Intelligent welding system and method based on visual recognition

Abstract

The application discloses an intelligent welding system and method based on visual identification, and relates to the technical field of laser welding.The system comprises a position adjusting assembly, which has an adjusting end with at least three degrees of freedom; a laser welding head installed to the adjusting end; a rotating seat arranged outside the laser welding head and rotating around the Y direction; an arm body, one end of which is connected to the rotating seat, and the other end of which is connected with a first visual identifier, which is used to obtain side welding information of first and second workpieces, and to obtain side welding depth information according to the side welding information of the first and second workpieces; wherein the laser welding head can adjust its own parameters according to the side welding depth information; wherein a synchronous monitoring assembly is further arranged below the first and second workpieces, which is used to synchronously monitor the welding depth of the first and second workpieces, so as to improve the effect of intelligent welding.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of laser welding, in particular to an intelligent welding system and method based on visual recognition. BACKGROUND

[0002] In the manufacturing industry, laser welding technology has become one of the preferred methods for connecting metal parts due to its high precision, high efficiency and high flexibility. However, in the face of increasingly complex and variable workpiece welding requirements, its inherent limitations gradually emerge, especially in terms of welding depth control and welding efficiency, there are still some deficiencies.

[0003] Firstly, the accuracy of welding depth control is crucial. Too deep welding may cause excessive melting of materials, affecting the strength and stability of the structure; while too shallow welding may leave un-melted defects, reducing the strength of the welded joint. Unfortunately, existing technologies often cannot monitor and control the welding depth in real time and accurately, lacking effective depth monitoring and feedback mechanisms, making it difficult to achieve the expected welding effect in the welding process, thereby affecting the overall performance of the product.

[0004] In terms of welding efficiency, existing technologies often require a lot of manual intervention and adjustment, which not only prolongs the welding cycle and increases production costs, but also limits the automation and intelligent level of the production line. With the transformation of manufacturing industry to intelligent manufacturing, higher requirements are put forward for the automation and intelligent control of the welding process, and existing technologies are obviously not up to the task in this regard.

[0005] Therefore, it is necessary to provide an intelligent welding system and method based on visual recognition to solve the above problems. SUMMARY

[0006] To solve the above problems, the present application provides the following technical solution: an intelligent welding system based on visual recognition for welding a first workpiece and a second workpiece, comprising:

[0007] A position adjustment assembly having an adjustment end, the adjustment end having at least three degrees of freedom;

[0008] A laser welding head mounted to the adjustment end;

[0009] A rotating seat arranged outside the laser welding head around the Y direction;

[0010] An arm body connected to the rotating seat at one end and having a first visual identifier connected at the other end, the first visual identifier being used to obtain side welding information of the first workpiece and the second workpiece, and to obtain side welding depth information according to the side welding information of the first workpiece and the second workpiece;

[0011] The laser welding joint can adjust its own parameters according to the side welding depth information.

[0012] The first workpiece and the second workpiece are further provided with a synchronous monitoring assembly below, which is used for synchronously monitoring the welding depth of the first workpiece and the second workpiece.

[0013] Further, as a preferred, the intelligent welding system further comprises a base and two jigs, wherein the jigs are arranged on the base, and the two jigs are respectively a first jig and a second jig, wherein the first jig is used for positioning the first workpiece, and the second jig is used for positioning the second workpiece.

[0014] The first workpiece and the second workpiece maintain a first interval;

[0015] The bottom of the first workpiece and the second workpiece maintains a second interval with the base, so that the synchronous monitoring assembly synchronously monitors the first workpiece and the second workpiece from the bottom of the first workpiece and the second workpiece.

[0016] Further, as a preferred, the synchronous monitoring assembly comprises:

[0017] A follow-up assembly having a moving end moving along the Z direction;

[0018] An offset adjustment assembly fixed to the moving end, and the offset adjustment assembly has an offset adjustment end moving along the X direction;

[0019] An angle adjustment assembly arranged on the offset adjustment end, and the angle adjustment assembly has an angle adjustment end rotating around the Z direction;

[0020] A second visual identifier fixed to the angle adjustment end.

[0021] Further, as a preferred, the follow-up assembly is any one of a screw nut mechanism, an electric telescopic rod, an air cylinder, and a hydraulic cylinder.

