Automatic welding system for flower basket barrel tank production

Abstract

The invention relates to the technical field of weld joints, in particular to an automatic weld joint system for flower basket, bucket and tank production, and aims at products such as flower baskets, buckets and tanks which may have curved surface connection and irregular structures, the system accurately analyzes the curved surface connection difference degree by extracting three-dimensional point cloud and calculating parameters such as the curvature difference and the normal vector included angle, and the welding quality of the products is improved. The system can meet the requirements of welding seams of non-planar and complex structures, the limitation of a traditional automatic welding seam system on a simple planar structure is broken through, the universality of equipment is improved, the data acquisition module periodically acquires welding seam process images, and the system can find abnormities in time in the welding seam process instead of post detection in combination with real-time defect recognition of the recognition module; if defects are found in the early stage, weld parameters can be adjusted or repair welding can be started in time through the data analysis module, reworking of a whole batch of products caused by defect accumulation is avoided, and production loss is remarkably reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of welding seams, and particularly relates to an automatic welding seam system for flower basket barrel production. BACKGROUND

[0002] In the production process of a flower basket barrel (a metal container commonly used in the fields of chemical industry and warehousing), the quality of a welding seam directly determines the sealing property, structural strength and service life of a product. The joining parts of a barrel body and a barrel bottom, a barrel body and a barrel cover of the flower basket barrel are mostly curved surfaces or special-shaped structures. The welding seam line has complex curved surface transitions. In traditional welding, a worker needs to judge difficult welding seam areas (such as unevenly joined curved surfaces) by experience. The adaptability of welding parameters (such as current and speed) is poor due to the deviation in judgment, and defects such as incomplete penetration, pores and cracks occur.

[0003] Chinese patent application No. CN202010372941.3 discloses a flower basket barrel production process, which includes the following steps: S1: standing a barrel, after the raw materials are seam-welded, the barrel is stood, and after processing, a coning and flanging process is performed to obtain a barrel body in the shape of a round barrel; S2: edge rolling, the barrel body is conveyed to an edge rolling machine to perform edge rolling, and after the edge rolling is completed, the barrel body is processed by using a flower basket barrel rib expanding and flanging machine; S3: bottom sealing, after the barrel body processed is vertically turned, a bottom sealing operation is performed, so that the bottom of the barrel body is sealed; S4: side welding of a sealing strip, after the side welding seam is found, the sealing strip is welded on the outer side wall of the barrel body at the welding seam; S5: welding lug, after the barrel body is rotated by 180 degrees, a barrel lug is welded on the outer side wall of the barrel body near one side of the upper cover; S6: painting, after the welding lug process is completed, the barrel body is conveyed to a paint spraying room to perform paint spraying, and after the paint is naturally dried, the barrel body is packed and stored in a warehouse. The sealing strip is added at the welding seam of the outer side wall of the flower basket barrel, which greatly increases the sealing property of the flower basket barrel.

[0004] However, the prior art still has the following problems: The joining parts of a barrel body and a barrel bottom, a barrel body and a barrel cover of the flower basket barrel are mostly curved surfaces or special-shaped structures. The welding seam line has complex curved surface transitions. In traditional welding, a worker needs to judge difficult welding seam areas (such as unevenly joined curved surfaces) by experience. The adaptability of welding parameters (such as current and speed) is poor due to the deviation in judgment, and defects such as incomplete penetration, pores and cracks occur. SUMMARY

[0005] Therefore, the present application provides an automatic welding seam system for flower basket barrel production, which overcomes the problem in the prior art that a worker needs to judge difficult welding seam areas (such as unevenly joined curved surfaces) by experience, the adaptability of welding parameters (such as current and speed) is poor due to the deviation in judgment, and defects such as incomplete penetration, pores and cracks occur.

[0006] To achieve the above object, the present application provides an automatic welding system for flower basket barrel production. a welding module for welding the target to be welded; a data acquisition module connected with the welding module for acquiring the image of the target to be welded before welding and periodically acquiring the image of the target to be welded during welding; an identification module connected with the data acquisition module for identifying the welding area to be welded in the image of the target to be welded and the defect type in the image of the target to be welded; a division module connected with the identification module for determining the welding attention area of the welding area to be welded based on the curvature connection difference degree of the identified welding area to be welded and the welding abnormal area; a type determination module connected with the division module for calculating the welding difficulty tendency value based on the number and distribution uniformity of the welding attention area and determining the welding difficulty type of the target to be welded based on the welding difficulty tendency value; a data analysis module connected with the welding module, the data acquisition module, the identification module, the division module and the type determination module respectively for determining the processing mode for the target to be welded based on the defect type in the image of the target to be welded and the welding difficulty type of the target to be welded, including: determining the repair welding area and analyzing the influence degree of the repair welding area and the adjacent welding difficult area to determine the secondary repair welding mode based on the influence degree, or, outputting the target to be welded.

[0007] Further, the division module determines the curvature connection difference degree of the welding area to be welded, including: extracting the three-dimensional point cloud data and the depth image of the two welding surfaces to be welded in the welding area to be welded, determining a plurality of sampling points of the predetermined welding line, calculating the curvature value, normal vector direction and surface gap value of the two welding surfaces to be welded at each sampling point respectively, after standardizing the curvature difference, normal vector angle and gap value of each sampling point, calculating the connection difference component of each sampling point by weighted summation, taking the average value of the connection difference components of all sampling points as the curvature connection difference degree of the welding area to be welded; wherein the curvature difference is the difference between the curvature values of the two welding surfaces to be welded, and the normal vector angle is the angle between the normal vectors of the two welding surfaces to be welded.

