Bucket rod box welding method and welding system based on machine vision
Through welding methods and systems based on machine vision, combined with welding robots and visual recognition technology, the problems of difficulty in ensuring welding quality and low efficiency in the prior art are solved, and efficient and high-quality rod box welding is achieved.
Patent Information
- Application Number
- CN202510177013.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-18
- Publication Date
- 2025-04-15
- Estimated Expiration
- 2045-02-18
AI Technical Summary
In the prior art, when welding a bucket rod box, it is difficult to ensure welding quality and the welding efficiency is low.
Using machine vision-based welding methods and systems, weld the initial welds are quickly welded by establishing a standard welding appearance database, and using welding robots to quickly weld the initial welds, and through machine vision identification and complementary welding, we ensure welding quality.
An efficient welding process is achieved, while ensuring high standards of welding molding quality, the yield is controlled above 99%, and the grain coarsification of welding defects and heat-affected zones is avoided.
Smart Images

Figure CN119820163A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of automatic welding, and in particular to a boom box welding method and a welding system based on machine vision. Background Art
[0002] The arm box, also often simply called the arm or forearm, is an important part of the excavator, mainly controlling the bucket for excavation and loading operations. Figure 1 The boom box 10 is generally welded by two parallel steel plates 11 with internal vertically distributed internal support ribs 12. It is a three-dimensional shape and is also provided with multiple axial holes 101. The thickness of the parallel steel plates 11 mainly depends on factors such as the specific machine model, design load, working conditions and material strength. Generally, the minimum thickness is 6 mm, and the thickness of a large excavator may be as thick as 30 mm.
[0003] In the prior art, various tooling or auxiliary tooling is usually set to complete the rapid welding of the boom box. Although it can improve work efficiency to a certain extent, the welding quality is difficult to guarantee.
[0004] For example, the invention patent with application number CN202410868911.X discloses an excavator boom welding mechanism and method thereof. By setting a lifting assembly and setting a leveling cylinder on the surface of the lifting platform, it is convenient to adjust the height of the two ends of the positioning bracket after the excavator boom main frame is lifted to the surface of the lifting platform, so as to facilitate the fixing of the excavator boom main frame, reduce the difficulty of fixing the excavator boom main frame, and improve the welding efficiency. By setting a rotating assembly, when one side of the excavator boom main frame is welded, the first docking plate is driven to rotate through the driving device, and the first docking plate drives the excavator boom main frame to flip through the second docking plate, so as to realize the rapid flipping of the excavator boom main frame and further improve the welding efficiency.
[0005] For example, the utility model patent with application number CN202220058116.0 discloses a boom processing and welding device. Through a sliding clamping structure and a driving structure, when the boom is welded and fixed, the driving structure drives the moving block to move on two slide rails. When the rotating clamping assembly contacts the boom, the four rotating clamping assemblies rotate on four hinge seats, so that both ends of the four rotating clamping assemblies contact the boom, thereby achieving the purpose of fixing the irregular boom and effectively reducing the clamping operation time of the boom.
[0006] Therefore, it is necessary to provide a technology that has both faster welding efficiency and self-checking welding quality, so as to be able to process a boom box with high welding quality in one go. Summary of the invention
[0007] The embodiments of the present application provide a boom box welding method and welding system based on machine vision, which not only have faster welding efficiency, but also have a self-inspection function, and can process a boom box with high welding quality in one go.
[0008] A first aspect of an embodiment of the present application provides a boom box welding method based on machine vision, which comprises the following steps in sequence:
[0009] S10, establishing a standard welding appearance database based on a boom box having qualified welding appearance information, wherein the qualified welding appearance information includes a weld path and a weld three-dimensional shape;
[0010] S20, based on the standard welding appearance database, the three-dimensional shape of the weld is reduced by 90%-95% in the same proportion to obtain initial welding parameters, and the welding robot quickly welds the boom box to be welded according to the weld path and the initial welding parameters to obtain an initial boom box;
[0011] S30, identifying appearance information of an initial boom box by machine vision, and based on the three-dimensional shape of the weld, precisely welding the initial boom box by the welding robot in a complementary manner;
[0012] S40, again identifying the appearance information of the initial boom box through machine vision to form primary appearance data information, comparing the primary appearance data information with the qualified welding appearance information, if the difference between the three-dimensional shape size of the weld in the primary appearance information and the size of the three-dimensional shape of the weld is within the first threshold range, it is determined that welding is completed.
