Special-shaped box girder production quality traceability management system

Abstract

The invention belongs to the technical field of special-shaped box girder production quality traceability, and discloses a special-shaped box girder production quality traceability management system, which comprises a three-dimensional geometric feature acquisition module, a morphological defect identification module, a morphological defect association traceability module and a solder distribution association traceability module. Scanning the internal three-dimensional morphological characteristics of the variable cross section of the special-shaped box girder, analyzing and positioning the defect position so as to associate the plate defect with the welding defect; and the over-welding phenomenon is judged by quantitatively analyzing the interlayer lap joint characteristics between the welding layers in the welding seam through the transverse offset and the penetration depth at the positions with the defects in the form, and whether the over-welding phenomenon exists or not is judged. According to the method, the association of the morphological defects and the welding process defects is fused, so that the whole-process accurate traceability from the macroscopic morphological defects to the microcosmic solder distribution is realized, and the problems that the defect attribution is fuzzy and the process deviation is difficult to trace in the traditional quality inspection are solved.

Description

Technical Field

[0001] The invention belongs to the technical field of special-shaped box girder production quality traceability, and relates to a special-shaped box girder production quality traceability management system. Background Art

[0002] In modern complex projects, special-shaped box girders are widely used in the construction of important infrastructure such as bridges and rail transit due to their strong structural adaptability and high space utilization.

[0003] Special-shaped box girders are usually made up of plates of different shapes and have a variable cross-section structural feature. This type of structure is composed of a top plate, a bottom plate, and ribs. The changes in its geometric shape may lead to uneven force in various parts, which is prone to stress concentration, thus affecting the safety and durability of the overall structure. At the same time, because the variable cross-section structure is made up of multiple steel plates connected by a high-precision welding process, its welding quality directly affects the overall strength and fatigue life of the box girder. Its morphological accuracy and weld quality are directly related to the mechanical properties and service life of the structure. Once structural abnormalities or welding quality problems occur during use, it will be difficult to quickly locate the source of the problem, affecting the analysis of the cause of the accident and the determination of responsibility. Therefore, it is of great significance to carry out production quality traceability management for special-shaped box girders.

[0004] Although the existing technology still has the following problems in variable-section quality traceability: the existing scheme regards plate size defects and welding defects as independent problems, and does not establish a traceability logic for the causal relationship between the two. When plate deformation is detected, it is directly attributed to "plate processing error", ignoring the secondary deformation that may be caused by welding thermal stress. When weld defects are found, they are simply attributed to "over-welding process", and it is not traced whether it is caused by the plate size error forcing the welder to compensate for the parameters. This leads to one-sided defect attribution and inaccurate production quality traceability, making it difficult to accurately implement quality control measures, which easily leads to repetitive quality problems and waste of resources. Summary of the Invention

[0005] In view of this, in order to solve the problems raised in the above background technology, a special-shaped box girder production quality traceability management system is proposed.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A special-shaped box girder production quality traceability management system, including the following modules: a three-dimensional geometric feature acquisition module, used to obtain the three-dimensional morphological characteristic parameters of the top plate, bottom plate and rib plate of the variable-section internal of the special-shaped box girder through a three-dimensional laser scanning device.

[0007] The morphological defect recognition module is used to identify and locate the defects in the internal structure of the special-shaped box girder with variable cross-section through three-dimensional morphological feature parameters.

[0008] The morphological defect association tracing module is used to scan the weld at the defect location through X-ray imaging equipment to generate a three-dimensional image, extract the overlap characteristics between weld layers, and determine whether there is over-welding. If over-welding exists, the defect cause at the defect location is traced back to over-welding in the welding process; otherwise, the defect cause is traced back to plate processing errors.

[0009] The solder distribution association traceability module is used to scan the weld at a defect-free position to generate a three-dimensional image, obtain the solder color distribution characteristics of the weld and the distribution characteristics of the voids inside the solder, analyze the uniformity of the solder distribution, focus on the uneven position when identifying solder unevenness, associate the production database to retrieve the welding process data of the corresponding position, and trace the cause of the unevenness by comparing the actual welding parameters at the uneven position with the deviation from the process standard.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention performs a three-dimensional morphological feature scan on the variable cross-section of the special-shaped box girder to analyze whether there are defective positions, conducts over-welding on the defective positions to determine the associated plate defects and welding defects, and traces the welding process quality of the non-defective positions. This overcomes the one-sided attribution problem caused by analyzing the plate defects and welding defects separately, and realizes the full-chain production quality traceability from the location of geometric defects to the causes of welding process, thereby accurately optimizing the production process and eliminating quality risks.

