Steel pipe high-frequency welding temperature zone monitoring method and device, electronic equipment and storage medium

Abstract

The present application relates to the technical field of high-frequency straight seam welded pipe production process detection, and particularly relates to a steel pipe high-frequency welding temperature zone monitoring method and device, electronic equipment and storage medium, the method of the present application first acquires a first thermal image sequence; then each thermal image in the first thermal image sequence is subjected to differential processing, and the differential amplitude graph obtained for representing the differential amplitude and the differential direction is constructed into a differential image sequence; then the boundary feature of each differential amplitude graph in the differential image sequence is extracted, and the boundary graph obtained for representing the temperature gradient edge line is constructed into a boundary image sequence; finally, the boundary form of the boundary image sequence is analyzed, and the stability of the welding state is evaluated according to the form analysis result. Through the infrared thermal imaging time sequence differential and boundary extraction technology, the temperature gradient edge of the steel pipe high-frequency welding temperature zone can be accurately positioned, the welding heat affected zone can be limited in the preset range in real time, and the uniformity and stability of the welding stress field are ensured.

Description

Technical Field

[0001] This invention relates to the field of high-frequency straight seam welded pipe production process testing technology, and in particular to a method, device, electronic equipment and storage medium for monitoring the high-frequency welding temperature zone of steel pipes. Background Technology

[0002] High-frequency straight seam welded steel pipes are widely used in oil and gas transportation, engineering structures, and machinery manufacturing. The quality of the weld and the dimensional accuracy of the pipe are core indicators for evaluating product performance. During high-frequency welding, the spatial range, boundary morphology, and distribution stability of the welding heat-affected zone directly determine the uniformity of the welding residual stress field, which is a key process for controlling the out-of-roundness of steel pipes and suppressing forming defects.

[0003] Currently, the control of welding temperature zone in welded pipe production relies heavily on operator experience and post-production spot checks. Online monitoring often uses single-point infrared temperature measurement, which can only obtain the local temperature peak of the weld and cannot fully characterize the two-dimensional shape and boundary distribution of the temperature zone, making it difficult to grasp the size fluctuation and shape deviation of the heat-affected zone in real time.

[0004] Monitoring schemes that use infrared thermal imaging often stop at temperature threshold alarms and simple temperature statistics for the whole image, without conducting fine extraction and morphological quantitative analysis of temperature zone boundaries. Furthermore, conventional edge detection algorithms are limited by on-site arc light, smoke and dust interference and pixel grid accuracy, resulting in insufficient boundary positioning accuracy. They cannot achieve quantitative assessment of the temporal fluctuations in temperature zone morphology, making it difficult to provide reliable real-time data support for welding stress control and non-roundness control.

[0005] The aforementioned technical limitations prevent the stability of the welding temperature zone morphology from being quantified online. The heat-affected zone range is prone to deviating from the preset process range, which in turn leads to uneven welding stress distribution. Ultimately, this results in quality problems such as excessive roundness of the steel pipe and fluctuations in the mechanical properties of the weld, thus hindering the improvement of the yield and product consistency of high-speed welded pipe production lines.

[0006] Therefore, it is necessary to develop and design a method for monitoring the high-frequency welding temperature zone of steel pipes. Summary of the Invention

[0007] The present invention provides a method, device, electronic device and storage medium for monitoring the high-frequency welding temperature zone of steel pipes, which is used to solve the problem in the prior art that the unstable welding temperature zone morphology leads to uneven stress distribution and thus affects the roundness of steel pipes.

[0008] In a first aspect, embodiments of the present invention provide a method for monitoring the temperature zone during high-frequency welding of steel pipes, comprising: A first thermal imaging sequence is obtained, which includes multiple thermal imaging images arranged in time sequence. The thermal imaging images are obtained based on the high-frequency welding area of ​​the steel pipe, and each pixel in the thermal imaging image corresponds to a temperature observation value. Each thermal image in the first thermal image sequence is subjected to differential processing, and the differential amplitude maps representing the differential amplitude and differential direction are constructed into a differential image sequence. Boundary features are extracted from each differential amplitude map in the differential image sequence, and the obtained boundary maps representing the edge lines of the temperature gradient are constructed into a boundary image sequence. Boundary form analysis is performed on the boundary image sequence, and the stability of the welding state is evaluated based on the form analysis results.

[0009] In one possible implementation, the differential processing of each thermal image in the first thermal image sequence includes: For each thermal image in the first thermal image sequence, perform the following steps: Obtain horizontal and vertical difference data blocks; According to the predetermined sliding direction, the data block to be differentiated is extracted from the thermal image, wherein the data block to be differentiated has the same specifications as the horizontal differential data block and the vertical differential data block; The horizontal difference value is obtained by multiplying the data block to be differencing with the horizontal difference data block by its nodes and summing the results. The vertical difference data block is multiplied by the vertical difference data block by its position and summed to obtain the vertical difference value; The horizontal difference value and the vertical difference value are stored in the horizontal difference map and the vertical difference map, respectively; If the traversal of the thermal image is not completed, proceed to the step of extracting the data block to be differentiated from the thermal image according to the predetermined sliding direction; Otherwise, based on the horizontal difference map and the vertical difference map, a difference magnitude map representing the difference magnitude and the difference direction is constructed.