[0013] In a possible implementation, the comparing the primary appearance data information and the qualified welding appearance information specifically includes: comparing the weld coverage of the three-dimensional shape of the weld and the convexity of the weld coverage between the two.
[0014] In a possible implementation, within the first threshold range means that the error ratio of the weld coverage of the three-dimensional shape of the weld between the two, and the error ratio of the convexity of the weld coverage of the three-dimensional shape of the weld between the two are within ±10%.
[0015] In one possible implementation, in step S30, the initial boom box is precisely welded by the welding robot in a complementary manner, specifically, the initial welding parameters are proportionally increased by 1.04 to 1.07 times to form a compensation parameter, and the increased portion is the compensation welding parameter. The initial boom box is precisely welded by the welding robot according to the weld path and the compensation welding parameter.
[0016] In one possible implementation, in step S40, if the difference between the size of the weld three-dimensional shape in the first appearance information and the size of the weld three-dimensional shape exceeds the first threshold range, the appearance information of the boom box is identified by machine vision for a third time, and based on the weld three-dimensional shape, the initial boom box is precision-welded a second time by the welding robot in a complementary manner.
[0017] In one possible implementation, the initial boom box is precision welded a second time by the welding robot in a complementary manner, specifically, the first compensation parameter is proportionally increased by 1.05 times to 1.08 times, and the increased portion is the secondary compensation welding parameter. The welding robot continues to precision weld the initial boom box according to the weld path and the secondary compensation welding parameter.
[0018] In one possible implementation, the boom box to be welded is directionally rotated in the horizontal direction at a rotation frequency of 180° by a rotary positioner, and the vision robot and the welding robot are symmetrically distributed on both sides of the rotation direction of the boom box to be welded, wherein the vision robot provides machine vision.
[0019] In one possible implementation, the interval between quickly welding the boom box to be welded in step S20 and precision welding the initial boom box in a complementary manner in step S30 is 2-4 minutes, and the interval between precision welding the initial boom box according to the compensation welding parameters in step S30 and precision welding the initial boom box for the second time in a complementary manner in step S40 is 2-4 minutes.
[0020] A second aspect of an embodiment of the present application provides a boom box welding system based on machine vision, which is used to implement the boom box welding method based on machine vision as described above, wherein the welding system includes a control cabinet, a rotary positioner, a welding robot and a visual robot, the rotary positioner having a side output end, the rotary positioner drives the side output end to rotate in a horizontal direction, the side output end is connected to a fixed tooling for fixing the boom box, wherein the boom boxes are symmetrically distributed on both sides of the fixed tooling, the welding robot and the visual robot are respectively facing the two sides of the fixed tooling, so that when the welding robot is welding the boom box on one side, the visual robot can visually scan and identify the boom box on the other side.
[0021] In a possible implementation, the fixing fixture includes a main support and an inner supporting mechanism and an outer pressing mechanism symmetrically distributed on both sides of the main support;
[0022] The inner support mechanism includes a plurality of telescopic cylinders distributed at intervals along the outer side of a predetermined area, wherein the predetermined area is used to place a boom box, and the output ends of the telescopic cylinders are connected to support blocks, so that the plurality of telescopic cylinders can correspondingly drive the support blocks to abut against the inner support ribs between the two parallel steel plates of the boom box above and below the boom box, respectively, to support and fix the boom box;
[0023] The external clamping mechanism includes a plurality of clamping cylinders spaced apart along the outer edge of the predetermined area, the output end of the clamping cylinder is vertically connected to a clamping frame, and the front end of the clamping frame is connected to a positioning piece that matches the axial hole of the boom box.
[0024] Beneficial effects: Compared with the prior art, the boom box welding method and welding system based on machine vision provided by the present application integrate both machine vision technology and lane welding technology. First, the initial weld is quickly welded by a welding robot, and then complementary precision welding is performed after machine vision analysis and judgment. It not only has faster welding efficiency, but also can effectively ensure the welding forming quality, and can avoid the risk of grain coarsening and welding defects in the heat-affected zone that are easily caused by temperature accumulation between weld layers.