[0011] (2) The present invention determines the over-welding phenomenon by quantitatively analyzing the interlayer overlap characteristics between the welding layers inside the weld by the lateral offset and the penetration depth at the morphological defect position, which effectively solves the problem of one-sided identification of welding quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 This is a schematic diagram of the connection of various modules of the system of the present invention.

[0014] Figure 2 This is a flowchart for implementing the morphological defect association tracing module in the present invention.

[0015] Figure 3 This is a flow chart for implementing the solder distribution association traceability module in the present invention. DETAILED DESCRIPTION

[0016] 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.

[0017] See also Figure 1 As shown, the present invention provides a traceability management system for the production quality of special-shaped box girders, which includes a three-dimensional geometric feature acquisition module, a morphological defect recognition module, a morphological defect correlation traceability module, and a solder distribution correlation traceability module. All modules are connected in the above order.

[0018] The three-dimensional geometric feature acquisition module obtains the three-dimensional morphological feature parameters of the top plate, bottom plate and rib plate inside the variable-section special-shaped box beam through a three-dimensional laser scanning device.

[0019] It should be noted that 3D laser scanning is a non-contact measurement technology, which means that any physical damage to the surface of the special-shaped box girder can be avoided to obtain the 3D geometric features of the variable cross-section inside the special-shaped box girder.

[0020] The three-dimensional laser scanning device is started to perform non-contact scanning on the interior of the variable-section box girder to generate continuous high-density point cloud data of the top plate, bottom plate and rib plate.

[0021] The point cloud data of the roof surface is extracted, the absolute value of the height difference between adjacent scanning points is calculated, and the average value of the height difference of several consecutive scanning points is taken as the roof corrugation amplitude.

[0022] It should be noted that adjacent scanning points are determined by the scanning resolution. The point cloud is traversed with a sliding window, and the average of the absolute values ​​of the height differences between adjacent points in the window is calculated as the roof corrugation amplitude, which directly reflects the local tiny corrugations or unevenness on the roof surface and quantitatively determines the deformation defects of the roof.

[0023] The ideal plane of the base plate is generated by the plane fitting tool, and the vertical distance from each point cloud to the fitting plane is calculated as the local depression depth of the base plate.

[0024] It should be noted that the plane fitting tool uses the least squares plane fitting function built into the industrial inspection software to generate the ideal plane of the base plate.

[0025] The normal vectors of the surface point clouds on both sides of the rib are extracted, and the complementary angle of the angle between the normal vectors on both sides is calculated as the inclination angle of the rib.

[0026] Specifically, the point cloud is divided into a left side surface point cloud set and a right side surface point cloud set of the rib, and normal vector calculation is performed on the left side surface point cloud set to generate a left side surface point cloud normal vector set. Similarly, normal vector calculation is performed on the right side surface point cloud set to generate a right side surface point cloud normal vector set. Based on the left and right side surface point cloud normal vector sets, vector average extraction is implemented to calculate the left and right side characteristic normal vectors representing the overall spatial orientation of the surface, and the spatial angle between the left characteristic normal vector and the right characteristic normal vector is calculated. Using the principles of spatial geometry, the complementary angle of the normal vector angle is defined as the actual inclination angle of the rib.

[0027] The top plate corrugation amplitude, bottom plate local depression depth and rib inclination angle are taken as the three-dimensional morphological characteristic parameters of the top plate, bottom plate and rib plate inside the variable-section special-shaped box girder.

[0028] It should be noted that the morphological characteristics of the top plate, bottom plate and ribs inside the variable-section special-shaped box girder are quantified by the top plate corrugation amplitude, the local depression depth of the bottom plate and the inclination angle of the ribs, the undulation degree of the top plate surface and the local deformation of the bottom plate bearing surface are quantified, and the verticality deviation of the ribs is detected, so that the subsequent positioning of morphological defects is more accurate.

[0029] The morphological defect recognition module is used to identify and locate the defects in the internal structure of the special-shaped box girder with variable cross-section through three-dimensional morphological feature parameters.

[0030] The roof is divided into several equal areas along the length direction, and the number of points where the corrugation amplitude exceeds the standard in each area is counted. When the proportion of the exceeding standard area exceeds the preset roof corrugation threshold, it is determined to be a roof deformation defect.

[0031] It should be noted that the top plate corrugation threshold is preset based on the elastic modulus and bending stiffness difference of the steel box girder and industry specifications.

[0032] A specific example of the above process is that the variable-section roof is 10 meters long and can be divided into 20 0.5-meter areas. If the corrugation amplitude of 50 out of 200 points in the fifth area exceeds the standard, the calculated excess ratio is 25%. The roof corrugation threshold is set to 20%. At this time, the excess ratio is greater than the roof corrugation threshold, and the area is determined to be deformed.

