Deviation analysis and detection system and method for high-toughness steel member

Abstract

The invention discloses a deviation analysis and detection system and method for a high-toughness steel component, and relates to the technical field of deviation analysis. The deviation analysis and detection system comprises a data acquisition module, a data processing module, a deviation analysis module, a decision application module and a visual control interface. According to the deviation analysis and detection system and method for the high-toughness steel component, the deviation analysis and detection system is arranged, data are extracted, a simulation analysis model is established, image data are extracted through the simulation analysis model and then analyzed, and the deviation abnormal problem in the welding process is determined in combination with a threshold value; the method comprises the following steps: extracting stress data based on a current deviation abnormal problem, tracing a welding position parameter, calculating an adjustment parameter according to the welding position parameter, and performing optimization operation on a simulation analysis model through the adjustment parameter, so that deviation analysis operation can be effectively completed; and the regulation and control compensation operation can be adaptively completed according to different conditions, so that the detection efficiency and stability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of deviation analysis, and in particular to a deviation analysis detection system and method for high-toughness steel components. Background Art

[0002] High-toughness steel components are widely used in numerous fields, including construction, bridges, and machinery manufacturing. Their quality is directly related to the safety and reliability of the entire engineering structure. During actual production and use, high-toughness steel components may exhibit dimensional and shape deviations due to factors such as manufacturing processes, fluctuations in material properties, and transportation and installation. If these deviations exceed the allowable range, they will seriously affect the assembly accuracy, load-bearing capacity, and service life of the steel components.

[0003] The reference patent name is: A steel structure in-depth design data management system (patent publication number: CN115423315A, patent publication date: 2022-12-02), including a data entry module to enter in-depth design data and node cycle data; a data acquisition module to collect unit cycle data; a calculation unit to calculate engineering quantitative data; a first processing unit to process the engineering calculation data; a comparison unit to subtract the engineering quantitative data from the engineering calculation data to obtain an engineering deviation value, and generate a detection instruction and a display instruction; the instruction generation unit inputs each three-dimensional component and the corresponding expected three-dimensional component into the deviation analysis model according to the detection instruction to obtain the component deviation value, and generates a marking instruction; the display unit generates display content according to the display instruction; the marking unit marks and displays the unit quantitative data and front-line personnel in the display content according to the marking instruction; the suggestion generation unit obtains work suggestions based on the component deviation value processing.

[0004] Based on the description in the above-mentioned documents, during the welding process of existing high-toughness steel components, the applied pressure during welding varies with the welding position, causing the pressure on the product to change, resulting in position or deformation problems of the steel components. Even after the welding is completed, there are still deviation problems, which exceed the normal range. For this reason, the present invention provides a deviation analysis and detection system and method for high-toughness steel components. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a deviation analysis and detection system and method for high-toughness steel components, which solves the problem that during the welding process of existing high-toughness steel components, the pressure applied during welding changes with the different welding positions, causing the pressure on the product to change, resulting in position or deformation of the steel components. Even after the welding is completed, there is still a problem of deviation.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A deviation analysis and detection system for high-toughness steel components, comprising:

[0007] A data acquisition module performs welding operations on high-toughness steel components, and sets image acquisition equipment and force sensors around the high-toughness steel components to realize data acquisition operations;

[0008] The data processing module pre-processes the raw data collected by the data acquisition module to complete the data classification;

[0009] The deviation analysis module extracts data to establish a simulation analysis model. The simulation analysis model is used to analyze the extracted image data and determine the deviation anomaly in the welding process based on the threshold. Based on the current deviation anomaly, the force data is extracted and the welding position parameters are traced. The adjustment parameters are calculated based on the welding position parameters. The simulation analysis model is optimized by adjusting the parameters, and then real-time data is introduced into the simulation analysis model for analysis and an adjustment strategy is generated to achieve compensation for the welding operation.

[0010] The decision-making application module uses the generated adjustment strategy to transmit it to the corresponding equipment to implement the control operation;

[0011] The visual control interface displays the processed data and generated strategies in the form of text and charts.

[0012] Preferably, the data processing module performs preprocessing operations on the raw data collected by the data collection module as follows:

[0013] The filtering algorithm is used to remove noise points from the collected parameter data, and the collected image data is enhanced and denoised;

[0014] Then set up a classification table to classify the collected parameter data;

[0015] The collected image data is processed by image stitching algorithm to form complete image data for analysis.

