Internal welding control system

Abstract

The invention relates to an internal welding control system which is used for welding the interior of a workpiece and comprises a moving module used for moving in the workpiece in the horizontal direction; the sweeping module is used for sweeping the moving path of the moving module, the end, provided with the sweeping module, of the moving module serves as the moving front end, and the other end of the moving module serves as the moving rear end. The guide module is used for guiding the movement of the moving module in the workpiece; the welding module is used for welding the interior of the workpiece; the wire feeding module is used for conveying welding wires to the welding module; the moving module, the cleaning module, the guiding module, the welding module and the wire feeding module are all in signal connection with the control module. The cleaning module is arranged at the moving front end of the moving module, so that in the moving process of the moving module in the workpiece, the cleaning module can remove impurities such as particles, stains and dust on the path in advance, the situation that the impurities affect stable movement of the moving module is avoided, and then the moving module can move stably.

Description

Technical Field

[0001] The present application relates to the field of welding technology, and particularly to an internal welding control system. Background Art

[0002] Internal welding is a type of practical problem often encountered in welding production, especially in the welding processes of large bridges, pressure vessels, oil and gas pipelines, etc. Since it is difficult for construction workers to enter the interior of the structure, internal welding is more often completed by using special internal welding mobile devices. Currently, most of these welding platforms use magnetic wheels to firmly adhere to the surface of the workpiece during the welding process.

[0003] However, during the welding process, a large amount of spattered metal is inevitably generated. These liquid metals spattered from the molten pool rapidly cool when encountering air to form irregular metal particles, and the spattered metal particles will adhere to the surface of the wheels, thus seriously hindering the stable rolling of the wheels and even causing the wheels to completely separate from the surface of the workpiece, making it difficult to maintain the accuracy of internal welding. Summary of the Invention

[0004] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present application is to provide an internal welding control system. By arranging the cleaning module at the moving front end of the moving module, during the movement of the moving module inside the workpiece, the cleaning module can pre-clean impurities such as particles, stains, and dust on the path, avoiding the influence of impurities on the smooth movement of the moving module. Furthermore, the moving module can move smoothly along the predetermined route and ensure the continuous and stable subsequent welding.

[0005] An internal welding control system described in the present application includes:

[0006] A moving module, which is used for horizontal movement inside the workpiece;

[0007] A cleaning module, which is arranged at one end of the moving module and is used for cleaning the moving path of the moving module. The moving module takes the end with the cleaning module as the moving front end and the other end as the moving rear end;

[0008] A guiding module, which is arranged on the moving module and is used for guiding the movement of the moving module inside the workpiece;

[0009] A welding module, which is arranged at the moving rear end of the moving module and is used for welding and processing the interior of the workpiece;

[0010] A wire feeding module, which is arranged on the moving module, and the wire feeding module is connected to the welding module for feeding welding wire to the welding module;

[0011] A control module, and the moving module, the cleaning module, the guiding module, the welding module and the wire feeding module are all signal-connected to the control module.

[0012] Preferably, the guiding module includes two relatively arranged moving guiding mechanisms, and the two moving guiding mechanisms are respectively arranged near both sides of the moving module, so that the two moving guiding mechanisms are respectively in contact with the two side walls inside the workpiece inside the workpiece.

[0013] Preferably, the control module controls the moving module, the cleaning module, the guiding module, the welding module and the wire feeding module respectively according to a welding control strategy to complete welding processing;

[0014] The welding control strategy includes the following steps:

[0015] S1. Select welding process parameters according to the parameter information of the workpiece to be welded and set them;

[0016] S2. Set an image acquisition component on the moving module, and collect image data of the internal path of the workpiece in real time through the image acquisition component, and select the moving speed of the moving module according to the impurity density on the internal path of the workpiece;

[0017] S3. Guide the movement of the moving module according to the guiding module;

[0018] S4. Build a weld seam recognition model, move the moving module to the required welding position, and dynamically adjust the welding current and welding speed according to the weld seam type.