[0033] Measuring points are arranged at fixed intervals on the bottom plate along the scanning path of the special-shaped box girder by the three-dimensional laser scanning device to detect whether the local depression depth of the bottom plate at several consecutive adjacent measuring points exceeds the preset bottom plate flatness tolerance range. If it exceeds, it is determined to be a local warping defect.

[0034] Specifically, the fixed spacing of measurement points is determined by the specific dimensions of the base plate, the desired accuracy requirements, and the inspection efficiency. For example, a measurement point can be set every 10 cm to ensure that any localized dents or warping on the base plate are captured.

[0035] More specifically, the base plate flatness tolerance range is determined based on design specifications and quality standards, which reflects the maximum deviation value allowed for the base plate during the manufacturing process to ensure that the final product meets structural performance and safety requirements.

[0036] Local warping defects may cause stress concentration and affect the stability and durability of the structure. Identifying local warping defects can accurately determine the type of morphological defects.

[0037] The deviation from the designed vertical angle is calculated based on the rib inclination angle. When the deviation value continuously exceeds the preset verticality tolerance range, it is determined to be a rib deformation defect.

[0038] Specifically, data representing the surfaces on both sides of the rib are extracted from the point cloud data obtained by the three-dimensional laser scanning device, and the normal vectors of the point cloud data on both sides of the rib are calculated respectively. The angle between the normal vectors on both sides is calculated, and the complementary angle is taken as the actual inclination angle of the rib. The actually measured inclination angle of the rib is compared with the vertical angle specified in the design, and the deviation value between the inclination angle of the rib and the designed vertical angle is calculated.

[0039] More specifically, the verticality tolerance range is determined based on a comprehensive consideration of factors such as engineering design specifications, material properties, and usage requirements.

[0040] It should be noted that the position of morphological defects is determined by top plate deformation defects, local warping defects and rib plate deformation defects, the positioning defects of internal plate problems of variable cross-sections are quantified, and the morphological defect problems are accurately determined.

[0041] See Figure 2 As shown, the morphological defect association tracing module is used to scan the weld at the defect location through X-ray imaging equipment to generate a three-dimensional image, extract the overlap characteristics between weld layers, and determine whether there is over-welding. If over-welding exists, the defect cause at the defect location is traced back to over-welding in the welding process; otherwise, the defect cause is traced back to plate processing error.

[0042] Specifically, the X-ray imaging device is started, aligned with the defect location and scanned from multiple angles to fully cover the weld area and collect information to construct a three-dimensional image.

[0043] It should be noted that X-ray imaging equipment is a non-destructive detection method that can perform detailed inspections of the internal structure without damaging special-shaped box girders. It provides high-definition images of the internal structure of the weld and can accurately observe key features such as the overlapping status and penetration depth between weld layers, thereby effectively identifying potential quality problems.

[0044] The generated three-dimensional image is segmented into several single-layer welding pool contours according to the layer thickness threshold.

[0045] It should be noted that the layer thickness threshold refers to the theoretical thickness of each layer of weld during the welding process, which is usually determined based on the requirements for weld size in the design documents of the special-shaped box girder.

[0046] Using the information of the single-layer welding pool profile, the lateral offset between adjacent layers can be accurately calculated, and the position of the unfused interface between the bottom of the pool and the base material can be located in the longitudinal cross-section of the single-layer weld pool, thereby measuring the penetration depth.

[0047] The weld pool contours of two adjacent welding layers are projected onto the same plane, with the base material reference plane as the spatial reference, ensuring that only the local relative position differences between the layers are retained.

[0048] Specifically, the contour geometric shape data of the two adjacent layers of welding pool are extracted from the generated three-dimensional image respectively, and the geometric shape data of the two layers of welding pool contours are adjusted so that they share the same coordinate system. For each layer of welding pool contour, its position in three-dimensional space is found, and the welding pool contour of each layer is projected onto the reference plane in a direction perpendicular to the base material reference plane, eliminating the influence of different layers due to the difference in welding height, and only retaining the information of the relative position between layers.

[0049] Get the leftmost and rightmost positions of the weld pool contours of two adjacent layers in the horizontal direction.

[0050] Specifically, all vertex coordinates of the projected outline polygon are traversed, and the vertex with the smallest coordinate value is selected as the leftmost position, and the vertex with the largest coordinate value is selected as the rightmost position.

[0051] The rightmost boundary of the two left ends is taken as the starting position of the overlap, and the leftmost boundary of the two right ends is taken as the ending position of the overlap.

[0052] Subtract the overlap start position from the overlap end position. If the result is positive, it is defined as the effective overlap area width between layers. If it is negative or zero, it means there is no effective overlap.