[0016] Preferably, the classification table classifies the collected parameter data as follows:

[0017] The classification table sets column and row titles, and the structure of the column title is device name + parameter category name, and the structure of the row title is timestamp;

[0018] Then, the collected parameter data content is matched with the content of the column header and the row header, and the parameter data with the same content is added to the classification table;

[0019] The intersection of the vertical position of the column title and the horizontal position of the row title is the position of the corresponding parameter data compensation, and finally a classification table with parameter results is formed.

[0020] Preferably, the image stitching algorithm implements the following image data processing operations:

[0021] According to the image acquisition sequence, the moving distance in each image acquisition process is determined to be L, and the horizontal distance of the image is H1, and the vertical distance is H2;

[0022] Starting from the first image, a segmentation line is set at the boundary of one side of the moving direction of the first image, and the distance between the segmentation line segment and the boundary of the moving direction of the image is H1-L or H2-L, and then the image segmentation operation is performed according to the segmentation line segment;

[0023] The image data before the segmentation is retained, and then the image data is spliced, such as connecting the pre-sequence part of the first image data and the pre-sequence part of the second image data, and sequentially splicing the two corner points of the first image in the moving direction with the two corner points of the second image relative to the first image, and finally forming complete image data.

[0024] Preferably, the operation of extracting data and establishing a simulation analysis model in the deviation analysis module is:

[0025] A preliminary model is established by extracting the equipment parameters in actual operation, and then historical data is introduced into the preliminary model to form a simulation analysis model that can be used for simulation;

[0026] The simulation analysis model extracts the corresponding image data and parameter data through the changes in welding position, and then performs analysis operations.

[0027] Preferably, the operation of extracting and analyzing the image data using the simulation analysis model in the deviation analysis module is:

[0028] After extracting the complete image data, the image data is grayscale processed, and comparison points are set at equal intervals on the image data. The characteristic parts in the image are determined by the grayscale value changes of the comparison points. Then, the known historical feature data is extracted and matched with the determined feature parts to obtain the interpretation of the feature parts.

[0029] After the welding position is determined, the corresponding image data is analyzed and the deviation and abnormality problems in the welding process are determined by combining the threshold.

[0030] Preferably, the deviation analysis module determines the deviation abnormality problem in the welding process in combination with the threshold value by:

[0031] Extract the image data of the front side of the weld and perform analysis operations, and use the corner points of the weld alignment position of the steel components in the image data as reference points;

[0032] The normal image data before welding is extracted, and then the image data during welding is extracted. The two images are overlaid based on the corresponding corner points of the images. The image data on the upper layer is blurred to maintain the required features of the upper and lower image data.

[0033] Then, a reference coordinate axis is established, the position of the reference point is calculated, and the historical safety threshold is introduced to compare with the calculated result.

[0034] Preferably, the calculation operation of establishing the reference coordinate axis to realize the reference point change is:

[0035] The lower left corner of the image data is used as the coordinate axis origin, and the X axis is established from the origin along the horizontal boundary of the image data, and the Y axis is established from the origin along the vertical boundary of the image data;

[0036] Then, based on the reference coordinate axes, the coordinates of the reference points are determined, and the coordinates of reference point A are set to (x1, y1), the coordinates of reference point B are set to (x2, y2), the coordinates of reference point C are set to (x3, y3), and the coordinates of reference point D are set to (x4, y4);

[0037] When x1=x2 and y1=y2, and x3=x4 and y3=y4, then there is no abnormality in the welding operation of the current steel component;

[0038] When x1≠x2 or y1≠y2, or x3≠x4 or y3≠y4, the welding operation of the current steel component is abnormal, and then the calculated deviation value is compared with the safety threshold;

[0039] The calculation formula for the deviation value is:

[0040]

[0041] And K1 belongs to the distance between reference point A and reference point B, while K2 belongs to the distance between reference point A and reference point B;

[0042] The safety threshold interval is set to [M, N], and when K1 or K2 both belong to [M, N], the current value is normal. Conversely, if either K1 or K2 does not belong to [M, N], the current value is abnormal.

[0043] Preferably, the deviation analysis module extracts force data based on the current deviation anomaly problem and performs the operation of tracing welding position parameters as follows:

[0044] If the value is abnormal, the pressure applied at the current welding position causes the steel structure to deviate, and the current position is marked;

[0045] Then, the current pressure on the steel structure is extracted as F, and a compensation strategy is generated. The force provided to the subsequent jacking equipment is the jacking force f in the opposite direction, and F=f.