[0010] In one possible implementation, constructing a difference magnitude map representing the difference magnitude and the difference direction based on the horizontal difference map and the vertical difference map includes: Get the location index; Based on the location pointed to by the location index, the horizontal difference value to be processed and the vertical difference value to be processed are extracted from the horizontal difference map and the vertical difference map, respectively. Calculate the sum of squares of the horizontal difference value to be processed and the vertical difference value to be processed, and take the square root of the sum of squares as the difference magnitude; Calculate the quotient of the vertical difference value to be processed and the horizontal difference value to be processed, and calculate the arctangent value of the obtained quotient as the difference direction angle; Divide the difference direction angle by The obtained quotient is rounded to obtain the difference direction; Based on the location index, the differential amplitude and differential direction are added to the differential amplitude map; If the position index has not completed the traversal of the difference map, the position index is adjusted, and the process jumps to the step of retrieving the horizontal difference value to be processed and the vertical difference value to be processed from the horizontal difference map and the vertical difference map respectively according to the position pointed to by the position index.

[0011] In one possible implementation, the boundary feature extraction for each difference magnitude map in the difference image sequence includes: For each difference magnitude plot, perform the following steps: Pixels are extracted from the difference magnitude map as candidate pixels through a traversal process; Along the difference direction of the candidate pixel, locate two neighboring comparison pixels of the candidate pixel on the difference magnitude map; The difference magnitude of the candidate pixel is compared with the difference magnitude of the two neighboring comparison pixels. If the difference magnitude of the candidate pixel is greater than the difference magnitude of the two neighboring comparison pixels, then the candidate pixel is retained as a boundary pixel. If the difference magnitude of the candidate pixel is less than or equal to the difference magnitude of any neighboring comparison pixel, then the candidate pixel is removed. If the traversal of the differential magnitude map is not completed, proceed to the step of extracting pixels from the differential magnitude map as candidate pixels.

[0012] In one possible implementation, locating two neighboring contrasting pixels of the candidate pixel on the difference magnitude map along the difference direction of the candidate pixel includes: Based on the differential direction angle of the candidate pixels, sub-pixel level directional offset components are obtained by decomposition. Based on the directional offset component, bilinear interpolation is performed on the adjacent pixels on both sides of the candidate pixel along the difference direction, and the interpolation difference magnitude corresponding to the comparison pixels in the neighboring areas on both sides is calculated. The interpolation difference magnitude is then used as the difference magnitude of the two comparison pixels in the neighboring areas.

[0013] In one possible implementation, performing boundary form analysis on the boundary image sequence and evaluating the stability of the weld state based on the form analysis results includes: For each boundary image in the boundary image sequence, the center coordinates of the boundary line are extracted, and a relative coordinate system is established based on the center coordinates; Based on each relative coordinate system, multiple boundary coordinates are constructed into multiple boundary complex numbers in a clockwise or counterclockwise order, and these multiple boundary complex numbers are arranged into a boundary queue according to their corresponding clockwise positions. Each boundary complex number is constructed based on a boundary coordinate. Frequency domain features are extracted from each boundary queue, and the obtained multiple frequency domain amplitudes are used to construct a first boundary form description vector; Formal analysis is performed based on multiple first boundary form description vectors and standard boundary form description vectors to obtain formal analysis results characterizing the stability of the welding state.

[0014] In one possible implementation, the frequency domain feature extraction for each boundary queue includes: Frequency domain features are extracted for each boundary queue according to the first formula, where the first formula is: In the formula, For the first Each frequency domain amplitude, This represents the total number of data items in the boundary queue. For the first A boundary complex number, It is a natural constant. The imaginary unit, Pi For the first boundary coordinates Axis coordinates For the first boundary coordinates Axis coordinates The fundamental frequency for frequency domain analysis; The formal analysis performed based on multiple first boundary form description vectors and standard boundary form description vectors to obtain formal analysis results characterizing the stability of the welding state includes: The similarity between each first boundary and the standard boundary is calculated based on the second formula, multiple first boundary form description vectors, and standard boundary form description vectors, wherein the second formula is: In the formula, For similarity, A is a standard boundary form describing the vector. The vector is described in the first boundary form; The average value of multiple similarities and the standard deviation of multiple similarities are used as the results of formal analysis.

[0015] Secondly, embodiments of the present invention provide a steel pipe high-frequency welding temperature zone monitoring device for implementing the steel pipe high-frequency welding temperature zone monitoring method as described in the first aspect or any possible implementation thereof, the steel pipe high-frequency welding temperature zone monitoring device comprising: The thermal imaging acquisition module is used to acquire a first thermal imaging sequence, wherein the first thermal imaging sequence includes multiple thermal imaging images arranged in time sequence, the thermal imaging images are obtained based on the high-frequency welding area of ​​the steel pipe, and each pixel in the thermal imaging image corresponds to a temperature observation value. The thermal imaging image preprocessing module is used to perform differential processing on each thermal imaging image in the first thermal imaging image sequence, and to construct a differential image sequence by obtaining differential amplitude maps that represent differential amplitude and differential direction. The boundary feature extraction module is used to extract boundary features from each difference magnitude map in the difference image sequence, and construct a boundary image sequence from the obtained boundary maps representing the edge lines of the temperature gradient. as well as, The welding stability assessment module is used to perform boundary form analysis on the boundary image sequence and assess the stability of the welding state based on the form analysis results.