[0025] The welding robot performs each welding in a complementary manner, including initial welding, primary compensation welding and secondary compensation welding, rather than direct full welding. The robot can make full use of the analysis and judgment of machine vision to ensure that the weld is extremely close to the three-dimensional shape of the weld of the qualified welding appearance information, so that the one-time welding forming of the boom box can be completed in the form of multiple welding passes, with high forming efficiency and high forming quality, and the yield rate of the one-time formed boom box can be controlled at more than 99%;
[0026] The interval between two adjacent supplementary weldings is controlled within 2-4 minutes, which can further ensure the welding quality of the boom box;
[0027] In the welding system, the boom box is first positioned and fixed by the telescopic cylinder in cooperation with the support block, which is easy to operate. Then the boom box is pressed by the clamping cylinder in cooperation with the clamping wheel or the clamping pad, which can effectively ensure the stability of the boom box in the welding process, thereby effectively ensuring the one-time forming welding quality.
[0028] These and other objects, features and advantages of the present invention will be fully reflected in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 The schematic diagram of the structure of the boom box in the prior art is shown.
[0030] Figure 2A flow chart of the boom box welding method based on machine vision of the present application is shown.
[0031] Figure 3 A schematic diagram of the process from initial welding to primary supplementary welding in the boom box welding method based on machine vision of the present application is shown.
[0032] Figure 4 A structural schematic diagram of the boom box welding system of the present application based on machine vision is shown.
[0033] Figure 5 A partially enlarged structural schematic diagram of the boom box welding system of the present application based on machine vision is shown. DETAILED DESCRIPTION
[0034] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.
[0035] Those skilled in the art should understand that, in the disclosure of the specification, the orientation or position relationship indicated by the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. are based on the orientation or position relationship shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the above terms should not be understood as limiting the present invention.
[0036] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.
[0037] refer to Figure 2 The first aspect of the embodiment of the present application provides a boom box welding method based on machine vision, which comprises the following steps in sequence:
[0038] S10, establishing a standard welding appearance database based on the boom box with qualified welding appearance information, wherein the qualified welding appearance information includes a weld path and a weld three-dimensional shape, wherein the weld three-dimensional shape includes an area covered or occupied by the weld on the boom box and a protruding portion (or convexity) on the covered or occupied area;
[0039] S20, based on the standard welding appearance database, the three-dimensional shape of the weld is reduced by 90%-95% in the same proportion to obtain initial welding parameters, that is, initial weld shape information, and the welding robot quickly welds the boom box to be welded according to the weld path and the initial welding parameters, thereby greatly ensuring the welding efficiency and obtaining the initial boom box (such as Figure 3 Part a), complete the initial welding;
[0040] S30, identifying the appearance information of the initial boom box through machine vision, mainly identifying the three-dimensional shape information of the weld obtained by the initial rapid welding, and based on the three-dimensional shape of the weld on the boom box with qualified welding appearance information, the initial boom box is precisely welded by the welding robot in a complementary manner, that is, supplemented on the basis of the initial welding, wherein the supplemented form is close to the three-dimensional shape of the weld, rather than directly forming a full weld or exceeding the range of the three-dimensional shape of the weld, and at this time, one supplementary welding is completed;
[0041] S40, the appearance information of the initial boom box is recognized again by machine vision to form a primary appearance data information, and then the primary appearance data information is compared with the qualified welding appearance information. If the difference between the three-dimensional shape size of the weld in the primary appearance information and the size of the three-dimensional shape of the weld on the boom box with qualified welding appearance information is within a first threshold range, it is determined that the welding is completed, that is, Figure 3 Part b.
[0042] In the process of welding the boom box in one go, it is necessary to ensure that the amount of initial welding is 90%-95% of the full welding amount, that is, the welding is performed by reducing the three-dimensional shape of the weld by 90%-95% in the same proportion, and the position of the weld remains unchanged. Then, the difference between the weld on the initial boom box obtained by the initial welding and the weld on the qualified boom box is confirmed by comparison through machine vision. Then, based on this difference, two passes of welding are performed on the basis of the initial boom box weld in the form of supplementation, so that after the two passes of welding are supplemented, the weld on the initial boom box can be greatly close to the weld on the qualified boom box. Then, the comparison is performed again through machine vision. If the difference between the two is within the first threshold range, it is determined that the welding is completed. In the process of supplementary welding to ensure that the weld on the qualified boom box can be greatly close, considering the existence of a certain welding error, it is best not to reach the level of full welding or exceed the range of the three-dimensional shape of the weld, as long as the difference between the welds between the two is within the first threshold range. Obviously, even if the welding error causes the welding to eventually exceed the range of the three-dimensional shape of the weld, as long as the difference between the two is within the first threshold range, the welding quality requirements are still met and the welding can be confirmed to be completed.