[0008] Further, the division module determines the welding attention area of the welding area to be welded, including: determine image feature parameters of the to-be-welded seam area, compare each image feature parameter of the to-be-welded seam area with a preset parameter range, and if a single image feature parameter exceeds the preset parameter range, determine that the to-be-welded seam area is an abnormal area; The image feature parameters include a weld seam width range and a weld seam surface flatness. If the curved surface connection difference of the to-be-welded seam area is greater than or equal to a preset curved surface connection difference and / or the to-be-welded seam area is an abnormal area, the to-be-welded seam area is determined as a weld seam attention area.

[0009] Further, the type determination module is configured to calculate a weld seam difficulty tendency value, including: calculating a ratio of the number of the weld seam attention areas to a preset number to obtain a first weld seam difficulty sub-parameter, calculating a ratio of the distribution uniformity to a preset distribution uniformity to obtain a second weld seam difficulty sub-parameter, performing weighted summation on the first weld seam difficulty sub-parameter and the second weld seam difficulty sub-parameter to obtain the weld seam difficulty tendency value.

[0010] Further, the type determination module is configured to determine a weld seam difficulty type of the weld seam target based on the weld seam difficulty tendency value, including: If the weld seam difficulty tendency value is greater than or equal to a preset weld seam difficulty tendency value, the type determination module determines that the weld seam difficulty type of the weld seam target is a strong weld seam difficulty tendency; If the weld seam difficulty tendency value is less than the preset weld seam difficulty tendency value, the type determination module determines that the weld seam difficulty type of the weld seam target is a weak weld seam difficulty tendency.

[0011] Further, the data analysis module is configured to determine a processing mode for the weld seam target based on a defect type in the weld seam target image and the weld seam difficulty type of the weld seam target, including: If the defect type in the weld seam target image is a strong defect type and / or the weld seam difficulty type of the weld seam target is a strong weld seam difficulty tendency, the data analysis module determines to determine a repair welding area and analyze an influence degree of the repair welding area and an adjacent weld seam difficult area to determine a secondary repair welding mode based on the influence degree; If the defect type in the weld seam target image is a weak defect type and the weld seam difficulty type of the weld seam target is a weak weld seam difficulty tendency, the data analysis module determines to output the weld seam target.

[0012] Further, the data analysis module is further configured to analyze the influence degree of the repair welding area and the adjacent weld seam difficult area, including, determining defects in the weld seam target image, extracting a defect area of a defect type being a strong defect type to obtain a repair welding area, Calculate the overlapping area of the repair welding area and the adjacent welding difficult area to obtain the influence degree of the repair welding area and the adjacent welding difficult area.

[0013] Further, the data analysis module is used to determine the secondary repair welding mode based on the influence degree, comprising: If the overlapping area is greater than or equal to a preset overlapping area, the data analysis module determines to adopt a reverse segmented repair welding mode for secondary repair welding processing; If the overlapping area is less than the preset overlapping area, the data analysis module determines to adopt a stepped heat input mode for secondary repair welding processing.

[0014] Further, the reverse segmented repair welding is to divide the repair welding area into a plurality of segment sub-repair welding areas, and perform repair welding processing from one end away from the adjacent difficult area, and the segment sub-repair welding areas are sequentially repaired at a predetermined time interval.

[0015] Further, the stepped heat input secondary repair welding processing mode is to reduce the repair welding starting current to 70% of the initial current in the initial repair welding processing, and increase the current to the initial current value by gradually increasing.

[0016] Compared with the prior art, the beneficial effects of the present application are that in the present application, for products such as flower basket cans that may have curved surface connections and irregular structures, the system can accurately analyze the curved surface connection difference degree by extracting three-dimensional point cloud, calculating curvature difference, normal vector angle and other parameters, and can adapt to the welding seam demand of non-planar and complex structures, breaking through the limitations of traditional automatic welding seam systems on simple planar structures, improving the versatility of the equipment, and the data acquisition module periodically acquires welding process images, and combines the real-time defect recognition of the recognition module, so that the system can timely discover abnormalities in the welding process (rather than post-detection), if the defect is discovered early, the welding parameter can be adjusted in time or the repair welding can be started, so as to avoid the accumulation of defects to cause the whole batch of products to be reworked, and significantly reduce the production loss.

[0017] Further, in the present application, the three-dimensional point cloud data is extracted, the curvature difference, the normal vector angle and the gap value of the sampling points are calculated, and the curved surface connection difference degree is obtained by standardization and weighted summation, the traditional "curved surface connection complexity" depending on artificial subjective judgment is converted into a quantifiable numerical index, this process avoids the fuzziness of the complexity judgment of the curved surface transition part (such as the curved surface connection part of the barrel body and the barrel bottom of the flower basket can) in manual evaluation, so that the system can accurately identify high-difficulty welding areas, and provide objective basis for subsequent focusing on key areas and optimizing welding strategy.

[0018] Further, in the present application, through the dual judgment standard of the curvature surface connection difference degree and the image feature parameter abnormality, the comprehensive capture of the high-risk area of the weld is realized, wherein the curvature surface connection difference degree focuses on the inherent structural complexity of the to-be-welded area (such as the geometric defects of the curvature surface connection of the flower basket can), and the image feature parameter abnormality such as weld width and surface flatness is aimed at the dynamic defects (such as width out-of-tolerance caused by welding parameter fluctuation) occurring in the welding process. The combination of the two covers both the inherent structural difficulties and the postnatal process abnormalities, so that the system can accurately lock all areas that need to be paid attention to, and avoid missed judgment caused by a single standard (such as only focusing on structural differences and ignoring sudden width abnormalities in the process).