[0022] Further, as a preferred, the offset adjustment assembly comprises:

[0023] A base fixed to the moving end;

[0024] An offset seat slidingly arranged on the base as the offset adjustment end;

[0025] Two first micro telescopic rods, both fixed to the base and located on both sides of the offset seat, and the output ends of the two first micro telescopic rods are in contact with the offset seat.

[0026] Further, as a preferred, the two sides of the offset seat and the side of the base close to the offset seat are both fixed with a mounting seat.

[0027] The elastic arc plate is fixed between the two mounting seats on the same side of the offset seat.

[0028] Further, preferably, an elastic rope is connected between the two mounting seats on the same side of the offset seat.

[0029] Further, preferably, the angle adjustment assembly comprises:

[0030] a roller body rotatably arranged on the offset seat and driven by a motor, and a support frame as an angle adjustment end is fixed above the roller body;

[0031] two locking arc plates arranged symmetrically on both sides of the roller body, one end of the locking arc plate is hinged to the offset seat, and an elastic pad is attached to the side of the locking arc plate close to the roller body;

[0032] a synchronous rod hinged to the two locking arc plates and driven by a second micro-motion telescopic rod, and the second micro-motion telescopic rod is fixed on the offset seat.

[0033] An intelligent welding method based on visual recognition, comprising the following steps:

[0034] S1. Obtain first workpiece information and second workpiece information;

[0035] S2. Obtain first spacing information and first welding depth information according to the first workpiece information and the second workpiece information;

[0036] S3. Position the first workpiece using a first jig and position the second workpiece using a second jig, and maintain a first spacing between the first workpiece and the second workpiece;

[0037] S4. The position adjustment assembly adjusts the laser welding head and welds the first workpiece and the second workpiece;

[0038] S5. The first visual recognizer moves to the side of the first workpiece and the second workpiece, obtains side welding depth information, and calculates the difference between the side welding depth information and the first welding depth information, and adjusts the parameters of the laser welding head according to the difference;

[0039] S6. Reset the first visual recognizer;

[0040] S7. Obtain first angle information and first position information according to the first spacing information and the first welding depth information;

[0041] S8. Adjust the second visual recognizer according to the first angle information and the first position information;

[0042] S9. Continue to weld the first workpiece and the second workpiece using the laser welding head, and the second visual recognizer synchronously monitors.

[0043] Compared with the prior art, the present application provides an intelligent welding system and method based on visual recognition, which has the following beneficial effects:

[0044] In the present application, the laser welding head automatically adjusts its welding parameters, such as laser power, welding speed, focal point position, etc., according to the side welding depth information provided by the first visual identifier, to ensure that the preset welding depth is reached.

[0045] In the present application, the synchronous monitoring assembly monitors the welding depth in real time to ensure the stability and accuracy of the welding process. If the monitored welding depth does not match the expected value, the system can correct it in time through the parameter adjustment of the position adjustment assembly and the laser welding head.

[0046] In the present application, through the visual recognition technology, the system can accurately obtain the welding information of the workpiece, realize high-precision welding operation, and realize high intelligence and automation in the entire welding process, reducing manual operation and improving production efficiency and safety. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is a structural schematic diagram of an intelligent welding system based on visual recognition;

[0048] Figure 2 is a structural schematic diagram of a follow-up assembly in an intelligent welding system based on visual recognition;

[0049] Figure 3 is a structural schematic diagram of a synchronous monitoring assembly in an intelligent welding system based on visual recognition;

[0050] Figure 4 is Figure 3 is an enlarged structural schematic diagram of position A in the figure;

[0051] Figure 5 is a flowchart of an intelligent welding method based on visual recognition;

[0052] In the figure: 1, position adjustment assembly; 2, laser welding head; 3, transfer seat; 4, arm body; 5, first visual identifier; 6, base; 7, jig; 8, follow-up assembly; 9, offset adjustment assembly; 10, angle adjustment assembly; 11, second visual identifier; 91, base; 92, offset seat; 93, mounting seat; 94, elastic arc plate; 96, elastic rope; 97, first micro-motion telescopic rod; 101, locking arc plate; 102, elastic pad; 103, roller body; 104, synchronous rod; 105, second micro-motion telescopic rod; 106, support frame. DETAILED DESCRIPTION