[0043] In addition, taking into account the existence of welding errors, the weld surface of the initial boom box obtained after the initial rapid welding may not be very smooth, that is, it may have slightly raised or sunken parts. In this case, in the second welding as supplementary welding, based on the specific surface information of the weld identified by machine vision, the supplementary welding as precision welding will compensate for this defect as much as possible, so that the final difference is within the first threshold range.
[0044] For each boom box, during the automatic welding process, the three-dimensional shape of the weld on the boom box with qualified welding appearance information is used as a reference comparison standard. Therefore, the welding forming quality of the boom box can be guaranteed.
[0045] In one embodiment, the comparing the primary appearance data information and the qualified welding appearance information specifically includes: comparing the weld coverage of the three-dimensional shape of the weld and the convexity of the weld coverage between the two.
[0046] In one embodiment, within the first threshold range refers to the error ratio of the weld coverage range of the three-dimensional shape of the weld between the two, and the error ratio of the convexity of the weld coverage range of the three-dimensional shape of the weld between the two is within ±10%. Both conditions need to be met at the same time to determine that the boom box welding is qualified, and when the machine vision determines that it is unqualified, it is necessary to perform supplementary welding again. According to relevant standards, such as ISO, AWS (American Welding Society Standard), GB (Chinese National Standard) and other standards, as well as relevant specifications or design requirements, the weld size error is usually controlled between ±10% and ±15%. In general, each supplementary welding is based on the close to qualified boom box weld, rather than directly full welding or exceeding the three-dimensional shape range of the weld, which also makes it possible that there is still room for supplementation after each welding and machine vision judgment, or directly reaches the qualified welding standard after welding and machine vision judgment, rather than directly exceeding the first threshold range. Obviously, the probability of the last case is extremely small, which is also the key to ensuring that the yield rate of the boom box formed by one-time welding is controlled at more than 99%.
[0047] In one embodiment, in step S30, the initial boom box is precisely welded by the welding robot in a complementary manner, specifically, the initial welding parameters are increased by 1.04 to 1.07 times in the same proportion, that is, Figure 3 In the process from a to b, a compensation parameter is formed, and the increase part ( Figure 3The gray area in part b) is the part that needs supplementary welding, and the compensating welding parameters are derived therefrom. The initial boom box is precisely welded by the welding robot according to the weld path and the compensating welding parameters. During this period, as mentioned above, due to the problem of welding error, although it is a very small error, in the supplementary welding as precision welding, the compensation parameters of the supplementary welding are flexibly adjusted between 1.04 times and 1.07 times of the increase ratio, rather than a fixed value. That is to say, the amount of supplementation may be larger in some areas and smaller in some areas.
[0048] Based on the initial welding ratio of 90%-95%, the supplementary welding increase ratio is 1.04-1.07. Therefore, the size ratio of the supplemented weld to the qualified boom box weld is between 0.9*1.04=0.936 and 0.95*1.07=1.0165, that is, between -6.4% and 1.6%, which fully meets the standard of the first threshold range (±10%). Therefore, the space for supplementary welding or the supplementary margin is relatively large.
[0049] Normally, when the second weld is formed by supplementary welding, the weld size of most boom boxes can meet the requirements. However, considering that in some extreme cases, the difference between the final product and the qualified product is less than -10% (i.e., the value obtained by subtracting the three-dimensional shape size of the weld of the qualified product from the three-dimensional shape size of the weld of the final product) due to the large welding error, because it is supplementary welding, the supplement amount is between -6.4% and 1.6%, and the supplement range tends to be negative (i.e., it does not reach the level of full welding) rather than positive (i.e., it exceeds the situation of full welding). Therefore, the difference between the two will basically not exceed 10%. In one embodiment, if the difference between the three-dimensional shape size of the weld in the first appearance information and the size of the three-dimensional shape of the weld exceeds the first threshold range, the appearance information of the boom box is recognized for the third time by machine vision, and the three-dimensional shape information of the weld after the first supplementary precision welding is recognized again. Based on the three-dimensional shape of the weld, the initial boom box is secondarily precisely welded by the welding robot in a supplementary manner, thereby further compensating for the welding defects and ensuring that the boom box can be welded and formed in one time.