[0053] It's important to note that quantifying the effective overlap width between weld layers not only identifies overlap issues but also allows for further analysis of the lateral offset and penetration depth of these issues. A positive result indicates a clear effective overlap area and a good physical connection between the two layers within this area. A negative or zero result indicates a lack of effective overlap, potentially indicating defects caused by improper welding procedures, such as misalignment or insufficient penetration, which can compromise weld quality and reliability.

[0054] Select measurement points at equal intervals along the welding direction within the effective overlap area, calculate the horizontal difference between the center lines of adjacent layers, and take the maximum horizontal difference as the lateral offset.

[0055] Specifically, for each selected measuring point, the centerline coordinates of the corresponding position on the two-layer welding pool contour are found respectively, the centerline coordinates of the two adjacent layers at the same position are compared, and the horizontal distance difference between them is calculated. All measuring points are traversed, and the maximum value of all calculated horizontal differences is found as the lateral offset.

[0056] It should be noted that the center line usually refers to the line closest to the geometric center on the cross section of the weld, which represents the approximate central axis of the weld pool of that layer.

[0057] The unfused interface between the bottom of the molten pool and the parent material is located in the longitudinal section of the single-layer molten pool. The penetration depth is the vertical distance from the lowest point of the molten pool to the interface of the parent material.

[0058] It should be noted that the unfused interface refers to the place where the molten pool metal fails to completely merge into the base material, which represents the ideal penetration limit.

[0059] Specifically, a longitudinal cross-sectional image of a single-layer welding molten pool is extracted from the three-dimensional image data, the boundary of the molten pool is identified and marked, the lowest point of the molten pool on the longitudinal cross-section, that is, the position of the bottom of the molten pool, is found, and a vertical line segment is drawn from the bottom of the molten pool to the unfused interface of the base material. The length of this line segment is the penetration depth.

[0060] The lateral offset and penetration depth are used as the overlap characteristics of the weld layers.

[0061] By quantifying the inter-layer overlap characteristics of the weld seams as lateral offset and penetration depth, the quality of each layer of welding can be effectively evaluated to ensure that each weld meets the expected design requirements, thereby guaranteeing the overall strength and safety of the special-shaped box girder.

[0062] Preset the allowable lateral deviation threshold and the upper limit of penetration depth safety.

[0063] Specifically, preset thresholds for allowable lateral deflection and upper safety limits for penetration depth are typically based on industry standards, welding procedure qualification and design document requirements, and relevant quality control specifications. By comparing the allowable lateral deflection thresholds and upper safety limits for penetration depth with actual welding conditions, it is possible to effectively determine whether over-welding has occurred.

[0064] If the lateral offset does not exceed the allowable lateral offset threshold and the penetration depth exceeds the safety upper limit of the penetration depth for two consecutive layers, it is determined that over-welding occurs.

[0065] It should be noted that over-welding is caused by excessive penetration due to excessive welding energy. If the lateral offset does not exceed the allowable lateral offset threshold, it means that the weld position deviation during welding is within the acceptable range, with no obvious misalignment or offset issues. If the penetration depth exceeds the safe upper limit for two consecutive layers, it indicates that the energy input during welding is too high.

[0066] If the lateral offset exceeds the allowable lateral offset threshold and the penetration depth does not exceed the safety upper limit of the penetration depth, it is determined that there is no over-welding phenomenon.

[0067] It should be noted that over-welding is usually caused by excessive welding energy, which may cause deformation of the internal plate of the variable cross-section. In areas where plate morphological defects and welding defects coexist, over-welding can provide a basis for further analysis of the defect type, thereby more effectively identifying the cause of the defect. However, its classification still needs to be comprehensively judged in combination with other verification methods.

[0068] See Figure 3 As shown, the solder distribution association traceability module is used to scan the weld at a defect-free position to generate a three-dimensional image, obtain the solder color distribution characteristics of the weld and the distribution characteristics of the voids inside the solder, analyze the uniformity of the solder distribution, focus on the uneven position when identifying solder unevenness, associate the production database to retrieve the welding process data of the corresponding position, and trace the cause of the unevenness by comparing the actual welding parameters at the uneven position with the deviation from the process standard.

[0069] The weld seam in the defect-free area is scanned at multiple angles using X-ray imaging equipment to obtain two-dimensional transmission images at different projection angles.

[0070] It should be noted that because welds and their internal structure can be complex and non-uniform, scanning from a single angle may not provide enough information to fully assess weld quality. Collecting data from multiple angles ensures a comprehensive view of the weld's internal structure, allowing for more accurate inspection and analysis of weld quality and filler metal distribution.

[0071] The three-dimensional volume data model of the weld area is reconstructed based on the filtered back-projection algorithm to generate a three-dimensional space matrix containing grayscale distribution.