[0019] Preferably, step S1 specifically includes:

[0020] The parameter information includes the workpiece thickness and the workpiece material type coefficient;

[0021] The welding process parameters include the initial welding current and the initial welding speed;

[0022] Obtain the workpiece thickness D and the workpiece material type coefficient K of the workpiece ω , and calculate the initial welding current I of the workpiece 0 and the initial welding speed V 0 :

[0023] I 0 = I L * K ω * D * α

[0024]

[0025] Among them, I L represents the theoretical welding current value; α represents the current adjustment coefficient; V L represents the theoretical welding speed value; K V represents the comprehensive influence coefficient of speed; β represents the speed adjustment coefficient.

[0026] Preferably, the step S1 further includes:

[0027] The workpiece material type coefficient K ω is selected according to the workpiece material type:

[0028] If the workpiece material type is carbon steel, the workpiece material type coefficient K ω takes a value of 1.0 to 1.3;

[0029] If the workpiece material type is stainless steel, the workpiece material type coefficient K ω takes a value of 1.4 to 1.7;

[0030] If the workpiece material type is alloy material, the workpiece material type coefficient K ω takes a value of 1.8 to 2.0.

[0031] Preferably, the step S2 specifically includes:

[0032] The moving module makes a forward movement towards the inside of the workpiece with the moving front end, and the image acquisition component performs real-time image acquisition on the path of the forward movement of the moving module to obtain a path image;

[0033] Transmit the path image to the edge computing gateway, and perform graying processing on the path image data by the weighted average method to obtain a grayscale image;

[0034] Adopt the median filtering algorithm to perform denoising processing on the grayscale image to obtain a first path image;

[0035] Adopt the histogram equalization technique to perform image enhancement processing on the first path image to obtain a second path image;

[0036] Perform image segmentation processing and morphological processing on the second path image in sequence to obtain a third path image, count the number of pixel points in the impurity part of the third path image, and calculate the impurity area S by using an image analysis algorithm. Calculate the impurity density M according to the following formula:

[0037]

[0038] Among them, the unit of the number N of pixel points is piece; the unit of the area S of the impurity region is square pixel; the unit of the impurity density M is piece / square pixel.

[0039] Preferably, step S2 further includes:

[0040] The moving speed of the moving module selects the moving speed V according to the impurity density M Y ;

[0041] Select the moving speed adjustment coefficient C according to the pre-established mapping relationship between the impurity density and the moving speed range, and dynamically adjust the moving speed V according to the moving speed adjustment coefficient C Y ;

[0042] If 0 ≤ M < 2, the value of the moving speed adjustment coefficient C is:

[0043]

[0044] If 2 ≤ M < 10, the value of the moving speed adjustment coefficient C is:

[0045]

[0046] If M ≥ 10, the value of the moving speed adjustment coefficient C is:

[0047]

[0048] The moving speed V Y is calculated according to the following formula:

[0049] V Y = V YL * C

[0050] where V YL represents the theoretical initial moving speed.

[0051] Preferably, step S3 specifically includes:

[0052] Pressure sensors are arranged at the positions where the two moving guiding mechanisms are in contact with the inner side wall of the workpiece, and the pressure values of the two moving guiding mechanisms are monitored in real time to obtain the real-time guiding pressure values F DX1 and F DX2 ;

[0053] If F DX1 = [F Dmin , F Dmax and F DX2 = [F Dmin , F Dmax , it is determined that the movement trajectory of the moving module is normal, and the original movement state is maintained;

[0054] If F DX1 ≠ [F Dmin , F Dmax and / or F DX2 ≠ [F Dmin , F Dmax , it is determined that the movement trajectory of the moving module is abnormal, an alarm message is sent, and a movement adjustment operation is performed;

[0055] Among them, F Dmin represents the minimum value of the preset guiding pressure; F Dmax represents the maximum value of the preset guiding pressure.