[0019] Further, in the present application, the number and distribution uniformity of the weld attention area are compared with the preset reference (preset number and preset distribution uniformity) to obtain the weld difficulty tendency value through ratio calculation and weighted summation, and the traditional welding difficulty depending on manual experience is converted into a quantifiable numerical index. This quantitative method avoids the ambiguity of subjective judgment, so that the difficulty evaluation of different weld targets has a unified standard, and the objectivity and comparability are improved. The weld difficulty tendency value integrates the two core dimensions of the number and distribution uniformity of the attention area: the more the number, the more intensive the high-risk areas that need to be focused on; the more uneven the distribution (such as concentrated in a key part or scattered in disorder), the higher the parameter adjustment frequency and the greater the operation complexity in the welding process. The weighted integration of the two overcomes the one-sidedness of a single index (such as only looking at the number), and can more comprehensively depict the actual complexity of the weld (such as the case of the flower basket can where the number of attention areas is small but the distribution is extremely scattered, and the difficulty can be reflected by the low distribution uniformity sub-parameter).

[0020] Further, in the present application, the determination result of the difficulty type directly determines the processing strategy of the data analysis module: the strong difficulty tendency target triggers the repair welding process (including repair welding area determination, influence degree analysis and secondary repair welding mode selection), and the weak difficulty tendency target with slight defects is directly output. This classification strategy enables the system to preferentially allocate resources (such as repair welding time and parameter adjustment effort) to high-risk targets, avoids excessive processing of low-difficulty targets (such as repeated detection without repair welding), reduces invalid operations while ensuring quality, and improves overall production efficiency.

[0021] Further, in the present application, for the target with strong defect type (such as crack and dense porosity) or strong weld difficulty tendency, the system preferentially starts the repair welding process to ensure that the high-risk weld is repaired in time; and for the target with weak defect type (such as slight undercut) and weak weld difficulty tendency, direct output can meet the quality requirements. This classification strategy avoids one-size-fits-all processing (such as repair welding or release for all defects), so that resources are concentrated on the welds that really need intervention, and the precision of quality control is improved.

[0022] Further, in the present application, by analyzing the correlation between the repair welding area and the difficult welding area (strong difficult tendency area), the repair welding process is included in the overall welding quality control system to avoid the problem of local optimization but overall failure caused by repair welding as an independent link. For example, for the repair welding area overlapping with the high-difficulty curved surface connection area, reverse segmented repair welding can reduce the influence of thermal stress on the curved surface connection, ensuring that the structural strength (such as pressure resistance and sealing) of the flower basket barrel is not affected by the secondary damage of repair welding.

[0023] Further, in the present application, by dividing the repair welding area into sub-areas ≤2mm, and welding again after 5-8s interval and cooling to below 300℃, the heat superposition caused by continuous repair welding is avoided, and the repair welding starts from one end away from the difficult area (such as sub-area 1 to sub-area 25), which can avoid the spread of repair welding heat to the difficult area with high stress, and reduce the risk of structural strength decline in the difficult area due to increased heat input (such as the curved surface connection of the flower basket barrel, excessive heat input will aggravate the deformation of the curved surface). This sequential design minimizes the impact of repair welding on the original high-difficulty area, ensuring the stability of the overall weld structure, and the starting current of repair welding is reduced to 70% of the initial current, which can reduce the thermal shock on the base material and the surrounding original weld of the repair welding area, and avoid problems such as base material burn-through and original weld fusion line cracking caused by instantaneous high temperature. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The structural block diagram of the automatic welding system for the flower basket barrel produced by the present application; Figure 2 The flowchart for determining the welding difficulty type of the welding target; Figure 3 The flowchart for determining the processing method for the welding target; Figure 4 The flowchart for determining the secondary repair welding method. DETAILED DESCRIPTION

[0025] In order to make the purpose and advantages of the present application more clear and explicit, the present application will be further described below in conjunction with examples; it should be understood that the specific examples described herein are only used to explain the present application, and do not limit the present application.

[0026] It should be noted that the data in the embodiment are obtained by comprehensive analysis and evaluation of historical data and corresponding historical determination results of the system in the application within 6 months before the present determination. Those skilled in the art can understand that the system in the application can select the value with the highest proportion as the preset standard parameter according to the data distribution, use weighted summation to obtain the value as the preset standard parameter, substitute each historical data into a specific formula and obtain the value by using the formula as the preset standard parameter, or other selection methods, as long as the system in the application can clearly define different specific situations in the single determination process through the obtained value.

[0027] The preferred embodiments of the present application will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application, and are not intended to limit the protection scope of the present application.

[0028] It should be noted that in the description of the present application, the terms "upper", "lower", "left", "right", "inner", "outer" and the like indicate the direction or positional relationship terms based on the direction or positional relationship shown in the drawings, which are only for the convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present application.

[0029] In addition, it should be noted that in the description of the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be connected inside two elements. Those skilled in the art can understand the specific meaning of the above terms in the present application according to the specific circumstances.

[0030] Please refer to Figure 1 As shown in the figure, it is a structural block diagram of the automatic welding system for flower basket can production of the present application.