[0053] The terms "first", "second", and the like, as used in the description and the claims of the present application and the above Abstract, are used for distinguishing between similar objects and not necessarily for describing a specific sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the present application described herein are, for example, capable of orderly or chronological sequential processes or procedures, or capable of stand alone or non-sequential processes or procedures, or capable of both. Furthermore, the terms "comprise", "comprising", "include", "including", and the like, as used herein, are specifically intended to be construed as open-ended terms i.e., the terms do not exclude additional, unrecited elements or steps. It is to be understood that the terms so used are also meant to be inclusive.

[0054] Reference will now be made to Figures 1-5 In the embodiments of the present application, a visual recognition based intelligent welding system is provided for welding a first workpiece and a second workpiece, comprising:

[0055] a position adjustment assembly 1 having an adjustment end with at least three degrees of freedom;

[0056] a laser welding head 2 mounted to the adjustment end;

[0057] a rotating seat 3 arranged outside the laser welding head 2 and rotating around the Y direction;

[0058] an arm body 4 connected to the rotating seat 3 at one end and having a first visual recognizer 5 connected at the other end, the first visual recognizer 5 being used to obtain side welding information of the first workpiece and the second workpiece and to obtain side welding depth information according to the side welding information of the first workpiece and the second workpiece;

[0059] wherein the laser welding head 2 is capable of adjusting its own parameters according to the side welding depth information;

[0060] wherein a synchronous monitoring assembly is further arranged below the first workpiece and the second workpiece for synchronously monitoring the welding depth of the first workpiece and the second workpiece.

[0061] In implementation, the following steps are included:

[0062] S1. Placing the first workpiece and the second workpiece on a welding workbench according to predetermined positions and manners.

[0063] S2. Initially, the position adjustment assembly 1 (having at least three degrees of freedom, such as movement in X, Y, Z axis directions and possible rotation) adjusts the initial position of the laser welding head 2 so as to align the welding starting point of the workpieces (the first workpiece and the second workpiece).

[0064] S3. The first visual identifier 5 obtains the side welding information of the first workpiece and the second workpiece from the side of the first workpiece and the second workpiece through the flexible rotation of the arm body 4 and the rotating seat 3. According to the side welding information, the first visual identifier 5 calculates the side welding depth information.

[0065] S4. The laser welding head 2 automatically adjusts its welding parameters, such as laser power, welding speed, focal point position, etc., according to the side welding depth information provided by the first visual identifier 5, to ensure that the preset welding depth is reached.

[0066] S5. The synchronous monitoring assembly (located below the first workpiece and the second workpiece) monitors the welding depth in real time to ensure the stability and accuracy of the welding process. If the monitoring shows that the welding depth does not meet the expectation, the system can timely correct it through the position adjustment assembly 1 and the parameter adjustment of the laser welding head 2.

[0067] In this embodiment, through the visual recognition technology, the system can accurately obtain the welding information of the workpiece, realizing high-precision welding operation. The automatic parameter adjustment function of the laser welding head 2 makes the welding process more efficient, reducing manual intervention and debugging time.

[0068] The whole welding process realizes high intelligence and automation, reduces manual operation, and improves production efficiency and safety. The system can dynamically adjust according to real-time welding depth information, making the welding quality more reliable and stable.

[0069] In this embodiment, the intelligent welding system further comprises a base 6 and two jigs 7, wherein the jigs 7 are arranged on the base 6, and the two jigs 7 are respectively a first jig and a second jig, wherein the first jig is used for positioning the first workpiece, and the second jig is used for positioning the second workpiece;

[0070] The first workpiece and the second workpiece maintain a first distance;

[0071] The bottom of the first workpiece and the second workpiece maintains a second distance with the base 6, so that the synchronous monitoring assembly can synchronously monitor the first workpiece and the second workpiece from the bottom of the first workpiece and the second workpiece.

[0072] It needs to be explained that the first jig and the second jig are mainly used to maintain the first distance between the first workpiece and the second workpiece. The first distance is actually the gap of the weld. The size of the weld gap usually needs to be determined according to the welding process, the type and thickness of the material, and the type of the welded joint. Too small gap may cause poor flow of the molten pool during welding, resulting in poor weld formation; while too large gap may make it difficult for the molten pool to fill during welding, resulting in internal defects in the weld, such as incomplete fusion and incomplete penetration, etc.