[0072] Specifically, the collected 2D transmission images from different projection angles undergo preprocessing, including image correction, noise removal, and registration, to ensure that all images are precisely aligned. The preprocessed images are then filtered, and the filtered data is used for backprojection. By accumulating the backprojection results from all angles, a 3D volumetric data model of the weld area is gradually constructed. As the backprojection process proceeds, the system gradually forms a 3D spatial matrix containing multiple volume pixels. Each volume pixel represents a small cube in 3D space, and its value represents the relative density or absorption coefficient of the material at that location, typically presented as a grayscale value.

[0073] It should be noted that back-projection refers to re-projecting each filtered two-dimensional projection image back into the three-dimensional space according to its corresponding viewing angle.

[0074] The detection sections are divided according to preset intervals along the length of the weld. In each detection section, collection points are arranged at equal angles along the circumference of the cross section to obtain the local grayscale value of each collection point.

[0075] It should be noted that the selection of the preset spacing mainly depends on the specific characteristics of the weld, the required detection accuracy and the feasibility in actual operation. Generally, a detection segment can be set every 1 mm.

[0076] Specifically, the geometric center of the weld cross section is found as a reference point, and collection points are evenly arranged along the circumference of the weld cross section starting from the reference point at predetermined angular intervals (eg, 10 degrees).

[0077] More specifically, the steps for obtaining the geometric center of the weld cross section are as follows: converting the grayscale image into a binary image, applying an edge algorithm to extract the weld contour, and obtaining the geometric center by calculating the centroid.

[0078] By evenly distributing sampling points around the circumference of the weld cross section, data is captured from multiple angles and positions, enabling a comprehensive assessment of weld quality. Single-angle or uneven sampling can miss important details, resulting in an incomplete understanding of the weld's internal structure. Providing evenly distributed sampling points helps ensure uniform and consistent data collection, avoiding oversampling in some areas and undersampling in others.

[0079] The absolute value of the grayscale difference between adjacent acquisition points is calculated and defined as the solder color distribution gradient.

[0080] It should be noted that a high gradient value indicates a large grayscale change in the area, which may mean that the solder is unevenly distributed or there are other abnormal conditions. A low gradient value indicates a small grayscale change in the area, which may mean that the solder is relatively evenly distributed.

[0081] The solder color distribution gradient can be used to simply and intuitively evaluate the color distribution of the solder inside the weld and identify possible unevenness or potential defects.

[0082] The defect-free area of ​​the weld is scanned by a multi-angle X-ray imaging device, and a digital grayscale image is collected at each rotation angle.

[0083] In the digital grayscale image, the grayscale value of each voxel is compared with the preset threshold. If the voxel grayscale value is lower than the threshold, the area is a void candidate area, and its spatial coordinates are recorded to form a void position set.

[0084] It should be noted that the preset threshold should be selected based on the quality control specifications for internal weld defect detection. The grayscale values ​​of weld material and void areas under X-ray imaging typically differ significantly. Normal weld material, due to its higher density, absorbs more X-rays, resulting in a higher grayscale value. Void areas, due to their lower density, absorb less X-rays, resulting in a lower grayscale value.

[0085] Calculate the dispersion of void distribution inside solder based on void location set , where Indicates the hole number, , represents the total number of voids, Indicates the The distance from the center of the cavity to the center of the weld, represents the average distance from all holes to the center, Indicates the maximum distance from the weld center to the section boundary.

[0086] It should be noted that the average distance from all cavities to the weld center is calculated , then calculate the sum of squared deviations between the distance from each void to the weld center and the average distance , reflects the dispersion of each cavity position relative to the average position, It directly reflects the geometric size and shape characteristics of the weld. In order to eliminate the influence of weld size, the square sum of the above deviations is divided by , and obtain the final degree of discretization ,in, , Indicates the A two-dimensional horizontal coordinate of a hole, Indicates the The two-dimensional vertical coordinate of a hole.

[0087] When the solder color distribution gradient and the degree of dispersion of the solder internal void distribution are both within their uniformity thresholds, the solder distribution uniformity is judged to meet the standard.

[0088] When any feature exceeds the uniformity threshold, the solder distribution is determined to be non-uniform.

[0089] It's important to note that the solder color distribution gradient reflects the consistency and uniformity of the material within the weld. A lower color distribution gradient indicates less variation in density and composition within the weld, indicating a more uniform solder distribution. If the color distribution gradient at all measurement points is less than the set uniformity threshold, the solder distribution is considered to be well-uniform.

[0090] The dispersion of voids within the solder reflects the spatial distribution of voids within the weld. The presence of voids within the solder itself indicates uneven distribution. When the dispersion of void distribution exceeds the set uniformity threshold, it further indicates that the void distribution is more dispersed and the solder filling quality distribution is less uniform.