[0031] The automatic welding system for flower basket can production provided in the embodiment comprises: A welding module is used to perform welding processing on a welding target; A data acquisition module is connected with the welding module, and is used to acquire an image of the welding target before welding processing and periodically acquire an image of the welding target during welding processing; An identification module is connected with the data acquisition module, and is used to identify a welding area in the image of the welding target and a defect type in the image of the welding target; a dividing module connected with the identifying module, configured to determine a welding seam attention area of the welding seam area to be welded based on the curvature connection difference degree of the identified welding seam area to be welded and the welding seam abnormal area; a type determining module connected with the dividing module, configured to calculate a welding seam difficulty tendency value based on the number and distribution uniformity of the welding seam attention area, and determine a welding seam difficulty type of the welding seam target based on the welding seam difficulty tendency value; a data analysis module connected with the welding seam module, the data acquisition module, the identifying module, the dividing module and the type determining module respectively, configured to determine a processing mode for the welding seam target based on the defect type in the welding seam target image and the welding seam difficulty type of the welding seam target, including: determining a repair welding area and analyzing the influence degree of the repair welding area and the adjacent welding seam difficult area to determine a secondary repair welding mode based on the influence degree, or, outputting the welding seam target.

[0032] Specifically, in the embodiment, the specific structure of the welding seam module is not limited, and only needs to realize the corresponding function, and a six-axis welding robot can be used.

[0033] Specifically, in the embodiment, the specific structure of the data acquisition module is not limited, and only needs to realize the corresponding function, and an industrial camera and a lens, a three-dimensional scanning device can be used.

[0034] Specifically, the specific structure of the identifying module, the dividing module, the type determining module and the data analysis module is not limited, which can be composed of a logic component including a field programmable processor, a computer and a microprocessor in the computer.

[0035] Specifically, in the embodiment, the data acquisition module acquires the image of the welding seam target to be welded (i.e. the image of the welding seam target to be welded) before welding, and acquires the image of the welding seam target (i.e. the image of the welding seam target) periodically during the welding process.

[0036] In the present application, for products such as flower basket cans that may have curved surface connection and irregular structure, the system can accurately analyze the curvature connection difference degree by extracting three-dimensional point cloud, calculating curvature difference, normal vector angle and other parameters, and can adapt to the welding seam demand of non-planar and complex structure, breaking through the limitation of traditional automatic welding seam system on simple planar structure, improving the universality of the equipment, and periodically acquiring the welding process image by the data acquisition module, combining with the real-time defect identification of the identifying module, so that the system can timely discover the abnormality during the welding process (rather than post-detection), if the defect is discovered early, the welding seam parameter can be adjusted in time or the repair welding is started, so as to avoid the accumulation of defects leading to the rework of the whole batch of products, and significantly reduce the production loss.

[0037] Specifically, the dividing module determines the surface connection difference degree of the to-be-welded seam region, including: extracting three-dimensional point cloud data and a depth image of two to-be-welded surfaces in the to-be-welded seam region, determining a plurality of sampling points of a predetermined weld line, respectively calculating the curvature value, the normal vector direction and the surface gap value of the two to-be-welded surfaces at each sampling point, after standardizing the curvature difference, the normal vector angle and the gap value of each sampling point, calculating the connection difference component of each sampling point by using a weighted summation method, taking the average value of the connection difference components of all sampling points as the surface connection difference degree of the to-be-welded seam region; wherein the curvature difference is the difference between the curvature values of the two to-be-welded surfaces, and the normal vector angle is the angle between the normal vectors of the two to-be-welded surfaces.

[0038] Specifically, in the embodiment, the data acquisition module acquires a to-be-welded seam target image, extracts three-dimensional point cloud data and a depth image of two to-be-welded surfaces (surface A and surface B) in the to-be-welded seam region, and the predetermined weld line is a theoretical connection line of the two surfaces, and five sampling points (S1-S5) are selected on the line; at S1, the curvature value of surface A is 0.02 mm⁻¹, the curvature value of surface B is 0.05 mm⁻¹, the curvature difference = 0.03 mm⁻¹; the angle between the normal vector of surface A and the normal vector of surface B is 15°; the surface gap value is 0.2 mm; similarly, the curvature difference (0.02 mm⁻¹, 0.04 mm⁻¹, 0.01 mm⁻¹, 0.03 mm⁻¹), the normal vector angle (10°, 20°, 8°, 12°), and the gap value (0.15 mm, 0.25 mm, 0.1 mm, 0.2 mm) at S2-S5 are calculated; the parameters are mapped to the [0, 1] interval, the maximum value of the curvature difference 0.04 mm⁻¹ corresponds to 1, the maximum value of the normal vector angle 20° corresponds to 1, and the maximum value of the gap value 0.25 mm corresponds to 1, to obtain the standardized value; the weights of the curvature difference, the normal vector angle and the gap value are respectively 0.3, 0.4 and 0.3, and the weighted summation is used to obtain the connection difference component of each sampling point: S1 is 0.3×0.75+0.4×0.75+0.3×0.8=0.75; S2-S5 are 0.6, 0.9, 0.4, 0.65 respectively; the average value of the connection difference components of S1-S5 is (0.75+0.6+0.9+0.4+0.65) / 5=0.66, that is, the surface connection difference degree of the to-be-welded seam region is 0.66.

[0039] In the present application, the "curved surface connection complexity" which is traditionally dependent on artificial subjective judgment is converted into a quantifiable numerical index by extracting three-dimensional point cloud data, calculating the curvature difference, normal vector angle and gap value of the sampling points, and obtaining the curved surface connection difference degree through standardization and weighted summation. This process avoids the ambiguity in artificial evaluation of the complexity of the curved surface transition part (such as the curved surface connection between the barrel body and the barrel bottom of a flower basket can), enabling the system to accurately identify high-difficulty welding areas and providing an objective basis for subsequent focusing on key areas and optimizing welding strategies.