[0073] During the welding process, it is very important to maintain a proper weld gap. This not only ensures the stability of the welding process, but also improves the quality and reliability of the weld.

[0074] In this embodiment, the synchronous monitoring assembly comprises:

[0075] A following assembly 8, which has a moving end moving along the Z direction;

[0076] An offset adjustment assembly 9, which is fixed to the moving end, and the offset adjustment assembly 9 has an offset adjustment end moving along the X direction;

[0077] An angle adjustment assembly 10, which is arranged at the offset adjustment end, and the angle adjustment assembly 10 has an angle adjustment end rotating around the Z direction;

[0078] A second visual identifier 11, which is fixed to the angle adjustment end.

[0079] Then, during the welding process, the second visual identifier 11 continuously monitors the welding depth of the first workpiece and the second workpiece from the bottom. After the welding is completed, the synchronous monitoring assembly can also provide detailed welding depth monitoring data for the system to analyze and evaluate. These data can be used to optimize welding parameters, improve process flow, or as a basis for quality control.

[0080] In this embodiment, through the precise adjustment of the following assembly 8, the offset adjustment assembly 9, and the angle adjustment assembly 10, the second visual identifier 11 realizes high-precision monitoring of the welding depth. The second visual identifier 11 observes the weld from the bottom, avoiding monitoring errors caused by uneven or blocked workpiece surfaces.

[0081] In this embodiment, the following assembly 8 is any one of a lead screw nut mechanism, an electric telescopic rod, a pneumatic cylinder, or a hydraulic cylinder.

[0082] In this embodiment, the offset adjustment assembly 9 comprises:

[0083] A base 91, which is fixed to the moving end;

[0084] An offset seat 92, which is arranged on the base 91 as a sliding offset adjustment end;

[0085] Two first micro telescopic rods 97, both of which are fixed to the base 91 and located on both sides of the offset seat 92, and the output ends of the two first micro telescopic rods 97 are in contact with the offset seat 92.

[0086] In this embodiment, the two sides of the offset seat 92 and the side of the base 91 close to the offset seat 92 are both fixed with a mounting seat 93;

[0087] The elastic arc plate 94 is fixed between two mounting seats 93 on the same side of the offset seat 92.

[0088] The elastic rope 96 is connected between two mounting seats 93 on the same side of the offset seat 92.

[0089] In this embodiment, the offset seat 92 is used as an offset adjustment end and is slidingly arranged on the base 91 and is pushed and limited by two first micro-motion telescopic rods 97. According to the welding requirement, the system sends an instruction to the first micro-motion telescopic rod 97, so that the output end of the first micro-motion telescopic rod 97 pushes the offset seat 92, thereby realizing the fine adjustment in the X direction. The elastic arc plate 94 and the elastic rope 96 play a buffering and stabilizing role when the offset seat 92 moves, thereby preventing impact and vibration caused by too fast or too large movement.

[0090] After the welding is completed, the offset adjustment assembly 9 returns to the initial position according to the system instruction, thereby preparing for the next welding. In the resetting process, the elastic arc plate 94 and the elastic rope 96 again play a role in ensuring that the offset seat 92 can stably return to the specified position.

[0091] In this embodiment, the angle adjustment assembly 10 comprises:

[0092] The roller body 103 is rotationally arranged on the offset seat 92 and is driven by a motor (not shown in the figure), and a support frame 106 as an angle adjustment end is fixed above the roller body 103;

[0093] Two locking arc plates 101 are symmetrically arranged on both sides of the roller body 103, one end of the locking arc plate 101 is hinged to the offset seat 92, and an elastic pad 102 is attached to one side of the locking arc plate 101 close to the roller body 103;

[0094] The synchronous rod 104 is hinged to the two locking arc plates 101 and is driven by a second micro-motion telescopic rod 105, and the second micro-motion telescopic rod 105 is fixed on the offset seat 92.