[0046] The present invention also discloses a deviation analysis and detection method for high-toughness steel components, which specifically includes the following steps:

[0047] S1. Real-time data collection and pre-processing of high-toughness steel component welding operations;

[0048] S2. Then extract historical data to establish a simulation analysis model, implement the search operation for abnormal deviation problems in the welding process, generate an adjustment strategy to achieve optimization compensation of the simulation analysis model, and introduce real-time data into the simulation analysis model for analysis;

[0049] S3. Transmit the final instruction to the corresponding device to implement the control and compensation operation.

[0050] The present invention provides a deviation analysis and detection system and method for high-toughness steel components. Compared with the existing technology, it has the following advantages:

[0051] 1. The deviation analysis and detection system and method of the high-toughness steel component are provided with a deviation analysis and detection system, which extracts data to establish a simulation analysis model, uses the simulation analysis model to analyze the extracted image data, and determines the deviation abnormality problem in the welding process in combination with the threshold. Based on the current deviation abnormality problem, the force data is extracted and the welding position parameters are traced, and the adjustment parameters are calculated based on the welding position parameters. By adjusting the parameters and optimizing the simulation analysis model, the deviation analysis operation can be effectively completed, and the regulation and compensation operation can be adaptively completed according to different situations, thereby improving the efficiency and stability of system detection.

[0052] 2. The deviation analysis and detection system and method of high-toughness steel components uses a filtering algorithm to remove noise points from the collected parameter data, and performs image enhancement and denoising on the collected image data. A classification table is then set to classify the collected parameter data, and the collected image data is processed by an image stitching algorithm to form complete image data for analysis, thereby enabling pre-order classification operations on the data, allowing for faster and more accurate data extraction operations in subsequent use, making subsequent use more convenient.

[0053] 3. The deviation analysis and detection system and method of the high-toughness steel component extracts the image data of the front side of the welding point for analysis, and uses the corner points of the steel component welding alignment position in the image data as reference points, and extracts the normal image data before welding, and then extracts the image data during welding, and implements the overlay operation of the two images with the corresponding corner point positions of the images. The image data on the upper layer is blurred to maintain the appearance of the required features of the upper and lower image data, and then establishes a reference coordinate axis, calculates the position of the reference point, and introduces a historical safety threshold for comparison with the calculated result. In this way, deviation detection and real-time analysis operations can be completed accurately, and abnormal conditions can be obtained and compensated in real time in subsequent processing to avoid deviation problems during welding. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a principle block diagram of the deviation analysis and detection system of the present invention;

[0055] Figure 2 This is an operational flow chart of the data processing module of the present invention;

[0056] Figure 3 This is an operational flow chart of data extraction and analysis of the present invention;

[0057] Figure 4 An operational flow chart for determining abnormal welding deviations according to the present invention;

[0058] Figure 5 Schematic diagram of the reference coordinate axis of the present invention. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0060] See also Figure 1-Figure 5 , the present invention provides two technical solutions:

[0061] Embodiment 1: A deviation analysis and detection system for high-toughness steel components, comprising:

[0062] A data acquisition module performs welding operations on high-toughness steel components, and sets image acquisition equipment and force sensors around the high-toughness steel components to realize data acquisition operations;

[0063] The data processing module pre-processes the raw data collected by the data acquisition module to complete the data classification;

[0064] The deviation analysis module extracts data to establish a simulation analysis model. The simulation analysis model is used to analyze the extracted image data and determine the deviation anomaly in the welding process based on the threshold. Based on the current deviation anomaly, the force data is extracted and the welding position parameters are traced. The adjustment parameters are calculated based on the welding position parameters. The simulation analysis model is optimized by adjusting the parameters, and then real-time data is introduced into the simulation analysis model for analysis and an adjustment strategy is generated to achieve compensation for the welding operation.

[0065] The decision-making application module uses the generated adjustment strategy to transmit it to the corresponding equipment to implement the control operation;

[0066] The visual control interface displays the processed data and generated strategies in the form of text and charts.

[0067] Among them, by setting up a deviation analysis detection system, extracting data to establish a simulation analysis model, using the simulation analysis model to analyze the image data after extraction, and combining with the threshold to determine the deviation abnormality problem in the welding process, based on the current deviation abnormality problem, the force data is extracted and the welding position parameters are traced, and the adjustment parameters are calculated based on the welding position parameters. By adjusting the parameters to optimize the simulation analysis model, the deviation analysis operation can be effectively completed, and the control compensation operation can be adaptively completed according to different situations, thereby improving the efficiency and stability of system detection.