[0040] Specifically, the division module determines a welding attention area of the to-be-welded seam area, including: determining image feature parameters of the to-be-welded seam area, comparing each image feature parameter of the to-be-welded seam area with a preset parameter range, and if a single image feature parameter exceeds the preset parameter range, determining that the to-be-welded seam area is an abnormal area; The image feature parameters include: welding seam width range, welding seam surface flatness; If the curved surface connection difference degree of the to-be-welded seam area is greater than or equal to a preset curved surface connection difference degree and / or the to-be-welded seam area is an abnormal area, the to-be-welded seam area is determined as a welding attention area.

[0041] Specifically, in the present embodiment, the preset curved surface connection difference degree can be determined in the following way: based on the material (such as cold-rolled steel sheet, thickness 1.2mm) and welding process (such as carbon dioxide gas shielded welding) of the to-be-welded seam target (such as the connection part of the barrel body and the barrel bottom of a flower basket can), the theoretically allowed maximum curved surface connection difference threshold is determined, and the geometric tolerance requirements (such as curved surface gap ≤0.3mm, normal vector angle ≤25°) for this type of welding seam in the industry standard are referred to, and the theoretically upper limit value of the connection difference component (such as the connection difference component of a single sampling point ≤1.0) is converted; 100 samples of the same type of to-be-welded seam are selected, and the actual difference degree values of the samples are obtained according to the calculation method of the curved surface connection difference degree (extracting three-dimensional point cloud, calculating sampling point parameters and weighted summation), and the defect rate (such as the proportion of defects such as incomplete penetration and porosity) after welding is recorded at the same time. Statistics show that when the difference degree is ≤0.6, the welding defect rate is stable within 3%; when the difference degree is >0.6, the defect rate increases linearly with the increase of the difference degree (such as 8% when the difference degree is 0.7, and 15% when the difference degree is 0.8); taking "defect rate ≤5%" as the quality control target, the preset curved surface connection difference degree is determined in combination with the statistical data. In the experiment, the defect rate corresponding to the difference degree 0.65 is 5%, so the preset curved surface connection difference degree is set to 0.65.

[0042] In the present application, the comprehensive capture of high-risk areas of the weld is realized through the dual judgment criteria of the curvature surface connection difference degree and the image feature parameter anomaly, wherein the curvature surface connection difference degree focuses on the inherent structural complexity of the to-be-welded area (such as the geometric defects of the curvature surface connection of the flower basket can), and the image feature parameter anomaly such as the weld width and the surface flatness is aimed at the dynamic defects (such as the width out-of-tolerance caused by the welding parameter fluctuation) occurring in the welding process. The combination of the two covers both the inherent structural difficulties and the acquired process anomalies, so that the system can accurately lock all areas that need to be paid attention to, and avoid the missed judgment caused by a single standard (such as only focusing on the structural difference and ignoring the sudden width anomaly in the process).

[0043] Specifically, the type determination module is used to calculate a weld difficulty tendency value, including: calculating a ratio of the number of the weld attention areas to a preset number to obtain a first weld difficulty sub-parameter, calculating a ratio of the distribution uniformity to a preset distribution uniformity to obtain a second weld difficulty sub-parameter, performing weighted summation on the first weld difficulty sub-parameter and the second weld difficulty sub-parameter to obtain the weld difficulty tendency value.

[0044] Specifically, in the present embodiment, the preset number can be determined by the following method: based on the overall weld length and the structure segmentation characteristics of the to-be-welded target, the total weld area is divided into a plurality of basic evaluation units (such as 100 mm of weld as one unit), and the upper limit of the maximum number of attention areas that may exist in theory is determined (such as 30 units corresponding to a maximum possible attention area number of 30); the qualified product data of the same type of weld target (such as the flower basket can) in the past three months is collected, and the actual number distribution of the weld attention areas is counted. For example, in 1000 qualified samples, the number of attention areas is concentrated in 2-6, 95% of the sample number is ≤8, and when the number exceeds 8, the probability of re-welding after welding increases to more than 15% (exceeding the quality target); taking “the re-welding rate after welding ≤5%” as the quality target, the critical number value is determined in combination with the statistical data. If the data shows that the re-welding rate is 5% when the number of attention areas is 8, the preset number is initially set to 8.

[0045] Specifically, in the present embodiment, the distribution uniformity is determined in the following manner: the three-dimensional space or the two-dimensional projection plane where the to-be-welded area is located is divided into a predetermined number of grid units; the area ratio or the number ratio of the weld attention areas in each grid unit is counted to generate a density distribution matrix; the discrete degree of the attention area density between the grid units is evaluated by using the standard deviation; and the discrete index value is mapped to the [0, 1] interval to obtain the distribution uniformity.

[0046] In the present application, the number and distribution uniformity of the weld attention area are subjected to ratio calculation with the preset reference (preset number and preset distribution uniformity), and then weighted summation is performed to obtain the weld difficulty tendency value, thereby converting the traditional welding difficulty depending on artificial experience into a quantifiable numerical index. This quantification method avoids the ambiguity of subjective judgment, enables the difficulty evaluation of different weld targets to have a unified standard, and improves the objectivity and comparability. The weld difficulty tendency value integrates the two core dimensions of the number and distribution uniformity of the attention area: the more the number, the more intensive the high-risk areas that need to be focused on; the more uneven the distribution (such as being concentrated in a certain key part or being scattered and disordered), the higher the parameter adjustment frequency and the greater the operation complexity in the welding process. The weighted integration of the two overcomes the one-sidedness of a single index (such as only looking at the number), and can more comprehensively depict the actual complexity of the weld (for example, in the girth weld of the flower basket barrel, the number of attention areas is small but the distribution is extremely scattered, and its difficulty can be reflected by the low distribution uniformity sub-parameter).