[0095] In the implementation, according to the welding requirement, the system sends an instruction to the second micro-motion telescopic rod 105, so that the output end of the second micro-motion telescopic rod 105 pushes or pulls the synchronous rod 104. The synchronous rod 104 is hinged to the two locking arc plates 101, so when the synchronous rod 104 moves, it will drive the locking arc plates 101 to rotate around the hinge point, thereby changing the included angle between the locking arc plates 101 and the roller body 103.

[0096] Since the locking arc plates 101 are attached with the elastic pads 102, they can provide stable support force when they contact the roller body 103, and prevent damage caused by friction. With the rotation of the locking arc plates 101, the roller body 103 will be subjected to a certain extrusion force, thereby being limited.

[0097] In the welding process, the angle adjusting assembly 10 cooperates with the second visual identifier 11 to monitor the position and depth of the weld seam in real time. During the welding process, as the weld seam moves and changes, the angle adjusting assembly 10 can adjust the rotation angle of the roller body 103 in real time to ensure that the second visual identifier 11 can always clearly observe the weld seam.

[0098] Based on this, the embodiment also provides an intelligent welding method based on visual identification, which comprises the following steps:

[0099] S1. Obtain first workpiece information and second workpiece information. Specifically, the images or three-dimensional data of the first workpiece and the second workpiece can be captured by using a high-precision scanner or a camera, the images or data are preprocessed, and key features such as size, shape, edge, etc. are extracted. Alternatively, the first workpiece information and the second workpiece information can be directly obtained from a database;

[0100] S2. Obtain first spacing information and first welding depth information according to the first workpiece information and the second workpiece information. Specifically, after obtaining the first workpiece information and the second workpiece information, the first spacing information and the first welding depth information are determined according to a data reference table. The first spacing information is actually preset spacing information, and the first welding depth information is actually preset welding depth information;

[0101] S3. Position the first workpiece by using a first jig and position the second workpiece by using a second jig, and maintain a first spacing between the first workpiece and the second workpiece. Specifically, the workpieces are placed on the jigs (the first jig and the second jig), and the positions and angles of the jigs are adjusted to ensure that the first spacing is maintained between the two workpieces to meet the requirements of the preset spacing information;

[0102] S4. The position adjusting assembly 1 adjusts the laser welding head 2 and welds the first workpiece and the second workpiece;

[0103] S5. The first visual identifier 5 moves to the side of the first workpiece and the second workpiece to obtain side welding depth information, and the side welding depth information and the first welding depth information are difference calculated, and the parameters of the laser welding head 2 are adjusted according to the difference. Specifically, during the welding process, the first visual identifier 5 moves to the side of the workpieces to capture images of the welding areas of the sides of the first workpiece and the second workpiece in real time. The side welding depth information is calculated by using an image processing algorithm. The side welding depth information is compared with the preset first welding depth information, and the difference is calculated. The parameters of the laser welding head 2, such as power and speed, are adjusted according to the difference to ensure the welding quality;

[0104] S6. Reset the first visual identifier 5;

[0105] S7. Obtain first angle information and first position information according to the first distance information and the first welding depth information, specifically, the optimal angle and position required for the second visual identifier 11 to perform visual identification are calculated through an algorithm using the first distance information and the first welding depth information, i.e. the first angle information and the first position information;

[0106] S8. Adjust the second visual identifier 11 according to the first angle information and the first position information;

[0107] S9. Continue to use the laser welding head 2 to weld the first workpiece and the second workpiece, and the second visual identifier 11 performs synchronous monitoring.

[0108] The entire welding process is highly automated, reducing manual intervention and errors, and the system can automatically adjust the welding parameters and the position and angle of the monitoring components according to the monitoring data, improving production efficiency and welding quality.

[0109] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can make equivalent replacements or changes within the technical range disclosed by the present application according to the technical solution and the inventive concept of the present application, which should be covered within the protection scope of the present application.