[0050] Further preferably, the initial boom box is precision welded a second time by the welding robot in a complementary manner, specifically, the primary compensation parameters are proportionally increased by 1.05 times to 1.08 times, the increased portion being the secondary compensation welding parameters, and the initial boom box is continued to be precision welded by the welding robot according to the weld path and the secondary compensation welding parameters.
[0051] The principle of secondary supplementary welding is similar to that of primary supplementary welding, wherein the amplification factor of secondary supplementary welding is 1.05-1.08, wherein secondary supplementary welding is basically local supplementary welding, that is, it mainly supplements the part where the difference between the two is less than -10%, and only a small amount of supplement or no supplement is made to other parts that need to be supplemented (the difference between the two is greater than -10% and less than 0), and the welding requirements can still be met. Therefore, the amplification factor is 1.05-1.08, which is greater than 1.04-1.07 in step S30. If it is less than or equal to the amplification factor in step 30, it may be necessary to perform supplementary welding again due to some special circumstances. Therefore, appropriately increasing this amplification factor can not only reduce the probability of needing to supplement welding again, thereby speeding up the welding forming rate, but also directly meet the welding requirements. 0.9*1.05=94.5%, 0.9*1.08=97.2%, that is, based on -10%, after the second supplement, the ratio of the weld to the weld on the qualified boom box is 94.5%-97.2%, which can meet the requirements of the first threshold range. In the working process, after experimental research and normal processing operation data acquisition, the probability of secondary supplement is 0.3‰, so the probability of secondary supplement welding is extremely low. Based on machine vision, the initial welding and one-time supplement welding can ensure that the one-time forming yield of the boom box is above 99%.
[0052] It is worth mentioning that, if conditions permit, the number of supplementary welding is not limited, and there may be a third supplementary welding, a fourth supplementary welding, etc. These conditions include the first threshold range value (which may be greater than ±10% and less than ±15%) and the supplementary ratio value. However, the final result is to ensure that the boom box can be welded and formed in one go. In addition, the more times of supplementary welding, the smaller the amount of supplementation each time. Although this will sacrifice a certain amount of welding time, it can obviously further improve the yield rate of one-time forming and further improve the weld quality of the boom box welding. The third supplementary welding, the fourth supplementary welding and subsequent supplementary welding (which can be referred to as several supplementary weldings after the second supplementary welding) processes are the same as the second supplementary welding process, that is, the amount of weld to be supplemented is first confirmed by machine vision recognition, and then supplementary welding is performed with an increase of 1.05-1.08.
[0053] In one embodiment, the boom box to be welded is directionally rotated in the horizontal direction at a rotation frequency of 180° by a rotary positioner, and at the same time, the vision robot and the welding robot are symmetrically distributed on both sides of the rotation direction of the boom box to be welded, wherein the vision robot provides machine vision, that is, the machine vision and the welding robot are located on two opposite sides in the horizontal direction, and welding and visual recognition do not interfere with each other, and do not affect the welding efficiency of the boom box.
[0054] The steel plates that make up the boom box are generally thick plates with a thickness of more than 6 mm. The thickness of the boom box steel plates of large excavators may reach 30 mm. In the welding of thick plates, full welding is often not performed directly, which is easy to directly affect the welding quality, residual stress and deformation control. Therefore, it is best to perform multiple passes of welding in an interval welding manner, but the interval time cannot be too short. If it is too short, it will easily lead to excessive interlayer temperature, thereby increasing the risk of grain coarsening and welding defects in the heat affected zone (HAZ). At the same time, the interval time should not be too long. If it is too long, it may reduce the welding efficiency and even cause cold cracks (especially high-strength steel or hardening-prone materials). Therefore, in one embodiment, the interval time between the rapid welding of the boom box to be welded in step S20 and the precision welding of the initial boom box in a complementary manner in step S30 is 2-4 minutes, and the interval time between the precision welding of the initial boom box according to the compensation welding parameters in step S30 and the second precision welding of the initial boom box in a complementary manner in step S40 is 2-4 minutes.