[0068] In the embodiment of the present invention, the data processing module performs preprocessing operations on the raw data collected by the data collection module as follows:

[0069] The filtering algorithm is used to remove noise points from the collected parameter data, and the collected image data is enhanced and denoised;

[0070] Then set up a classification table to classify the collected parameter data;

[0071] The collected image data is processed by image stitching algorithm to form complete image data for analysis.

[0072] Among them, the noise points of the collected parameter data are removed by using a filtering algorithm, and the collected image data are enhanced and denoised. Then, a classification table is set to classify the collected parameter data, and the collected image data is processed by relying on the image stitching algorithm to form complete image data for analysis, so that the data can be classified in advance, which is convenient for subsequent use to complete data extraction operations faster and more accurately, making subsequent use more convenient.

[0073] In the embodiment of the present invention, the classification table performs the following operations to classify the collected parameter data:

[0074] The classification table sets column and row titles, and the structure of the column title is device name + parameter category name, and the structure of the row title is timestamp;

[0075] Then, the collected parameter data content is matched with the content of the column header and the row header, and the parameter data with the same content is added to the classification table;

[0076] The intersection of the vertical position of the column title and the horizontal position of the row title is the position of the corresponding parameter data compensation, and finally a classification table with parameter results is formed.

[0077] In the embodiment of the present invention, the image stitching algorithm implements the following image data processing operations:

[0078] According to the image acquisition sequence, the moving distance in each image acquisition process is determined to be L, and the horizontal distance of the image is H1, and the vertical distance is H2;

[0079] Starting from the first image, a segmentation line is set at the boundary of one side of the moving direction of the first image, and the distance between the segmentation line segment and the boundary of the moving direction of the image is H1-L or H2-L, and then the image segmentation operation is performed according to the segmentation line segment;

[0080] The image data before the segmentation is retained, and then the image data is spliced, such as connecting the pre-sequence part of the first image data and the pre-sequence part of the second image data, and sequentially splicing the two corner points of the first image in the moving direction with the two corner points of the second image relative to the first image, and finally forming complete image data.

[0081] In the embodiment of the present invention, the operation of extracting data and establishing a simulation analysis model in the deviation analysis module is as follows:

[0082] A preliminary model is established by extracting the equipment parameters in actual operation, and then historical data is introduced into the preliminary model to form a simulation analysis model that can be simulated;

[0083] The simulation analysis model extracts the corresponding image data and parameter data through the changes in welding position, and then performs analysis operations.

[0084] In the embodiment of the present invention, the operation of analyzing the image data after extraction using the simulation analysis model in the deviation analysis module is as follows:

[0085] After extracting the complete image data, the image data is grayscale processed, and comparison points are set at equal intervals on the image data. The characteristic parts in the image are determined by the grayscale value changes of the comparison points. Then, the known historical feature data is extracted and matched with the determined feature parts to obtain the interpretation of the feature parts.

[0086] After the welding position is determined, the corresponding image data is analyzed and the deviation and abnormality problems in the welding process are determined by combining the threshold.

[0087] In the embodiment of the present invention, the deviation analysis module determines the deviation abnormality problem in the welding process by combining the threshold value as follows:

[0088] Extract the image data of the front side of the weld and perform analysis operations, and use the corner points of the weld alignment position of the steel components in the image data as reference points;

[0089] The normal image data before welding is extracted, and then the image data during welding is extracted. The two images are overlaid based on the corresponding corner points of the images. The image data on the upper layer is blurred to maintain the required features of the upper and lower image data.

[0090] Then, a reference coordinate axis is established, the position of the reference point is calculated, and the historical safety threshold is introduced to compare with the calculated result.

[0091] In the embodiment of the present invention, the calculation operation of establishing the reference coordinate axis to realize the reference point change is as follows:

[0092] The lower left corner of the image data is used as the coordinate axis origin, and the X axis is established from the origin along the horizontal boundary of the image data, and the Y axis is established from the origin along the vertical boundary of the image data;

[0093] Then, based on the reference coordinate axes, the coordinates of the reference points are determined, and the coordinates of reference point A are set to (x1, y1), the coordinates of reference point B are set to (x2, y2), the coordinates of reference point C are set to (x3, y3), and the coordinates of reference point D are set to (x4, y4);

[0094] When x1=x2 and y1=y2, and x3=x4 and y3=y4, then there is no abnormality in the welding operation of the current steel component;

[0095] When x1≠x2 or y1≠y2, or x3≠x4 or y3≠y4, the welding operation of the current steel component is abnormal, and then the calculated deviation value is compared with the safety threshold;

[0096] The calculation formula for the deviation value is:

[0097]

[0098] And K1 belongs to the distance between reference point A and reference point B, while K2 belongs to the distance between reference point A and reference point B;

[0099] The safety threshold interval is set to [M, N], and when K1 or K2 both belong to [M, N], the current value is normal. Conversely, if either K1 or K2 does not belong to [M, N], the current value is abnormal.