Claims

1. An intelligent welding system based on visual recognition for welding a first workpiece and a second workpiece, characterized in that, The utility model relates to a laser welding system, comprising: a position adjusting assembly (1) having an adjusting end with at least three degrees of freedom; a laser welding head (2) mounted to the adjusting end; a rotating seat (3) arranged outside the laser welding head (2) and rotating around the Y direction; an arm body (4) having one end connected to the rotating seat (3) and the other end connected to a first visual identifier (5) for obtaining side welding information of first and second workpieces and obtaining side welding depth information according to the side welding information of the first and second workpieces; wherein the laser welding head (2) can adjust its parameters according to the side welding depth information; wherein a synchronous monitoring assembly is further arranged below the first and second workpieces for synchronously monitoring the welding depth of the first and second workpieces; the synchronous monitoring assembly comprising: a follow-up assembly (8) having a moving end moving along the Z direction; an offset adjusting assembly (9) fixed to the moving end and having an offset adjusting end moving along the X direction; an angle adjusting assembly (10) arranged on the offset adjusting end and having an angle adjusting end rotating around the Z direction; a second visual identifier (11) fixed to the angle adjusting end.

2. The intelligent welding system based on visual recognition according to claim 1, characterized in that, The intelligent welding system further comprises a base (6) and two jigs (7), wherein the jigs (7) are arranged on the base (6), and the two jigs (7) are respectively a first jig and a second jig, wherein the first jig is used for positioning the first workpiece, and the second jig is used for positioning the second workpiece; the first and second workpieces maintain a first interval; the bottom of each of the first and second workpieces maintains a second interval with the base (6) so that the synchronous monitoring assembly synchronously monitors the first and second workpieces from the bottom of each of the first and second workpieces.

3. The intelligent welding system based on visual recognition as claimed in claim 1, wherein, The follow-up assembly (8) is any one of a screw-nut mechanism, an electric telescopic rod, an air cylinder, or a hydraulic cylinder.

4. The intelligent welding system based on visual recognition according to claim 1, wherein, The offset adjusting assembly (9) comprises: a base (91) fixed to the moving end; an offset seat (92) slidingly arranged on the base (91) as the offset adjusting end; two first micro telescopic rods (97) both fixed to the base (91) and located on both sides of the offset seat (92), and the output ends of the two first micro telescopic rods (97) are both in contact with the offset seat (92).

5. The intelligent welding system based on visual recognition as claimed in claim 4, wherein, Both sides of the offset seat (92) and the side of the base (91) close to the offset seat (92) are both fixed with mounting seats (93). The two mounting seats (93) on the same side of the offset seat (92) are both fixed with elastic arc plates (94).

6. The intelligent welding system based on visual recognition according to claim 5, wherein, The two mounting seats (93) on the same side of the offset seat (92) are connected with elastic ropes (96).

7. The intelligent welding system based on visual recognition as claimed in claim 4, wherein, The angle adjusting assembly (10) comprises: a roller body (103) rotatingly arranged on the offset seat (92) and driven by a motor, and a support frame (106) as the angle adjusting end is fixed above the roller body (103); Two locking arc plates (101) are symmetrically arranged on both sides of the roller body (103), one end of the locking arc plate (101) is hinged to the offset seat (92), and the side of the locking arc plate (101) close to the roller body (103) is attached with an elastic pad (102); A synchronous rod (104) is hinged to the two locking arc plates (101) and is driven by a second micro-motion telescopic rod (105), and the second micro-motion telescopic rod (105) is fixed on the offset seat (92).

8. A visual recognition based intelligent welding method using the visual recognition based intelligent welding system according to any one of claims 1 to 7, characterized by, The method comprises the following steps: S1. obtaining first workpiece information and second workpiece information; S2. obtaining first spacing information and first welding depth information according to the first workpiece information and the second workpiece information; S3. positioning the first workpiece by using a first jig and positioning the second workpiece by using a second jig, and maintaining a first spacing between the first workpiece and the second workpiece; S4. adjusting the laser welding head (2) by the position adjustment assembly (1) and welding the first workpiece and the second workpiece; S5. moving the first visual recognizer (5) to the side of the first workpiece and the second workpiece, obtaining side welding depth information, calculating the difference between the side welding depth information and the first welding depth information, and adjusting the parameters of the laser welding head (2) according to the difference; S6. resetting the first visual recognizer (5); S7. obtaining first angle information and first position information according to the first spacing information and the first welding depth information; S8. adjusting the second visual recognizer (11) according to the first angle information and the first position information; S9. continuing to weld the first workpiece and the second workpiece by using the laser welding head (2), and synchronously monitoring by using the second visual recognizer (11).

Images