[0055] Combination Figure 1 , Figure 4 and Figure 5 In a second aspect of an embodiment of the present application, a boom box welding system based on machine vision is provided, which is used to implement the boom box welding method based on machine vision as described above, wherein the welding system includes a control cabinet 20, a rotary positioner 30, a welding robot 40 and a visual robot (not shown in the figure), wherein the rotary positioner 30 has a side output terminal 31, and at the same time, the rotary positioner 30 drives the side output terminal 31 to rotate in a horizontal direction, and the circumferential rotation amount is basically controlled to be 180° each time. In addition, the side output terminal 31 is connected to a fixing tool 32 for fixing the boom box 10, wherein the boom box 10 are symmetrically distributed on both sides of the fixed tooling 32, and the welding robot 40 and the visual robot are respectively facing the two sides of the fixed tooling 32, so that when the welding robot 40 is welding the boom box 10 on one side, the visual robot can visually scan and identify the boom box 10 on the other side. In this way, welding and visual recognition will not interfere with each other. At the same time, each rotation of the rotary positioner 30 by 180° can ensure that the working ranges of the welding robot 40 and the visual robot are consistent, and the information recognized by the visual robot only needs to be rotated 180° to fully correspond to the welding robot 40, and the control is convenient and accurate.
[0056] Further preferably, the fixing tool 32 includes a main bracket 321 and an inner supporting mechanism and an outer pressing mechanism symmetrically distributed on both sides of the main bracket 321;
[0057] The internal support mechanism includes a plurality of telescopic cylinders 322 distributed at intervals along the outer side of a predetermined area, wherein the predetermined area is used to place the boom box 10, and the output end of the telescopic cylinder 322 is connected to a support block 323, so that the plurality of telescopic cylinders 322 can correspondingly drive the support block 323 to abut against the inner support rib 12 between the two parallel steel plates 11 of the boom box 10 above and below the boom box 10, respectively, to position, support and fix the boom box 10, and the operation is convenient.
[0058] The external clamping mechanism includes a plurality of clamping cylinders 324 spaced apart along the outer edge of the predetermined area, wherein the output end of the clamping cylinder 324 is vertically connected to a clamping frame 325, and the front end of the clamping frame 325 is connected to a positioning piece 326 that cooperates with the axial hole 101 of the boom box 10, so that the position of the boom box 10 can be accurately positioned to ensure the welding accuracy of the boom box 10, wherein the positioning piece 326 can be a positioning pin or a positioning shaft.
[0059] Therefore, the positioning, support and fixation of the boom box 10 formed by the internal support mechanism, combined with the precise positioning of the boom box 10 formed by the external clamping mechanism, can firmly fix the boom box 10 in place, while being easy to operate and maintain.
[0060] The operation process is roughly as follows: first, move the lower support block 323 into position, then support the boom box 10 on the lower support block 323, and then the clamping cylinder 3231 drives the positioning piece 326 to be inserted into the corresponding axial hole 101 to accurately position the boom box 10, and finally move the upper support block 323 so that the upper support block 323 cooperates with the lower support block 323 to firmly clamp the boom box 10, wherein the support block 323 abuts against the inner support rib 12 to clamp and fix the boom box 10, and can also abut against the upper and lower edges of the boom box 10 (i.e., the edges of the parallel steel plate 11) and the inner support rib 12 at the same time to clamp and fix the boom box 10.
[0061] It should be noted that the terms "first, second" in the present application are only used for descriptive purposes and do not indicate any order. They cannot be understood as indicating or implying relative importance. These terms can be interpreted as names.
[0062] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The advantages of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.
Claims
1. A boom box welding method based on machine vision, characterized in that: The following steps are included in sequence: S10, establishing a standard welding appearance database based on a boom box having qualified welding appearance information, wherein the qualified welding appearance information includes a weld path and a weld three-dimensional shape; S20, based on the standard welding appearance database, the three-dimensional shape of the weld is reduced by 90%-95% in the same proportion to obtain initial welding parameters, and the welding robot quickly welds the boom box to be welded according to the weld path and the initial welding parameters to obtain an initial boom box; S30, identifying appearance information of an initial boom box by machine vision, and based on the three-dimensional shape of the weld, precisely welding the initial boom box by the welding robot in a complementary manner; S40, again identifying the appearance information of the initial boom box through machine vision to form primary appearance data information, comparing the primary appearance data information with the qualified welding appearance information, if the difference between the three-dimensional shape size of the weld in the primary appearance information and the size of the three-dimensional shape of the weld is within the first threshold range, it is determined that welding is completed.