[0100] Among them, by extracting the image data of the front side of the welding point for analysis, and taking the corner points of the welding alignment position of the steel structure in the image data as reference points, and extracting the normal image data before welding, and then extracting the image data during welding, the overlay operation of the two images is realized with the corresponding corner point positions of the images, and the image data on the upper layer is blurred to keep the required features of the upper and lower image data visible, and then establish a reference coordinate axis, calculate the position of the reference point, and introduce historical safety thresholds for comparison with the calculated results, so that deviation detection and real-time analysis operations can be completed accurately, and abnormal conditions can be obtained and compensated in real time in subsequent processing to avoid deviation problems during welding.

[0101] In the embodiment of the present invention, the deviation analysis module extracts the force data based on the current deviation anomaly problem and performs the operation of tracing the welding position parameters as follows:

[0102] If the value is abnormal, the pressure applied at the current welding position causes the steel structure to deviate, and the current position is marked;

[0103] Then, the current pressure on the steel structure is extracted as F, and a compensation strategy is generated. The force provided to the subsequent jacking equipment is the jacking force f in the opposite direction, and F=f.

[0104] The difference between the second embodiment and the first embodiment is that the present invention further discloses a deviation analysis and detection method for high-toughness steel components, which specifically includes the following steps:

[0105] S1. Real-time data collection and pre-processing of high-toughness steel component welding operations;

[0106] S2. Then extract historical data to establish a simulation analysis model, implement the search operation for abnormal deviation problems in the welding process, generate an adjustment strategy to achieve optimization compensation of the simulation analysis model, and introduce real-time data into the simulation analysis model for analysis;

[0107] S3. Transmit the final instruction to the corresponding device to implement the control and compensation operation.

[0108] At the same time, the contents not described in detail in this specification belong to the existing technology well known to those skilled in the art.

[0109] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0110] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A deviation analysis and detection system for high-toughness steel components, characterized by: include: A data acquisition module performs welding operations on high-toughness steel components, and sets image acquisition equipment and force sensors around the high-toughness steel components to realize data acquisition operations; The data processing module pre-processes the raw data collected by the data acquisition module to complete the data classification; The deviation analysis module extracts data to establish a simulation analysis model. The simulation analysis model is used to analyze the extracted image data and determine the deviation anomaly in the welding process based on the threshold. Based on the current deviation anomaly, the force data is extracted and the welding position parameters are traced. The adjustment parameters are calculated based on the welding position parameters. The simulation analysis model is optimized by adjusting the parameters, and then real-time data is introduced into the simulation analysis model for analysis and an adjustment strategy is generated to achieve compensation for the welding operation. The decision-making application module uses the generated adjustment strategy to transmit it to the corresponding equipment to implement the control operation; The visual control interface displays the processed data and generated strategies in the form of text and charts.

2. The deviation analysis and detection system for high-toughness steel components according to claim 1, characterized in that: The data processing module performs preprocessing operations on the raw data collected by the data collection module as follows: The filtering algorithm is used to remove noise points from the collected parameter data, and the collected image data is enhanced and denoised; Then set up a classification table to classify the collected parameter data; The collected image data is processed by image stitching algorithm to form complete image data for analysis.

3. The deviation analysis and detection system for high-toughness steel components according to claim 2, characterized in that: The operation of the classification table to classify the collected parameter data is: The classification table sets column and row titles, and the structure of the column title is device name + parameter category name, and the structure of the row title is timestamp; Then, the collected parameter data content is matched with the content of the column header and the row header, and the parameter data with the same content is added to the classification table; The intersection of the vertical position of the column title and the horizontal position of the row title is the position of the corresponding parameter data compensation, and finally a classification table with parameter results is formed.