[0091] By comprehensively analyzing the color distribution gradient of the solder and the degree of discreteness of the internal void distribution, a more comprehensive and accurate solder distribution uniformity assessment system can be constructed. A single indicator is often unable to fully reflect the quality issues within the weld. The solder color distribution gradient is mainly used to characterize the spatial continuity and distribution trend of the solder, while the presence of voids means that there is no solder filling in the area at all, which usually leads to the loss of color distribution gradient information, making it impossible to effectively identify potential defects. By introducing the analysis of the discrete degree of void distribution within the solder, quality abnormalities in the weld can be more accurately identified, compensating for the lack of color gradient information. The combination of the two not only improves the accuracy of defect detection, but also helps to more effectively locate and analyze potential welding process problems.

[0092] In the uneven area marked by the solder color distribution gradient, the distance between the collection points is further reduced to form an encrypted detection grid.

[0093] It should be noted that the aforementioned further reduction of the collection point spacing is based on the initially set collection point spacing. For example, if the initially set collection point spacing is 1 mm, it can be reduced to 0.5 mm in the uneven area.

[0094] The acquisition points are rearranged to ensure that the newly formed grid can cover the entire uneven area, thereby providing a higher resolution dataset.

[0095] Statistical grayscale difference rate within the encrypted grid , where Indicates the highest gray value point in the encrypted grid. Indicates the lowest gray value point in the encrypted grid. Indicates the average gray value of the detection segment.

[0096] It should be noted that the difference between the maximum and minimum grayscale values ​​within the encrypted grid reflects the range of grayscale value variation within the encrypted grid. In order to eliminate the influence of absolute values, the grayscale difference is divided by the average grayscale value of the detection segment to obtain the relative difference, which reflects the degree of grayscale variation relative to the average grayscale value. The smaller the value, the more uniform the grayscale value in the encrypted grid area and the more consistent the solder distribution. A large value indicates that the grayscale value in the encrypted grid area varies greatly, and there may be a problem of uneven solder distribution.

[0097] As a specific example of the above formula, there is a dense grid area containing the following grayscale values ​​(in pixels): 50, 60, 70, 80, 90. The grayscale difference rate within the dense grid area is calculated to be approximately 57.14%.

[0098] When the grayscale difference rate exceeds a preset threshold, the encrypted grid is located at a position where the solder color distribution is uneven, where the center coordinates of the weld are located.

[0099] Specifically, locating the center coordinates of the refined grid involves the following steps: For each marked area, the minimum and maximum coordinates represent the minimum and maximum values ​​of the spatial range across all dimensions, respectively. Next, the average of the minimum and maximum values ​​in each dimension is calculated to obtain the x, y, and z coordinates of the center point.

[0100] The locations of voids and uneven solder color distribution are defined as uneven locations.

[0101] It should be noted that defining the void locations and the locations of uneven color distribution together as uneven locations can help accurately locate areas that require special attention and comprehensively evaluate the distribution of solder.

[0102] According to the coordinate information of the uneven position, the welding process record corresponding to the position is searched and extracted in the production database.

[0103] It should be noted that by comparing the actual welding parameters with the standard process parameters, the root cause of uneven solder distribution can be determined.

[0104] The process parameters recorded during the actual welding process are compared item by item with the standard process parameters of the corresponding process.

[0105] It should be noted that the process parameters may be welding temperature, welding voltage, welding current, etc.

[0106] The entire welding process is divided into multiple time periods according to the start and end time of welding.

[0107] Specifically, the length of each time period is determined based on the characteristics of the welding process. For example, each 1-minute time interval can be used as a time period.

[0108] It's important to note that transient disturbances can occur during the welding process, causing certain parameters to briefly exceed standard ranges. These transient deviations often have no real impact on the process, and judging based solely on a single point in time can easily lead to misjudgments. Therefore, analyzing welding parameters over a specific time period can more accurately pinpoint deviations from standard parameters, thereby enabling targeted improvements to the welding process and improving product quality.

[0109] For each welding parameter recorded in each time period, the duration of its deviation from the standard range is calculated and compared with the total duration of the time period to obtain the parameter deviation ratio.

[0110] Specifically, the specific contents of calculating the duration of deviation from the standard range are as follows: obtain the welding parameter records in each time period, which usually include a series of timestamps and their corresponding parameter values, traverse all data points in each time period, identify records that exceed the pre-set standard range, and mark these data points that exceed the standard.

[0111] If there are multiple discontinuous intervals that exceed the standard range, the duration of all these intervals is accumulated to obtain the total duration of the welding parameters deviating from the standard range during the entire time period.

[0112] The parameter deviation ratio can clearly show the deviation of welding parameters from the standard in each time period, which is helpful for subsequent quality analysis and process improvement.