[0016] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the computer program to implement the steps of the method as described in the first aspect or any possible implementation of the first aspect.

[0017] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any possible implementation thereof.

[0018] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: This invention discloses a method for monitoring the temperature zone during high-frequency welding of steel pipes. First, a first thermal imaging sequence is acquired, comprising multiple thermal images arranged chronologically. These thermal images are obtained based on the high-frequency welding zone of the steel pipe, with each pixel corresponding to a temperature observation value. Then, each thermal image in the first thermal imaging sequence undergoes differential processing, and the resulting differential amplitude maps, representing the differential amplitude and direction, are constructed into a differential image sequence. Next, boundary features are extracted from each differential amplitude map in the differential image sequence, and the resulting boundary maps representing the edge lines of the temperature gradient are constructed into a boundary image sequence. Finally, boundary form analysis is performed on the boundary image sequence, and the stability of the welding state is evaluated based on the form analysis results.

[0019] This invention utilizes infrared thermal imaging time-series difference and boundary extraction technology to accurately locate the edge of the temperature gradient in the high-frequency welding temperature zone of steel pipes, and to limit the welding heat-affected zone within a preset range in real time, thereby achieving precise control over the size and distribution of the stress zone and ensuring a uniform and stable welding stress field.

[0020] Based on frequency domain morphological analysis, this invention can quantify the similarity between the boundaries of multiple temperature zones and the standard steady-state morphology online. It evaluates welding process fluctuations through dual indicators of mean and standard deviation, and promptly captures process parameter shifts.

[0021] This invention ensures the consistency of temperature zone morphology from the heat input source, providing a preliminary process basis for controlling the non-roundness of straight seam welded pipes. It can effectively reduce the forming dimensional deviation caused by uneven residual stress, and improve the dimensional accuracy and welding quality of steel pipes. The non-contact monitoring does not interfere with the production rhythm and is suitable for online quality control of high-speed welding production lines. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of the high-frequency welding temperature zone monitoring method for steel pipes provided by the embodiments of the present invention; Figure 2 This is a block diagram illustrating the principle of boundary feature extraction provided in the embodiments of the present invention; Figure 3 This is a functional block diagram of the high-frequency welding temperature zone monitoring device for steel pipes provided in the embodiments of the present invention; Figure 4 This is a functional block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, so as to provide a thorough understanding of embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0025] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0026] The embodiments of the present invention will be described in detail below. This example is implemented based on the technical solution of the present invention, and provides detailed implementation methods and specific operation processes. However, the protection scope of the present invention is not limited to the following embodiments.

[0027] Figure 1 A flowchart of a method for monitoring the high-frequency welding temperature zone of steel pipes provided for embodiments of the present invention.

[0028] like Figure 1 As shown, a flowchart illustrating the implementation of the high-frequency welding temperature zone monitoring method for steel pipes provided by an embodiment of the present invention is presented below in detail: In step 101, a first thermal imaging sequence is obtained, wherein the first thermal imaging sequence includes multiple thermal imaging images arranged in chronological order, the thermal imaging images are obtained based on the high-frequency welding area of ​​the steel pipe, and each pixel in the thermal imaging image corresponds to a temperature observation value.

[0029] In step 102, each thermal image in the first thermal image sequence is subjected to differential processing, and the differential amplitude maps representing the differential amplitude and differential direction are constructed into a differential image sequence.

[0030] In some embodiments, the differential processing of each thermal image in the first thermal image sequence includes: For each thermal image in the first thermal image sequence, perform the following steps: Obtain horizontal and vertical difference data blocks; According to the predetermined sliding direction, the data block to be differentiated is extracted from the thermal image, wherein the data block to be differentiated has the same specifications as the horizontal differential data block and the vertical differential data block; The horizontal difference value is obtained by multiplying the data block to be differencing with the horizontal difference data block by its nodes and summing the results. The vertical difference data block is multiplied by the vertical difference data block by its position and summed to obtain the vertical difference value; The horizontal difference value and the vertical difference value are stored in the horizontal difference map and the vertical difference map, respectively; If the traversal of the thermal image is not completed, proceed to the step of extracting the data block to be differentiated from the thermal image according to the predetermined sliding direction; Otherwise, based on the horizontal difference map and the vertical difference map, a difference magnitude map representing the difference magnitude and the difference direction is constructed.

[0031] In some embodiments, constructing a difference magnitude map representing the difference magnitude and the difference direction based on the horizontal difference map and the vertical difference map includes: Get the location index; Based on the location pointed to by the location index, the horizontal difference value to be processed and the vertical difference value to be processed are extracted from the horizontal difference map and the vertical difference map, respectively. Calculate the sum of squares of the horizontal difference value to be processed and the vertical difference value to be processed, and take the square root of the sum of squares as the difference magnitude; Calculate the quotient of the vertical difference value to be processed and the horizontal difference value to be processed, and calculate the arctangent value of the obtained quotient as the difference direction angle; Divide the difference direction angle by The obtained quotient is rounded to obtain the difference direction; Based on the location index, the differential amplitude and differential direction are added to the differential amplitude map; If the position index has not completed the traversal of the difference map, the position index is adjusted, and the process jumps to the step of retrieving the horizontal difference value to be processed and the vertical difference value to be processed from the horizontal difference map and the vertical difference map respectively according to the position pointed to by the position index.