[0056] Preferably, the step S4 specifically includes:

[0057] Obtain historical weld image data and corresponding weld type label data, and the weld type labels of the weld type label data include butt weld, fillet weld, T-joint weld, lap weld, plug weld and slot weld;

[0058] Perform grayscale processing, denoising processing and image enhancement processing on the historical weld image data to obtain the processed first historical weld image data;

[0059] Adopt the Canny algorithm and the gray-level co-occurrence matrix method to extract the edge features, geometric features and texture features of the first historical weld image data to obtain a weld feature vector;

[0060] Adopt a support vector machine, use the weld feature vector as the input and the weld type label as the output, perform model training and testing on the historical weld image data and the weld type label data, and obtain a trained weld recognition model;

[0061] The moving module moves to the required welding position, the image acquisition component acquires the weld image of the required welding position, and inputs the weld image into the weld recognition model to identify the weld type of the required welding position, and adjusts the welding current and the welding speed according to the weld type.

[0062] Preferably, the step S4 further includes:

[0063] Select a welding adjustment coefficient P according to the weld type, and adjust the welding initial current I 0 and the welding initial speed V 0 to obtain the adjusted actual welding current I S and the actual welding speed V S , and the adjustment method is as follows:

[0064] I S = I0 *P

[0065] V S = V 0 *P

[0066] If the weld type is a butt weld, the welding adjustment coefficient P is taken as 1.0 to 1.2;

[0067] If the weld type is a fillet weld, the welding adjustment coefficient P is taken as 0.9 to 1.1;

[0068] If the weld type is a T-joint weld, the welding adjustment coefficient P is taken as 1.2 to 1.4;

[0069] If the weld type is a lap weld, the welding adjustment coefficient P is taken as 0.8 to 1.0;

[0070] If the weld type is a plug weld, the welding adjustment coefficient P is taken as 1.3 to 1.5;

[0071] If the weld type is a slot weld, the welding adjustment coefficient P is taken as 1.1 to 1.3.

[0072] An internal welding control system according to the present application has the following advantages:

[0073] In an internal welding control system of the present application, by arranging the cleaning module at the moving front end of the moving module, during the movement of the moving module inside the workpiece, the cleaning module can pre-clean impurities such as particles, stains, and dust on the path, avoiding the influence of impurities on the stable movement of the moving module. Furthermore, the moving module can move smoothly along the predetermined route and ensure the continuous stability of subsequent welding; the guiding module can guide the moving module, enabling the moving module to move along the predetermined path inside the workpiece and accurately reach the position where welding is required, effectively improving the accuracy and reliability of subsequent welding. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 is a schematic structural diagram of an internal welding control system according to the present application;

[0075] Figure 2 is a schematic structural diagram of the moving module of an internal welding control system according to the present application;

[0076] Figure 3 is a schematic structural diagram of the cleaning module of an internal welding control system according to the present application;

[0077] Figure 4 is a schematic structural diagram of the T-shaped support column of an internal welding control system according to the present application;

[0078] Figure 5 is a schematic structural diagram of a moving guiding mechanism of an internal welding control system described in this application;

[0079] Figure 6 is a flowchart of a welding control strategy in an internal welding control system described in this application.

[0080] Description of reference numerals:

[0081] 10 - Moving module; 101 - Base; 102 - Moving wheel; 103 - Handle;

[0082] 20 - Cleaning module; 201 - Cleaning rake mechanism; 202 - Pressure spring piece;

[0083] 30 - Guiding module; 301 - Moving guiding mechanism; 3011 - Support plate 3011; 3012 - First rotating plate; 3013 - Second rotating plate; 3014 - First guiding ball; 3015 - Second guiding ball;

[0084] 40 - Welding module;

[0085] 50 - Wire feeding module;