[0113] A parameter deviation ratio threshold is set, and all parameter items and time periods are traversed. If the deviation ratio of a certain parameter in a certain time period exceeds the parameter deviation ratio threshold, the parameter is regarded as the uneven cause.

[0114] It should be noted that the setting of the parameter deviation ratio refers to the quality control standards of the relevant industry or past experience data to determine a reasonable threshold. For example, according to historical data analysis, when the welding current deviates from the standard by more than 20%, the weld quality decreases significantly. In this case, the current deviation ratio threshold can be set to 20%.

[0115] By comparing actual welding parameters with process standards, we can ensure that the welding process meets the established quality requirements and provide data and experience for the subsequent production of special-shaped box girders.

[0116] The parameters involved in the above formulas are all dimensionless and calculated using their numerical values. The preset parameters in the formulas are set by those skilled in the art according to actual conditions.

[0117] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0118] Those skilled in the art will appreciate that the modules and algorithm steps of each example 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 performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0119] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0120] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0121] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A special-shaped box girder production quality traceability management system, characterized by: Includes the following modules: The 3D geometric feature acquisition module uses a 3D laser scanning device to obtain the 3D morphological characteristic parameters of the top plate, bottom plate and rib plate inside the variable-section special-shaped box girder; The morphological defect recognition module uses three-dimensional morphological characteristic parameters to identify and locate defects in the internal structure of the special-shaped box girder with variable cross-section; The morphological defect correlation traceability module uses X-ray imaging equipment to scan the weld at the defect location to generate a three-dimensional image, extract the overlap characteristics of the weld layers, and determine whether there is over-welding. If over-welding is present, the defect at the defect location is traced back to over-welding during the welding process. Otherwise, the defect is traced back to plate processing errors. The solder distribution association traceability module scans the weld at a defect-free position to generate a three-dimensional image, obtains the solder color distribution characteristics of the weld and the distribution characteristics of the voids inside the solder, analyzes the uniformity of the solder distribution, focuses on the uneven position when identifying solder unevenness, associates with the production database to retrieve the welding process data of the corresponding position, and traces the cause of the unevenness by comparing the actual welding parameters at the uneven position with the deviation from the process standard.

2. The special-shaped box girder production quality traceability management system according to claim 1, characterized in that: The specific contents of the three-dimensional geometric feature acquisition module are as follows: Start the 3D laser scanning device to perform non-contact scanning on the interior of the variable cross-section box girder to generate continuous high-density point cloud data of the top plate, bottom plate and rib plate; Extract the point cloud data of the roof surface, calculate the absolute value of the height difference between adjacent scanning points, and take the average value of the height difference of several consecutive scanning points as the roof corrugation amplitude; Generate an ideal bottom plate plane using a plane fitting tool, and calculate the vertical distance from each point cloud to the fitting plane as the local depression depth of the bottom plate; Extract the normal vectors of the surface point clouds on both sides of the rib, and calculate the supplementary angle of the angle between the normal vectors on both sides as the rib inclination angle; The top plate corrugation amplitude, bottom plate local depression depth and rib inclination angle are taken as the three-dimensional morphological characteristic parameters of the top plate, bottom plate and rib plate inside the variable-section special-shaped box girder.

3. The special-shaped box girder production quality traceability management system according to claim 2, characterized in that: The specific contents of the morphological defect recognition module are as follows: The roof is divided into several equal areas along the length direction, and the number of points where the corrugation amplitude exceeds the standard in each area is counted. When the proportion of the areas exceeding the standard exceeds the preset roof corrugation threshold, it is determined to be a roof deformation defect; Measuring points are arranged at fixed intervals on the bottom plate along the scanning path of the special-shaped box girder by the 3D laser scanning device to detect whether the local depression depth of the bottom plate at several consecutive adjacent measuring points exceeds the preset bottom plate flatness tolerance range. If it exceeds, it is determined to be a local warping defect; The deviation from the designed vertical angle is calculated based on the rib inclination angle. When the deviation value continuously exceeds the preset verticality tolerance range, it is determined to be a rib deformation defect.