[0032] For example, the present invention aims to achieve online stability assessment of the welding temperature zone based on thermal imaging time-series data through gradient calculation, edge extraction and shape frequency domain analysis.

[0033] This invention utilizes an online infrared thermal imager deployed at a high-frequency welding station for steel pipes. It continuously captures images of the fusion zone and adjacent heat-affected zone in front of the welding extrusion roller at a preset sampling frame rate, resulting in a first sequence of thermal images arranged chronologically. Each thermal image in the sequence corresponds to a two-dimensional temperature field distribution at a specific moment during the welding process. The value of each pixel within the image represents the observed temperature at that spatial location. This observed value is obtained through a radiation calibration model of the infrared thermal imager, specifically by quantifying the infrared radiation energy emitted from the target surface into the corresponding temperature physical quantity based on Planck's blackbody radiation law.

[0034] In actual engineering deployments, the thermal imaging sampling frame rate can be set to match the steel pipe welding feed speed. For example, when the steel pipe feed speed is 15m / min, the sampling frame rate can be set to 25fps, which can ensure that the axial displacement of the weld between adjacent frames is less than 10mm, and realize continuous and complete monitoring of the temperature zone morphology. The temperature range usually covers 200℃~1600℃ to fully include the temperature distribution of the entire temperature range of the base material, including the ambient temperature zone, heat-affected zone, and melting zone.

[0035] In terms of thermal imaging differential processing and differential image sequence construction, this invention performs spatial differential operation frame-by-frame on each thermal imaging image in the first thermal imaging image sequence to extract the gradient distribution features of the temperature field. The final output is a differential image sequence composed of differential amplitude maps, where each differential amplitude map simultaneously contains the differential amplitude and differential direction information for each pixel location. The core principle of differential processing is to perform convolution operations on the temperature field image using spatial differential operators, enhancing edge regions of abrupt temperature changes and suppressing background noise in regions of uniform temperature, thereby highlighting the boundary features of the welding temperature zone.

[0036] The specific execution process is as follows: Predefined horizontal and vertical difference data blocks are used, both being difference convolution kernels of the same size, to extract the gradient components of the temperature field in the horizontal and vertical directions, respectively. For example, a 3×3 difference kernel can be used, where the horizontal difference kernel has a positive and negative difference weight distribution along the horizontal direction, and the vertical difference kernel has a positive and negative difference weight distribution along the vertical direction, to solve the first-order spatial differential of the temperature field in a numerical approximation manner.

[0037] According to the predetermined sliding traversal rules (usually using row-first raster scanning order with a sliding step size of 1 pixel), the data blocks to be differentiated with the same specifications as the differential data blocks are extracted sequentially from the current thermal image, that is, a local temperature data window of the corresponding size is cropped with one pixel as the center each time.

[0038] The current data block to be differentiated is multiplied by the horizontal difference data block and then summed to obtain the horizontal difference value at the center pixel position, which corresponds to the gradient component of the temperature field in the horizontal direction. Similarly, the current data block to be differentiated is multiplied by the vertical difference data block and then summed to obtain the vertical difference value at the center pixel position, which corresponds to the gradient component of the temperature field in the vertical direction.

[0039] The horizontal and vertical difference values ​​of all pixel locations are stored in their respective locations to generate the horizontal and vertical difference maps corresponding to the entire thermal image.

[0040] After completing the full image traversal, the gradient magnitude and gradient direction of each pixel are calculated based on the horizontal and vertical difference maps, and a difference magnitude map containing both magnitude and direction information is constructed.

[0041] The calculation of the differential magnitude and differential direction follows the mathematical definition of the two-dimensional image gradient, and the specific execution process is as follows: Traverse all pixel positions in the difference map, locate the current pixel to be processed by its position index, and read the horizontal and vertical difference values ​​at that position from the horizontal and vertical difference maps respectively. Take the square root of the sum of the squares of the two difference values ​​and use the result as the difference amplitude of the pixel. This amplitude represents the intensity of the temperature gradient at that position. The larger the amplitude, the more drastic the temperature change, and the greater the probability of corresponding to the temperature zone boundary. Calculate the ratio of the vertical difference value to the horizontal difference value, and take the arctangent of this ratio to obtain the difference direction angle. This angle represents the spatial orientation of the temperature gradient, that is, the direction of the fastest temperature rise.

[0042] To reduce the complexity of subsequent calculations, the continuous differential direction angles are discretized: the differential direction angle is divided by π / 4 and rounded to the nearest integer, resulting in eight discrete differential direction categories from 0 to 7, corresponding to eight direction intervals: 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°. Finally, the differential magnitude of each pixel is stored in the image matrix along with the corresponding discrete differential direction, resulting in a complete differential magnitude map. After frame-by-frame processing, all differential magnitude maps are arranged in temporal order to form a differential image sequence.

[0043] In step 103, boundary features are extracted from each differential amplitude map in the differential image sequence, and the obtained boundary maps representing the edge lines of the temperature gradient are constructed into a boundary image sequence.

