second group of chain code values. On the acoustic boundary, based on the difference between the chain code values ​​of each group of chain code values ​​and the numerical difference of a preset positive integer, as well as the difference in the number of boundary points corresponding to the chain code values, a confidence coefficient is obtained for each group of chain code values ​​corresponding to the chain code values ​​of a set of opposite sides of a rectangle; the numerical difference and the number difference are both negatively correlated with the confidence coefficient. Multiply the confidence coefficients of the two sets of chain code values ​​to obtain the class moment degree corresponding to the acoustic boundary.

4. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 2, characterized in that, The methods for obtaining the second confidence level include: Take any of the suspected silencing groove areas as the target area; within the neighborhood of the target area, take the two suspected silencing groove areas on the left and right as the horizontal reference areas, and take the two suspected silencing groove areas on the top and bottom as the vertical reference areas; The first distance and the second distance between the centroids of the two horizontal reference regions and the target region are obtained respectively. The difference between the first distance and the second distance is negatively correlated and mapped to obtain the horizontal uniform distribution parameters. The third and fourth distances between the centroids of the two vertical reference regions and the target region are obtained respectively. The differences between the third and fourth distances are negatively correlated and mapped to obtain the vertical uniform distribution parameters. The second confidence level of each target region is obtained by combining the horizontal uniform distribution parameters and the vertical uniform distribution parameters.

5. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 1, characterized in that, The method for obtaining the acoustic deformation coefficient includes: Based on the difference in sound wave energy distribution on both sides of the central axis of each anechoic groove region, a first acoustic deformation parameter is obtained for each anechoic groove region; the central axis is parallel to the edge of the anechoic groove, and the distribution difference is positively correlated with the first acoustic deformation parameter; Based on the attenuation difference of the sound wave energy on both sides of each anechoic groove edge, the acoustic deformation parameter of each anechoic groove edge is obtained; based on the difference of the acoustic deformation parameters of the two anechoic groove edges in the anechoic groove region, the second acoustic deformation parameter of each anechoic groove region is obtained. Based on the first acoustic deformation parameter and the second acoustic deformation parameter, acoustic sub-parameters of each of the silencing groove regions are obtained, and both the first acoustic deformation parameter and the second acoustic deformation parameter are positively correlated with the acoustic sub-parameters; By combining the acoustic sub-parameters of all the aforementioned silencing groove regions, the acoustic deformation coefficient of the pipe surface is obtained.

6. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 5, characterized in that, The method for obtaining the acoustic deformation parameters includes: In the acoustic signal, all outermost acoustic points and all innermost acoustic points on the edge of each silencing groove are obtained respectively. All acoustic points adjacent to the outermost acoustic points are taken as external acoustic points of the region, and all acoustic points adjacent to the innermost acoustic points are taken as internal acoustic points of the region. Neither the external acoustic points nor the internal acoustic points are on the edge of the silencing groove. The acoustic deformation parameters of each edge of the anechoic groove are obtained based on the energy difference between the total energy of the acoustic points outside the region of each edge of the anechoic groove and the total energy of the acoustic points inside the region.

7. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 6, characterized in that, The method for obtaining the coefficient of thermal expansion deformation includes: In each of the silencing groove regions, the thermal expansion torsion parameters of each silencing groove region are obtained based on the distance between all the innermost acoustic points on the edges of the two silencing grooves. The radial distance of each of the silencing groove regions is negatively correlated and then used as a weight to weight the corresponding thermal expansion torsion parameter, thereby obtaining the thermal expansion deformation sub-parameter of each of the silencing groove regions. By combining the thermal expansion deformation sub-parameters of all the aforementioned silencing groove regions, the thermal expansion deformation coefficient of the pipe surface is obtained.

8. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 7, characterized in that, The method for obtaining the thermal expansion distortion parameter includes: In each of the silencing groove regions, all the innermost acoustic points on the edge of each silencing groove are sequentially sorted. Based on the distance between the innermost acoustic points with the same sorting number on the edges of two silencing grooves, the average width and width range of the corresponding silencing groove are obtained. The relative flatness of each silencing groove region is obtained based on the average width of the groove corresponding to each silencing groove region and the degree of deviation of the average width of the grooves corresponding to all silencing groove regions; the width range is used as the torsion coefficient of the silencing groove region. The thermal expansion torsion parameters of each of the silencing groove regions are obtained based on the relative flatness and the torsion coefficient, and both the relative flatness and the torsion coefficient are positively correlated with the thermal expansion torsion parameters.

9. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 1, characterized in that, The method for obtaining the comprehensive deformation coefficient includes: The product of the acoustic deformation coefficient and the thermal expansion deformation coefficient is taken as the comprehensive deformation coefficient of the pipe surface.

10. The high-temperature pipeline phased array noise reduction and visualization detection sensing method according to claim 1, characterized in that, The method for obtaining the defect location result includes: When the comprehensive deformation coefficient is greater than the preset threshold, the defect location result is determined to be that there is a defect; otherwise, the defect location result is determined to be that there is no defect. When it is determined that there is a defect, the specific location of the defect is determined according to the maximum value of the comprehensive deformation coefficient, and a thermal map of the wall thickness of the high-temperature irregular pipe is generated.

Images