4. The special-shaped box girder production quality traceability management system according to claim 1, characterized in that: The specific steps of scanning the weld at the defect location by X-ray imaging equipment to generate a three-dimensional image and extracting the overlap characteristics of the weld layers are as follows: The generated three-dimensional image is segmented into several single-layer welding pool contours according to the layer thickness threshold; Project the weld pool contours of two adjacent weld layers onto the same plane, using the parent material reference plane as the spatial reference, ensuring that only the local relative position differences between the layers are retained; Obtain the leftmost and rightmost positions of the weld pool contours of two adjacent layers in the horizontal direction; The rightmost boundary of the two left ends is taken as the starting position of the overlap, and the leftmost boundary of the two right ends is taken as the ending position of the overlap; Subtract the overlap start position from the overlap end position. If the result is positive, it is defined as the effective overlap area width between layers; if it is negative or zero, it means there is no effective overlap; Select measurement points at equal intervals along the welding direction within the effective overlap area, calculate the horizontal difference between the center lines of adjacent layers, and take the maximum horizontal difference as the lateral offset; In the longitudinal section of a single-layer molten pool, the unfused interface between the bottom of the molten pool and the parent material is located, and the penetration depth is the vertical distance from the lowest point of the molten pool to the interface of the parent material; The lateral offset and penetration depth are used as the overlap characteristics of the weld layers.

5. The special-shaped box girder production quality traceability management system according to claim 4, characterized in that: The specific contents of judging whether there is over-welding are as follows: Preset the allowable lateral offset threshold and the upper limit of penetration depth safety; If the lateral offset does not exceed the allowable lateral offset threshold and the penetration depth exceeds the safety upper limit of the penetration depth for two consecutive layers, it is determined that over-welding occurs; If the lateral offset exceeds the allowable lateral offset threshold and the penetration depth does not exceed the safety upper limit of the penetration depth, it is determined that there is no over-welding phenomenon.

6. The special-shaped box girder production quality traceability management system according to claim 1, characterized in that: The specific steps of obtaining the solder color distribution characteristics of the weld are as follows: Use X-ray imaging equipment to perform multi-angle scanning of the weld in the defect-free area to obtain two-dimensional transmission images at different projection angles; Reconstruct the 3D volume data model of the weld area based on the filtered back-projection algorithm to generate a 3D spatial matrix containing grayscale distribution; The weld length is divided into test sections at preset intervals. Within each test section, collection points are arranged at equal angles along the circumference of the cross section to obtain the local grayscale value of each collection point. The absolute value of the grayscale difference between adjacent acquisition points is calculated and defined as the solder color distribution gradient.

7. The special-shaped box girder production quality traceability management system according to claim 6, characterized in that: The specific content of the distribution characteristics of the voids inside the solder is as follows: Perform multi-angle X-ray imaging scanning on the defect-free area of ​​the weld, and collect digital grayscale images at each rotation angle; In the digital grayscale image, the grayscale value of each voxel is compared with the preset threshold. If the voxel grayscale value is lower than the threshold, the area is a void candidate area, and its spatial coordinates are recorded to form a void position set; Calculate the dispersion of void distribution inside solder based on void location set , where Indicates the hole number, , represents the total number of voids, Indicates the The distance from the center of the cavity to the center of the weld, represents the average distance from all holes to the center, Indicates the maximum distance from the weld center to the section boundary.

8. The special-shaped box girder production quality traceability management system according to claim 7, characterized in that: The specific contents of analyzing the uniformity of solder distribution are as follows: When the solder color distribution gradient and the degree of dispersion of the solder internal void distribution are both within their uniformity thresholds, the solder distribution uniformity is determined to meet the standard; When any feature exceeds the uniformity threshold, the solder distribution is determined to be non-uniform.

9. The special-shaped box girder production quality traceability management system according to claim 8, characterized in that: The specific contents of focusing on the uneven position when identifying the uneven solder are as follows: In the uneven area marked by the solder color distribution gradient, the distance between the collection points is further reduced to form a dense detection grid; Statistical grayscale difference rate within the encrypted grid , where Indicates the highest gray value point in the encrypted grid. Indicates the lowest gray value point in the encrypted grid. Indicates the average gray value of the detection segment; When the grayscale difference rate exceeds the preset threshold, the center coordinate of the encrypted grid is located at the position where the solder color distribution is uneven at the weld; The locations of voids and uneven solder color distribution are defined as uneven locations.

10. The special-shaped box girder production quality traceability management system according to claim 1, characterized in that: The associated production database retrieves the welding process data of the corresponding position, and traces the cause of the unevenness by comparing the actual welding parameters at the uneven position with the deviation of the process standard. The specific steps are as follows: Searching and extracting the welding process record corresponding to the uneven position in the production database according to the coordinate information of the uneven position; Compare the process parameters recorded during the actual welding process with the standard process parameters of the corresponding process item by item; The entire welding process is divided into multiple time periods according to the welding start and end time; For each welding parameter recorded in each time period, calculate the duration of its deviation from the standard range and compare it with the total duration of the time period to obtain the parameter deviation ratio; A parameter deviation ratio threshold is set, and all parameter items and time periods are traversed. If the deviation ratio of a certain parameter in a certain time period exceeds the parameter deviation ratio threshold, the parameter is regarded as the uneven cause.

Images