[0044] In some implementations, the step of extracting boundary features from each difference magnitude map in the difference image sequence includes: For each difference magnitude plot, perform the following steps: Pixels are extracted from the difference magnitude map as candidate pixels through a traversal process; Along the difference direction of the candidate pixel, locate two neighboring comparison pixels of the candidate pixel on the difference magnitude map; The difference magnitude of the candidate pixel is compared with the difference magnitude of the two neighboring comparison pixels. If the difference magnitude of the candidate pixel is greater than the difference magnitude of the two neighboring comparison pixels, then the candidate pixel is retained as a boundary pixel. If the difference magnitude of the candidate pixel is less than or equal to the difference magnitude of any neighboring comparison pixel, then the candidate pixel is removed. If the traversal of the differential magnitude map is not completed, proceed to the step of extracting pixels from the differential magnitude map as candidate pixels.

[0045] In some implementations, locating two neighboring contrasting pixels of the candidate pixel on the difference magnitude map along the difference direction of the candidate pixel includes: Based on the differential direction angle of the candidate pixels, sub-pixel level directional offset components are obtained by decomposition. Based on the directional offset component, bilinear interpolation is performed on the adjacent pixels on both sides of the candidate pixel along the difference direction, and the interpolation difference magnitude corresponding to the comparison pixels in the neighboring areas on both sides is calculated. The interpolation difference magnitude is then used as the difference magnitude of the two comparison pixels in the neighboring areas.

[0046] For example, in terms of boundary feature extraction and boundary image sequence construction, this invention extracts boundary features from each differential amplitude map in the differential image sequence, filters out the true temperature gradient edge pixels using a boundary processing algorithm, generates a single-pixel-width boundary map, and arranges all boundary maps in chronological order to form a boundary image sequence. The core principle of this step is that the true temperature zone boundary corresponds to the local peak point of the gradient amplitude. Only the pixel with the largest amplitude in the gradient direction is retained, which can eliminate redundant responses near the edge, achieve fine-grained edge localization, and improve the accuracy of subsequent shape analysis.

[0047] The boundary extraction process for a single difference magnitude map is as follows: Traverse the entire differential amplitude map pixel by pixel, and use the currently traversed pixel as a candidate boundary pixel.

[0048] Along the discrete difference direction corresponding to the candidate pixel, locate a neighboring comparison pixel on each side before and after it for local amplitude comparison.

[0049] To improve edge positioning accuracy and avoid quantization errors in discrete pixel grids, sub-pixel level interpolation is used for neighboring comparison pixels: First, the sub-pixel offset components in the horizontal and vertical directions are obtained by decomposing the differential direction angle of the candidate pixels; then, the offset is performed by one unit step along the positive and negative gradient directions, and bilinear interpolation is performed on the four integer pixels around the offset position to calculate the interpolation difference magnitude of the two sub-pixel positions, which is used as the equivalent difference magnitude of the two neighboring comparison pixels.

[0050] For example, if the differential direction angle of a candidate pixel is 45°, offset this component along the positive gradient direction to obtain a sub-pixel coordinate, and offset it along the negative direction to obtain another sub-pixel coordinate. Perform bilinear interpolation on the adjacent pixels around the two coordinates to obtain the contrast amplitude on both sides.

[0051] The differential magnitude of the candidate pixel is compared with the interpolation magnitude of the two neighboring contrasting pixels. If the magnitude of the candidate pixel is greater than the contrasting magnitudes on both sides, it indicates that the pixel is a local peak in the gradient direction, and it is determined to be a boundary pixel and retained. If the magnitude of the candidate pixel is less than or equal to the contrasting magnitude on either side, it is determined to be a non-boundary pixel and discarded.

[0052] After traversing and filtering the entire differential amplitude map, a binary boundary map containing only boundary pixels is obtained, where the boundary pixels represent the edge lines of the temperature gradient in the welding temperature zone. After frame-by-frame processing, all boundary maps are arranged in time sequence to form a boundary image sequence.

[0053] In step 104, boundary form analysis is performed on the boundary image sequence, and the stability of the welding state is evaluated based on the form analysis results.

[0054] In some embodiments, performing boundary form analysis on the boundary image sequence and evaluating the stability of the welding state based on the form analysis results includes: For each boundary image in the boundary image sequence, the center coordinates of the boundary line are extracted, and a relative coordinate system is established based on the center coordinates; Based on each relative coordinate system, multiple boundary coordinates are constructed into multiple boundary complex numbers in a clockwise or counterclockwise order, and these multiple boundary complex numbers are arranged into a boundary queue according to their corresponding clockwise positions. Each boundary complex number is constructed based on a boundary coordinate. Frequency domain features are extracted from each boundary queue, and the obtained multiple frequency domain amplitudes are used to construct a first boundary form description vector; Formal analysis is performed based on multiple first boundary form description vectors and standard boundary form description vectors to obtain formal analysis results characterizing the stability of the welding state.

[0055] In some implementations, the frequency domain feature extraction for each boundary queue includes: Frequency domain features are extracted for each boundary queue according to the first formula, where the first formula is: In the formula, For the first Each frequency domain amplitude, This represents the total number of data items in the boundary queue. For the first A boundary complex number, It is a natural constant. The imaginary unit, Pi For the first boundary coordinates Axis coordinates For the first boundary coordinates Axis coordinates The fundamental frequency for frequency domain analysis; The formal analysis performed based on multiple first boundary form description vectors and standard boundary form description vectors to obtain formal analysis results characterizing the stability of the welding state includes: The similarity between each first boundary and the standard boundary is calculated based on the second formula, multiple first boundary form description vectors, and standard boundary form description vectors, wherein the second formula is: In the formula, For similarity, A is a standard boundary form describing the vector. The vector is described in the first boundary form; The average value of multiple similarities and the standard deviation of multiple similarities are used as the results of formal analysis.