[0047] Referring to Figure 2 as shown in the figure, which is a determination flowchart of the weld difficulty type of the weld target.

[0048] Specifically, the type determination module is configured to determine the weld difficulty type of the weld target based on the weld difficulty tendency value, including: If the weld difficulty tendency value is greater than or equal to the preset weld difficulty tendency value, the type determination module determines that the weld difficulty type of the weld target is a strong weld difficulty tendency. If the weld difficulty tendency value is less than the preset weld difficulty tendency value, the type determination module determines that the weld difficulty type of the weld target is a weak weld difficulty tendency.

[0049] Specifically, in the present embodiment, the preset weld difficulty tendency value can be determined by the following method: the mapping relationship between the weld difficulty tendency value and the welding quality is determined: the higher the tendency value, the more the number of weld attention areas or the more uneven the distribution, the greater the welding difficulty, and the higher the defect rate (such as incomplete penetration, crack) and the re-welding rate after welding. Based on this, the re-welding rate after welding ≤5% is taken as the core quality target, and the corresponding weld difficulty tendency value critical value is inversely deduced; 1000 sample data of the same type of weld target (such as flower basket barrel) within the last 3 months are collected, and the actual tendency value of each sample is obtained according to the calculation logic of the number of attention areas, the distribution uniformity, and the weld difficulty tendency value, and the re-welding rate after welding is recorded. Statistics show that when the tendency value ≤0.8, the re-welding rate is stable within 3%; when the tendency value >0.8, the re-welding rate increases significantly with the increase of the tendency value (for example, the re-welding rate is 8% when the tendency value is 0.9, and 12% when the tendency value is 1.0); combined with the quality target "re-welding rate ≤5%", the corresponding tendency value is selected from the statistical data, if the data shows that the re-welding rate is 5% when the tendency value is 0.85, the preset weld difficulty tendency value is initially set to 0.85.

[0050] In the present application, the determination result of the difficulty type directly determines the processing strategy of the data analysis module: strong difficulty tends to trigger the target repair welding process (including repair welding area determination, impact degree analysis and secondary repair welding mode selection), and the target with weak difficulty and slight defect is directly output. This classification strategy mode enables the system to preferentially allocate resources (such as repair welding time and parameter adjustment effort) to high-risk targets, avoids excessive processing of low-difficulty targets (such as repeated detection without repair welding), reduces invalid operations while ensuring quality, and improves overall production efficiency.

[0051] Please refer to Figure 3 for the determination flowchart of the processing mode for the weld target.

[0052] Specifically, the data analysis module is configured to determine the processing mode for the weld target based on the defect type in the weld target image and the weld difficulty type of the weld target, including: If the defect type in the weld target image is a strong defect type and / or the weld difficulty type of the weld target is a strong weld difficulty tendency, the data analysis module determines to determine a repair welding area and analyze the impact degree of the repair welding area and the adjacent weld difficult area to determine the secondary repair welding mode based on the impact degree. If the defect type in the weld target image is a weak defect type and the weld difficulty type of the weld target is a weak weld difficulty tendency, the data analysis module determines to output the weld target.

[0053] In the present application, for the target with strong defect type (such as crack, dense porosity) or strong weld difficulty tendency, the system preferentially starts the repair welding process to ensure that the high-risk weld is repaired in time; and for the target with only weak defect type (such as slight undercutting) and weak weld difficulty tendency, direct output can meet the quality requirements. This classification strategy avoids one-size-fits-all processing (such as repair welding or release for all defects), concentrates resources on welds that really need intervention, and improves the accuracy of quality control.

[0054] Specifically, the data analysis module is further configured to analyze the impact degree of the repair welding area and the adjacent weld difficult area, including, determining the defect in the weld target image, extracting the defect area with a strong defect type to obtain a repair welding area, calculating the overlapping area of the repair welding area and the adjacent weld difficult area to obtain the impact degree of the repair welding area and the adjacent weld difficult area.

[0055] Please refer to Figure 4 for the determination flowchart of the secondary repair welding mode.

[0056] Specifically, the data analysis module is used to determine the secondary repair welding mode based on the influence degree, including: If the overlap area is greater than or equal to the preset overlap area, the data analysis module determines to use the reverse segmented repair welding mode for secondary repair welding processing. If the overlap area is less than the preset overlap area, the data analysis module determines to use the stepped heat input mode for secondary repair welding processing.

[0057] Specifically, in the embodiment, the preset overlap area can be determined by the following method: taking the actual area of the repair welding area as the reference standard, defining the overlap area of the repair welding area and the adjacent difficult area of the weld as the intersection area in three-dimensional space, and the unit is mm² (because the repair welding area of the metal member such as the flower basket tank is usually 100-300 mm², the overlap area needs to adapt to this scale); collect repair welding records of the same type of weld target (such as flower basket tank) in the past three months, a total of 500 samples, and statistics the secondary defect rate (such as cracks, unfused, etc. at the junction of the repair welding area and the difficult area) after repair welding under different overlap areas; the data shows that when the overlap area is ≤40 mm², the secondary defect rate is stable within 3%; when the overlap area is >40 mm², the secondary defect rate increases significantly with the increase of the area (such as 6% at 50 mm², 10% at 60 mm²), because the heat affected zone of the overlap area is easy to cause stress concentration; taking “secondary defect rate ≤5% after repair welding” as the quality target, the critical overlap area is selected from the statistical data; if the data shows that the secondary defect rate is 5% when the overlap area is 50 mm², the preset overlap area is initially set to 50 mm²; if the repair welding process is adjusted (such as using low-stress repair welding materials), 300 samples need to be collected for verification, if the secondary defect rate corresponding to the overlap area of 60 mm² under the new process is still ≤5%, the preset value is adjusted to 60 mm², which adapts to the improved process capacity.