4. The deviation analysis and detection system for high-toughness steel components according to claim 2, characterized in that: The image stitching algorithm realizes the following image data processing operations: According to the image acquisition sequence, the moving distance in each image acquisition process is determined to be L, and the horizontal distance of the image is H1, and the vertical distance is H2; Starting from the first image, a segmentation line is set at the boundary of one side of the moving direction of the first image, and the distance between the segmentation line segment and the boundary of the moving direction of the image is H1-L or H2-L, and then the image segmentation operation is performed according to the segmentation line segment; The image data before the segmentation is retained, and then the image data is spliced, such as connecting the pre-sequence part of the first image data and the pre-sequence part of the second image data, and sequentially splicing the two corner points of the first image in the moving direction with the two corner points of the second image relative to the first image, and finally forming complete image data.

5. The deviation analysis and detection system for high-toughness steel components according to claim 1, characterized in that: The operation of extracting data and establishing a simulation analysis model in the deviation analysis module is as follows: A preliminary model is established by extracting the equipment parameters in actual operation, and then historical data is introduced into the preliminary model to form a simulation analysis model that can be used for simulation; The simulation analysis model extracts the corresponding image data and parameter data through the changes in welding position, and then performs analysis operations.

6. The deviation analysis and detection system for high-toughness steel components according to claim 4, characterized in that: The operation of analyzing the image data extracted by the simulation analysis model in the deviation analysis module is as follows: After extracting the complete image data, the image data is grayscale processed, and comparison points are set at equal intervals on the image data. The characteristic parts in the image are determined by the grayscale value changes of the comparison points. Then, the known historical feature data is extracted and matched with the determined feature parts to obtain the interpretation of the feature parts. After the welding position is determined, the corresponding image data is analyzed and the deviation and abnormality problems in the welding process are determined by combining the threshold.

7. The deviation analysis and detection system for high-toughness steel components according to claim 1, characterized in that: The deviation analysis module determines the abnormal deviation problem in the welding process by combining the threshold value as follows: Extract the image data of the front side of the weld and perform analysis operations, and use the corner points of the weld alignment position of the steel components in the image data as reference points; The normal image data before welding is extracted, and then the image data during welding is extracted. The two images are overlaid based on the corresponding corner points of the images. The image data on the upper layer is blurred to maintain the required features of the upper and lower image data. Then, a reference coordinate axis is established, the position of the reference point is calculated, and the historical safety threshold is introduced to compare with the calculated result.

8. The deviation analysis and detection system for high-toughness steel components according to claim 7, characterized in that: The calculation operation of establishing the reference coordinate axis to realize the reference point change is: The lower left corner of the image data is used as the coordinate axis origin, and the X axis is established from the origin along the horizontal boundary of the image data, and the Y axis is established from the origin along the vertical boundary of the image data; Then, based on the reference coordinate axes, the coordinates of the reference points are determined, and the coordinates of reference point A are set to (x1, y1), the coordinates of reference point B are set to (x2, y2), the coordinates of reference point C are set to (x3, y3), and the coordinates of reference point D are set to (x4, y4); When x1=x2 and y1=y2, and x3=x4 and y3=y4, then there is no abnormality in the welding operation of the current steel component; When x1≠x2 or y1≠y2, or x3≠x4 or y3≠y4, the welding operation of the current steel component is abnormal, and then the calculated deviation value is compared with the safety threshold; The calculation formula for the deviation value is: or And K1 belongs to the distance between reference point A and reference point B, while K2 belongs to the distance between reference point A and reference point B; The safety threshold interval is set to [M, N], and when K1 or K2 both belong to [M, N], the current value is normal. Conversely, if either K1 or K2 does not belong to [M, N], the current value is abnormal.

9. The deviation analysis and detection system for high-toughness steel components according to claim 8, characterized in that: The deviation analysis module extracts force data based on the current deviation anomaly problem and traces the position parameters of the welding as follows: If the value is abnormal, the pressure applied at the current welding position causes the steel structure to deviate, and the current position is marked; Then, the current pressure on the steel structure is extracted as F, and a compensation strategy is generated. The force provided to the subsequent jacking equipment is the jacking force f in the opposite direction, and F=f.

10. A method for analyzing and detecting deviations of high-toughness steel components, using a system for analyzing and detecting deviations of high-toughness steel components according to any one of claims 1 to 9, characterized in that: The specific steps include: S1. Real-time data collection and pre-processing of high-toughness steel component welding operations; S2. Then extract historical data to establish a simulation analysis model, implement the search operation for abnormal deviation problems in the welding process, generate an adjustment strategy to achieve optimization compensation of the simulation analysis model, and introduce real-time data into the simulation analysis model for analysis; S3. Transmit the final instruction to the corresponding device to implement the control and compensation operation.

Images