[0056] For example, in terms of boundary form analysis and welding state stability assessment, this invention quantitatively analyzes the boundary shape of a time-series boundary image sequence, extracts the morphological features of the boundary through frequency domain analysis, and compares them with the boundary features of standard steady-state welding, thereby quantitatively assessing the stability of the welding process. This method possesses translation and rotation invariance, effectively eliminating interference such as steel pipe feed jitter and field of view shift, and accurately reflecting the fluctuations in the temperature zone morphology itself.

[0057] The specific implementation process is as follows: For each boundary map, the coordinate set of all pixels along the boundary line is first extracted, the center coordinates of the boundary contour are calculated, and a relative coordinate system is established with these center coordinates as the origin. All boundary coordinates are then converted into coordinate values ​​relative to the center. This step eliminates the positional deviation caused by the overall translation of the temperature region, achieving translational normalization of shape features.

[0058] In a relative coordinate system, all coordinate points on the boundary contour are traversed in a fixed clockwise or counterclockwise order. The x-coordinate of each coordinate point is taken as the real part and the y-coordinate as the imaginary part to construct the corresponding boundary complex number. All boundary complex numbers are arranged in the traversal order to form an ordered boundary queue. , where n is the index of the boundary point (n=0,1,...,N-1), and N is the total number of pixels on the boundary.

[0059] Frequency domain feature extraction is performed on each boundary queue. The spatial boundary shape sequence is converted into frequency domain amplitude features by discrete Fourier transform. The first k frequency domain amplitudes are selected to form the first boundary form description vector.

[0060] Frequency domain feature extraction is calculated according to the first formula: In the formula, For the first Each frequency domain amplitude, This represents the total number of data items in the boundary queue. For the first A boundary complex number, It is a natural constant. The imaginary unit, Pi For the first boundary coordinates Axis coordinates For the first boundary coordinates Axis coordinates This is the fundamental angular frequency for frequency domain analysis.

[0061] In practical applications, the first 10 to 20 low-frequency amplitudes are usually selected to form the description vector, which can fully characterize the overall shape features of the temperature zone boundary, while filtering out high-frequency noise and interference from small contour fluctuations. This description vector has rotation invariance and scaling invariance, and only reflects the geometric morphological features of the boundary.

[0062] The similarity between the first boundary form description vector corresponding to each frame and the pre-stored standard boundary form description vector is calculated. Based on the similarity statistics of multiple frames, the stability evaluation result of the welding state is obtained. The standard boundary form description vector is obtained by averaging multiple frames under steady-state welding conditions, representing the temperature zone boundary morphology characteristics when the process parameters are optimal.

[0063] Similarity calculation follows the second formula, which uses cosine similarity to measure the degree of morphological similarity between two feature vectors: In the formula, The similarity score ranges from [0,1]; A is a standard boundary form describing vector. The first boundary form describes the vector; the numerator is the dot product of the two vectors, and the denominator is the product of the magnitudes of the two vectors.

[0064] A similarity score closer to 1 indicates better consistency between the current temperature zone boundary morphology and the steady-state standard morphology. The average similarity and standard deviation of all frames in the time series are then used as formal analysis results: the average reflects the overall deviation of the current welding condition from the steady-state standard; a closer average to 1 indicates better welding quality. The standard deviation reflects the dynamic fluctuation of the welding process; a smaller standard deviation indicates a more stable temperature zone morphology and more stable welding process parameters. When the standard deviation exceeds a preset threshold, or the average is below a preset threshold, an abnormal welding condition is determined, triggering a warning signal. The present invention provides an implementation method for monitoring the temperature zone of high-frequency welding of steel pipes. First, a first thermal imaging sequence is acquired, comprising multiple thermal imaging images arranged chronologically. These thermal imaging images are obtained based on the high-frequency welding zone of the steel pipe, and each pixel in the thermal imaging image corresponds to a temperature observation value. Then, each thermal imaging image in the first thermal imaging sequence undergoes differential processing, and the resulting differential amplitude maps, representing the differential amplitude and differential direction, are constructed into a differential image sequence. Next, boundary features are extracted from each differential amplitude map in the differential image sequence, and the resulting boundary maps representing the edge lines of the temperature gradient are constructed into a boundary image sequence. Finally, boundary form analysis is performed on the boundary image sequence, and the stability of the welding state is evaluated based on the form analysis results.

[0065] This invention utilizes infrared thermal imaging time-series difference and boundary extraction technology to accurately locate the edge of the temperature gradient in the high-frequency welding temperature zone of steel pipes, and to limit the welding heat-affected zone within a preset range in real time, thereby achieving precise control over the size and distribution of the stress zone and ensuring a uniform and stable welding stress field.

[0066] Based on frequency domain morphological analysis, this invention can quantify the similarity between the boundaries of multiple temperature zones and the standard steady-state morphology online. It evaluates welding process fluctuations through dual indicators of mean and standard deviation, and promptly captures process parameter shifts.