[0058] In the present application, by analyzing the correlation between the repair welding area and the difficult area of the weld (strong and difficult area), the repair welding process is included in the overall weld quality control system, avoiding the problem of local optimization but overall failure caused by repair welding as an independent link. For example, for the repair welding area overlapping with the high-difficulty curved surface connection area, reverse segmented repair welding can reduce the influence of thermal stress on the curved surface connection, and ensure that the structural strength (such as pressure resistance and sealing performance) of the flower basket tank is not affected by the secondary damage of repair welding.

[0059] Specifically, the reverse segmented repair welding is to divide the repair welding area into a plurality of segment repair welding areas, and perform repair welding processing from one end away from the adjacent difficult area, and the segment repair welding areas are sequentially repaired at a predetermined time interval.

[0060] Specifically, in the embodiment, the length of each sub-welding area is ≤2mm, the predetermined time is 5s-8s, and the welding treatment is performed after the temperature drops below 300℃ to avoid continuous heat accumulation.

[0061] Specifically, in the embodiment, the welding area (length 50mm, width 8mm, total area 400mm²) of the 200L flower basket barrel body and barrel bottom ring weld is taken as an example, the overlapping area of the area and the adjacent welding difficult area (high stress area with a difference of 0.72 in curved surface connection) is 60mm² (≥ preset overlapping area 50mm²), and reverse segmented welding is adopted: the welding area is divided into 25 sub-welding areas along the length direction, each with a length of 2mm (meeting the requirement of “≤2mm”), and labeled as sub-area 1 to sub-area 25 in order of distance from the difficult area (sub-area 1 far from the difficult area, and sub-area 25 adjacent to the difficult area); starting from sub-area 1, carbon dioxide gas shielded welding is adopted, with a welding current of 180A and a voltage of 22V, and a single segment welding time of about 3s; after completion, pause for 5s (in the interval of “5s-8s”), and after the temperature of sub-area 1 drops to 280℃ (<300℃) is monitored by an infrared thermometer, the adjacent sub-area 2 is welded, and so on, until the welding of sub-area 25 is completed; through reverse order and interval cooling, the heat accumulation to the difficult area is avoided, the maximum hardness of the heat affected zone (HAZ) after welding is controlled below 250HV (120% lower than the hardness of the base material), and the risk of cracks caused by heat accumulation is eliminated.

[0062] Specifically, the secondary welding treatment mode of the stepped heat input is to reduce the welding starting current to 70% of the initial current in the primary welding treatment, and increase the current to the initial current value by gradually increasing.

[0063] Specifically, in this embodiment, the same type of flower basket barrel tank barrel body longitudinal weld repair area (length 30mm, width 6mm, total area 180mm²) is taken as an example, the area overlaps with the adjacent weld difficult area by 30mm² (< preset overlap area 50mm²), the initial current of the first repair welding is 200A, and the repair welding is performed by using a step heat input: the starting current of the repair welding is reduced to 70% of the initial current of the first repair welding, that is, 140A; the current is increased by 20A for three times, and finally reaches the initial current 200A, and the specific parameters are as follows: the first section (10mm in length): current 140A, voltage 20V, welding time 4s; the second section (10mm in length): current 160A, voltage 21V, welding time 3.5s; the third section (10mm in length): current 180A, voltage 21.5V, welding time 3s; the fourth section (remaining 10mm in length): current 200A, voltage 22V, welding time 3s; the heat shock on the base material is reduced by using a low starting current, and the current is gradually increased to ensure that the penetration depth meets the standard (the final penetration depth is greater than or equal to 1.2mm, which is matched with the thickness of the base material), and the surface flatness error of the weld after repair welding is less than or equal to 0.3mm, without secondary defects such as pores and unfused.

[0064] In the present application, by dividing the repair welding area into sub-areas of ≤2mm, and welding again after 5-8s interval and cooling to below 300℃, the heat superposition caused by continuous repair welding is avoided, the repair welding starts from one end away from the difficult area (such as sub-area 1 to sub-area 25), which can avoid the spread of repair welding heat to the difficult area with high stress, and reduce the risk of structural strength decline of the difficult area caused by increased heat input (such as the curved surface connection of the flower basket barrel, excessive heat input will aggravate the deformation of the curved surface), this sequential design minimizes the impact of repair welding on the original difficult area, ensures the stability of the overall weld structure, and reduces the heat shock on the base material and the original weld around the repair welding area by reducing the starting current of repair welding to 70% of the initial current, thereby avoiding problems such as base material burn-through and original weld fusion line cracking caused by instantaneous high temperature.

[0065] Thus, the technical solutions of the present application have been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical solutions after the changes or replacements will fall within the protection scope of the present application.