2. The method for welding a boom box based on machine vision according to claim 1, characterized in that: The comparing of the primary appearance data information and the qualified welding appearance information specifically includes: comparing the weld coverage of the three-dimensional shape of the weld and the convexity of the weld coverage between the two.
3. The machine vision-based boom box welding method according to claim 2, characterized in that: Within the first threshold range means that the error ratio of the weld coverage of the three-dimensional shape of the weld between the two, and the error ratio of the convexity of the weld coverage of the three-dimensional shape of the weld between the two are both within ±10%.
4. The machine vision-based boom box welding method according to claim 3, characterized in that: In step S30, the initial boom box is precisely welded by the welding robot in a complementary manner, specifically, the initial welding parameters are proportionally increased by 1.04 to 1.07 times to form a compensation parameter, and the increased portion is the compensation welding parameter. The initial boom box is precisely welded by the welding robot according to the weld path and the compensation welding parameter.
5. The method for welding a boom box based on machine vision according to claim 4, characterized in that: In step S40, if the difference between the size of the weld three-dimensional shape in the first appearance information and the size of the weld three-dimensional shape exceeds the first threshold range, the appearance information of the boom box is identified by machine vision for the third time, and based on the weld three-dimensional shape, the initial boom box is precision-welded a second time by the welding robot in a complementary manner.
6. The method for welding a boom box based on machine vision according to claim 5, characterized in that: The welding robot performs a second precision welding of the initial boom box in a complementary manner, specifically, the first compensation parameter is increased by 1.05 to 1.08 times in the same proportion, and the increased portion is the secondary compensation welding parameter. The welding robot continues to perform precision welding of the initial boom box according to the weld path and the secondary compensation welding parameter.
7. The machine vision-based boom box welding method according to claim 6, characterized in that: The boom box to be welded is directionally rotated in the horizontal direction at a rotation frequency of 180° by a rotary positioner, and the vision robot and the welding robot are symmetrically distributed on both sides of the rotation direction of the boom box to be welded, wherein the vision robot provides machine vision.
8. The boom box welding method based on machine vision according to claim 6 or 7, characterized in that: The interval time between quickly welding the boom box to be welded in step S20 and precision welding the initial boom box in a complementary manner in step S30 is 2-4 minutes, and the interval time between precision welding the initial boom box according to the compensation welding parameters in step S30 and precision welding the initial boom box for the second time in a complementary manner in step S40 is 2-4 minutes.
9. A boom box welding system based on machine vision, used to implement the boom box welding method based on machine vision as described in any one of claims 1 to 8, characterized in that: The welding system includes a control cabinet, a rotary positioner, a welding robot and a visual robot. The rotary positioner has a side output end. The rotary positioner drives the side output end to rotate in a horizontal direction. The side output end is connected to a fixed tooling for fixing a boom box, wherein the boom boxes are symmetrically distributed on both sides of the fixed tooling. The welding robot and the visual robot are respectively facing the two sides of the fixed tooling, so that when the welding robot is welding the boom box on one side, the visual robot can visually scan and identify the boom box on the other side.
10. The machine vision-based boom box welding system according to claim 9, characterized in that: The fixing fixture comprises a main support and an inner supporting mechanism and an outer pressing mechanism symmetrically distributed on both sides of the main support; The inner support mechanism includes a plurality of telescopic cylinders distributed at intervals along the outer side of a predetermined area, wherein the predetermined area is used to place a boom box, and the output ends of the telescopic cylinders are connected to support blocks, so that the plurality of telescopic cylinders can correspondingly drive the support blocks to abut against the inner support ribs between the two parallel steel plates of the boom box above and below the boom box, respectively, to support and fix the boom box; The external clamping mechanism includes a plurality of clamping cylinders spaced apart along the outer edge of the predetermined area, the output end of the clamping cylinder is vertically connected to a clamping frame, and the front end of the clamping frame is connected to a positioning piece that matches the axial hole of the boom box.
Citation Information
Patent Citations
Excavator bucket rod welding mechanism and method thereof
CN118848408A
Bucket rod machining and welding device
CN216882475U
Welding method based on digital modeling and flexible weld joint recognition
CN116275742A
Parameter debugging method and system for laser hybrid welding process
CN116441730A
Control method and system for self-adaptive positioning equipment of welding seam
CN116727956A