[0067] This invention ensures the consistency of temperature zone morphology from the heat input source, providing a preliminary process basis for controlling the non-roundness of straight seam welded pipes. It can effectively reduce the forming dimensional deviation caused by uneven residual stress, and improve the dimensional accuracy and welding quality of steel pipes. The non-contact monitoring does not interfere with the production rhythm and is suitable for online quality control of high-speed welding production lines.

[0068] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0069] The following are embodiments of the apparatus of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0070] Figure 3 This is a functional block diagram of the high-frequency welding temperature zone monitoring device for steel pipes provided in the embodiments of the present invention, with reference to... Figure 3 The high-frequency welding temperature monitoring device for steel pipes includes: a thermal imaging acquisition module 301, a thermal imaging preprocessing module 302, a boundary feature extraction module 303, and a welding stability assessment module 304, wherein: The thermal imaging acquisition module 301 is used to acquire a first thermal imaging sequence, wherein the first thermal imaging sequence includes multiple thermal imaging images arranged in time sequence, the thermal imaging images are obtained based on the high-frequency welding area of ​​the steel pipe, and each pixel in the thermal imaging image corresponds to a temperature observation value. The thermal imaging image preprocessing module 302 is used to perform differential processing on each thermal imaging image in the first thermal imaging image sequence, and to construct a differential image sequence by obtaining differential amplitude maps that represent differential amplitude and differential direction. The boundary feature extraction module 303 is used to extract boundary features from each difference magnitude map in the difference image sequence, and construct a boundary image sequence from the obtained boundary maps representing the edge lines of the temperature gradient. The welding stability assessment module 304 is used to perform boundary form analysis on the boundary image sequence and assess the stability of the welding state based on the form analysis results.

[0071] Figure 4 This is a functional block diagram of the electronic device provided in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device 4 in this embodiment includes a processor 400 and a memory 401, wherein the memory 401 stores a computer program 402 that can run on the processor 400. When the processor 400 executes the computer program 402, it implements the steps of the various high-frequency welding temperature zone monitoring methods and embodiments for steel pipes described above, for example... Figure 1 Steps 101 to 104 are shown.

[0072] For example, the computer program 402 may be divided into one or more modules / units, which are stored in the memory 401 and executed by the processor 400 to complete the present invention.

[0073] The electronic device 4 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The electronic device 4 may include, but is not limited to, a processor 400 and a memory 401. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 4 may also include input / output devices, network access devices, buses, etc.

[0074] The processor 400 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0075] The memory 401 can be an internal storage unit of the electronic device 4, such as a hard disk or memory. The memory 401 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 401 can include both internal and external storage units of the electronic device 4. The memory 401 is used to store the computer program 402 and other programs and data required by the electronic device 4. The memory 401 can also be used to temporarily store data that has been output or will be output.

[0076] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0077] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0078] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0079] In the embodiments provided by this invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0080] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0081] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0082] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-described embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various methods and apparatus embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0083] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for monitoring the temperature zone during high-frequency welding of steel pipes, characterized in that, include: A first thermal imaging sequence is obtained, which includes multiple thermal imaging images arranged in time sequence. The thermal imaging images are obtained based on the high-frequency welding area of ​​the steel pipe, and each pixel in the thermal imaging image corresponds to a temperature observation value. Each thermal image in the first thermal image sequence is subjected to differential processing, and the differential amplitude maps representing the differential amplitude and differential direction are constructed into a differential image sequence. Boundary features are extracted from each differential amplitude map in the differential image sequence, and the obtained boundary maps representing the edge lines of the temperature gradient are constructed into a boundary image sequence. Boundary form analysis is performed on the boundary image sequence, and the stability of the welding state is evaluated based on the form analysis results.

2. The method for monitoring the high-frequency welding temperature zone of steel pipes according to claim 1, characterized in that, The differential processing of each thermal image in the first thermal image sequence includes: For each thermal image in the first thermal image sequence, perform the following steps: Obtain horizontal and vertical difference data blocks; According to the predetermined sliding direction, the data block to be differentiated is extracted from the thermal image, wherein the data block to be differentiated has the same specifications as the horizontal differential data block and the vertical differential data block; The horizontal difference value is obtained by multiplying the data block to be differencing with the horizontal difference data block by its nodes and summing the results. The vertical difference data block is multiplied by the vertical difference data block by its position and summed to obtain the vertical difference value; The horizontal difference value and the vertical difference value are stored in the horizontal difference map and the vertical difference map, respectively; If the traversal of the thermal image is not completed, proceed to the step of extracting the data block to be differentiated from the thermal image according to the predetermined sliding direction; Otherwise, based on the horizontal difference map and the vertical difference map, a difference magnitude map representing the difference magnitude and the difference direction is constructed.