[0066] The above description is only the preferred embodiments of the present application and is not used to limit the present application; for those skilled in the art, the present application can have various changes and variations, and any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An automated welding system for the production of flower baskets, barrels, and cans, characterized in that, include: Welding module, used to perform weld treatment on the target weld; The data acquisition module, which is connected to the weld module, is used to acquire the target image of the weld before weld processing and to periodically acquire the target image of the weld during the weld processing process. The identification module, which is connected to the data acquisition module, is used to identify the weld area in the weld target image and the defect type in the weld target image; A segmentation module, which is connected to the identification module, is used to determine the weld interest area of ​​the weld area based on the identified surface connection difference of the weld area and the weld abnormal area. A type determination module, which is connected to the division module, is used to calculate the weld difficulty tendency value based on the number and distribution uniformity of the weld attention area, and to determine the weld difficulty type of the weld target based on the weld difficulty tendency value. A data analysis module, connected to the weld seam module, the data acquisition module, the identification module, the segmentation module, and the type confirmation module, is used to determine the processing method for the weld seam target based on the defect type in the weld seam target image and the weld seam difficulty type of the weld seam target, including: The area to be repaired is determined, and the degree of influence between the repair area and the adjacent difficult areas of the weld is analyzed in order to determine the secondary repair welding method based on the degree of influence. Alternatively, output the weld target.

2. The automated welding system for producing flower basket barrels and cans according to claim 1, characterized in that, The segmentation module determines the surface connection difference degree of the region to be welded, including: Extract the three-dimensional point cloud data and depth image of the two surfaces to be welded in the area to be welded. Multiple sampling points are determined for the predetermined weld line, and the curvature value, normal vector direction, and surface gap value of the two surfaces to be welded are calculated at each sampling point. After standardizing the curvature difference, normal vector angle, and gap value at each sampling point, a weighted summation method is used to calculate the connection difference component at each sampling point. The average value of the connection difference components of all sampling points is taken as the surface connection difference degree of the area to be welded. Wherein, the curvature difference is the difference in curvature values ​​between the two surfaces to be welded, and the angle between the normal vectors is the angle between the normal vectors of the two surfaces to be welded.

3. The automated welding system for producing flower basket barrels and cans according to claim 2, characterized in that, The segmentation module determines the weld interest area of ​​the region to be welded, including: The image feature parameters of the area to be welded are determined, and each image feature parameter of the area to be welded is compared with a preset parameter range. If a single image feature parameter exceeds the preset parameter range, the area to be welded is determined to be an abnormal area. Image feature parameters include: weld width range and weld surface flatness; If the surface connection difference of the area to be welded is greater than or equal to the preset surface connection difference and / or the area to be welded is an abnormal area, then the area to be welded is determined as the weld interest area.

4. The automated welding system for producing flower basket barrels and cans according to claim 3, characterized in that, The type determination module is used to calculate the weld difficulty tendency value, including: The ratio of the number of weld seam interest areas to the preset number is used to obtain the first weld seam difficulty sub-parameter. The ratio of the distribution uniformity to the preset distribution uniformity is calculated to obtain the second weld difficulty sub-parameter. The first weld difficulty sub-parameter and the second weld difficulty sub-parameter are weighted and summed to obtain the weld difficulty tendency value.

5. The automated welding system for producing flower basket barrels and cans according to claim 4, characterized in that, The type determination module is used to determine the weld difficulty type of the weld target based on the weld difficulty tendency value, including: If the weld difficulty tendency value is greater than or equal to the preset weld difficulty tendency value, the type determination module determines that the weld difficulty type of the weld target is strong weld difficulty tendency. If the weld difficulty tendency value is less than the preset weld difficulty tendency value, then the type determination module determines that the weld difficulty type of the weld target is weak weld difficulty tendency.

6. The automated welding system for producing flower basket barrels and cans according to claim 5, characterized in that, The data analysis module is used to determine the processing method for the weld target based on the defect type in the weld target image and the weld difficulty type of the weld target, including: If the defect type in the weld target image is a strong defect type and / or the weld difficulty type of the weld target is a strong weld difficulty tendency, then the data analysis module determines the repair welding area and analyzes the degree of influence between the repair welding area and the adjacent difficult weld area to determine the secondary repair welding method based on the degree of influence. If the defect type in the weld target image is a weak defect type and the weld difficulty type of the weld target is a weak weld difficulty tendency, then the data analysis module determines and outputs the weld target.

7. The automated welding system for producing flower basket barrels and cans according to claim 6, characterized in that, The data analysis module is also used to analyze the degree of influence between the repair welding area and the adjacent difficult welding areas, including, Defects in the target weld image are identified, and defect areas with strong defect types are extracted to obtain the repair weld area. Calculate the overlap area between the repair welding area and the adjacent difficult welding area to obtain the degree of influence between the repair welding area and the adjacent difficult welding area.

8. The automated welding system for producing flower basket barrels and cans according to claim 7, characterized in that, The data analysis module is used to determine the secondary welding method based on the degree of impact, including: If the overlapping area is greater than or equal to the preset overlapping area, the data analysis module determines that a reverse segmented welding method should be used for secondary welding. If the overlapping area is less than the preset overlapping area, the data analysis module determines that a stepped heat input method should be used for secondary welding.

9. The automated welding system for producing flower basket barrels and cans according to claim 8, characterized in that, The reverse segmented welding process involves dividing the welding area into several welding segments, performing welding from the end furthest from the adjacent difficult area, and sequentially performing welding on each welding segment at predetermined intervals.

10. The automated welding system for producing flower basket barrels and cans according to claim 8, characterized in that, The secondary welding process of the stepped heat input involves reducing the welding start current to 70% of the initial current during the first welding process, and then gradually increasing the current to the initial current value.

Images