3. The method for monitoring the high-frequency welding temperature zone of steel pipes according to claim 2, characterized in that, The step of constructing a difference magnitude map representing the difference magnitude and the difference direction based on the horizontal difference map and the vertical difference map includes: Get the location index; Based on the location pointed to by the location index, the horizontal difference value to be processed and the vertical difference value to be processed are extracted from the horizontal difference map and the vertical difference map, respectively. Calculate the sum of squares of the horizontal difference value to be processed and the vertical difference value to be processed, and take the square root of the sum of squares as the difference magnitude; Calculate the quotient of the vertical difference value to be processed and the horizontal difference value to be processed, and calculate the arctangent value of the obtained quotient as the difference direction angle; Divide the difference direction angle by The obtained quotient is rounded to obtain the difference direction; Based on the location index, the differential amplitude and differential direction are added to the differential amplitude map; If the position index has not completed the traversal of the difference map, the position index is adjusted, and the process jumps to the step of retrieving the horizontal difference value to be processed and the vertical difference value to be processed from the horizontal difference map and the vertical difference map respectively according to the position pointed to by the position index.

4. The method for monitoring the high-frequency welding temperature zone of steel pipes according to claim 1, characterized in that, The step of extracting boundary features from each difference magnitude map in the difference image sequence includes: For each difference magnitude plot, perform the following steps: Pixels are extracted from the difference magnitude map as candidate pixels through a traversal process; Along the difference direction of the candidate pixel, locate two neighboring comparison pixels of the candidate pixel on the difference magnitude map; The difference magnitude of the candidate pixel is compared with the difference magnitude of the two neighboring comparison pixels. If the difference magnitude of the candidate pixel is greater than the difference magnitude of the two neighboring comparison pixels, then the candidate pixel is retained as a boundary pixel. If the difference magnitude of the candidate pixel is less than or equal to the difference magnitude of any neighboring comparison pixel, then the candidate pixel is removed. If the traversal of the differential magnitude map is not completed, proceed to the step of extracting pixels from the differential magnitude map as candidate pixels.

5. The method for monitoring the high-frequency welding temperature zone of steel pipes according to claim 4, characterized in that, The step of locating two neighboring comparison pixels of the candidate pixel on the difference magnitude map along the difference direction of the candidate pixel includes: Based on the differential direction angle of the candidate pixels, sub-pixel level directional offset components are obtained by decomposition. Based on the directional offset component, bilinear interpolation is performed on the adjacent pixels on both sides of the candidate pixel along the difference direction, and the interpolation difference magnitude corresponding to the comparison pixels in the neighboring areas on both sides is calculated. The interpolation difference magnitude is then used as the difference magnitude of the two comparison pixels in the neighboring areas.

6. The method for monitoring the high-frequency welding temperature zone of steel pipes according to any one of claims 1-5, characterized in that, The step of performing boundary form analysis on the boundary image sequence and evaluating the stability of the welding state based on the form analysis results includes: For each boundary image in the boundary image sequence, the center coordinates of the boundary line are extracted, and a relative coordinate system is established based on the center coordinates; Based on each relative coordinate system, multiple boundary coordinates are constructed into multiple boundary complex numbers in a clockwise or counterclockwise order, and these multiple boundary complex numbers are arranged into a boundary queue according to their corresponding clockwise positions. Each boundary complex number is constructed based on a boundary coordinate. Frequency domain features are extracted from each boundary queue, and the obtained multiple frequency domain amplitudes are used to construct a first boundary form description vector; Formal analysis is performed based on multiple first boundary form description vectors and standard boundary form description vectors to obtain formal analysis results characterizing the stability of the welding state.

7. The method for monitoring the high-frequency welding temperature zone of steel pipes according to claim 6, characterized in that, The frequency domain feature extraction for each boundary queue includes: Frequency domain features are extracted for each boundary queue according to the first formula, where the first formula is: In the formula, For the first Each frequency domain amplitude, This represents the total number of data items in the boundary queue. For the first A boundary complex number, It is a natural constant. The imaginary unit, Pi For the first boundary coordinates Axis coordinates For the first boundary coordinates Axis coordinates The fundamental frequency for frequency domain analysis; The formal analysis performed based on multiple first boundary form description vectors and standard boundary form description vectors to obtain formal analysis results characterizing the stability of the welding state includes: The similarity between each first boundary and the standard boundary is calculated based on the second formula, multiple first boundary form description vectors, and standard boundary form description vectors, wherein the second formula is: In the formula, For similarity, A is a standard boundary form describing the vector. The vector is described in the first boundary form; The average value of multiple similarities and the standard deviation of multiple similarities are used as the results of formal analysis.

8. A high-frequency welding temperature monitoring device for steel pipes, characterized in that, For implementing the high-frequency welding temperature zone monitoring method for steel pipes as described in any one of claims 1-7, the high-frequency welding temperature zone monitoring device for steel pipes comprises: The thermal imaging acquisition module is used to acquire a first thermal imaging sequence, wherein the first thermal imaging sequence includes multiple thermal imaging images arranged in time sequence, the thermal imaging images are obtained based on the high-frequency welding area of ​​the steel pipe, and each pixel in the thermal imaging image corresponds to a temperature observation value. The thermal imaging image preprocessing module is used to perform differential processing on each thermal imaging image in the first thermal imaging image sequence, and to construct a differential image sequence by obtaining differential amplitude maps that represent differential amplitude and differential direction. The boundary feature extraction module is used to extract boundary features from each difference magnitude map in the difference image sequence, and construct a boundary image sequence from the obtained boundary maps representing the edge lines of the temperature gradient. as well as, The welding stability assessment module is used to perform boundary form analysis on the boundary image sequence and assess the stability of the welding state based on the form analysis results.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7 above.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7 above.

Images