[0086] 60 - T-shaped support column. Detailed implementation manners

[0087] As Figure 1 - Figure 6 shown, an internal welding control system described in this application includes:

[0088] A moving module 10, and the moving module 10 is used for horizontal movement inside a workpiece; specifically, the moving module 10 includes a base 101, moving wheels 102, a handle 103, and a driving motor disposed inside the base 101, and the driving motor is used for driving the moving module 10 to move. A plurality of moving wheels 102 are respectively rotatably connected to both sides of the base 101, and moving wheels 102 are rotatably connected to both sides of the base 101 near the front end and near the rear end, that is, moving wheels 102 are provided on both sides of the front end and the rear end of the base 101;

[0089] The handle 103 is disposed at the front end of the base 101. When the driving motor is not in use, the base 101 can be pulled or pushed by applying an external force to the handle 103, so that the base 101 moves in the horizontal direction;

[0090] The cleaning module 20 is disposed at one end of the moving module 10 and is used to clean the moving path of the moving module 10. The moving module 10 uses the end where the cleaning module 20 is provided as the moving front end and the other end as the moving rear end. Specifically, the cleaning module 20 includes a cleaning scraper mechanism 201 and a pressure spring piece 202.

[0091] One end of the pressure spring piece 202 is connected to the moving module 10, and the other end is connected to the cleaning scraper mechanism 201.

[0092] The cleaning scraper mechanism 201 is slidably connected to the front end of the moving module 10, and the cleaning scraper mechanism 201 moves upward or downward along the vertical direction on the moving module 10 through the elastic force of the pressure spring piece 202. That is, the pressure spring piece 202 has an elastic force downward along the vertical direction. When an external force upward in the vertical direction acts on the cleaning scraper mechanism 201, the cleaning scraper mechanism 201 presses against the pressure spring piece 202, causing the cleaning scraper mechanism 201 to move upward along the vertical direction. After no external force acts on the cleaning scraper mechanism 201, the cleaning scraper mechanism 201 moves downward along the vertical direction through the elastic force of the pressure spring piece 202, so that when passing through a non-flat road condition, the cleaning scraper mechanism 201 can be adjusted along the vertical direction to avoid jamming and affecting the movement of the moving module 10.

[0093] The guiding module 30 is disposed on the moving module 10 and is used to guide the movement of the moving module 10 inside the workpiece.

[0094] The welding module 40 is disposed at the moving rear end of the moving module 10 and is used to perform welding processing on the inside of the workpiece. The welding module 40 includes two sets of welding torch assemblies arranged oppositely. The welding end of the welding torch assembly extends in a direction away from the moving rear end of the moving module 10, and the wire feeding end of the welding torch assembly is close to the moving rear end of the moving module 10.

[0095] The wire feeding module 50 is disposed on the moving module 10, and the wire feeding module 50 is connected to the welding module 40 and is used to convey welding wire to the welding module 40. The wire feeding module 50 includes a wire feeding motor and a wire feeding mechanism. The output end of the wire feeding motor is rotationally connected to the wire feeding mechanism, and the wire feeding end of the wire feeding mechanism is connected to the wire feeding end of the welding torch assembly. The wire feeding mechanism conveys the welding wire to the welding torch assembly through the drive of the wire feeding motor.

[0096] The welding module 40 and the wire feeding module 50 are connected to the moving module 10 through a T-shaped support column 60.

[0097] The control module. The moving module 10, the cleaning module 20, the guiding module 30, the welding module 40, and the wire feeding module 50 are all signal-connected to the control module.

[0098] The internal welding control system sets the cleaning module 20 at the moving front end of the moving module 10. During the movement of the moving module 10 inside the workpiece, the cleaning module 20 can pre-clean impurities such as particles, stains, and dust on the path, avoiding the influence of impurities on the smooth movement of the moving module 10. Furthermore, the moving module 10 can move smoothly along the predetermined route and ensure the continuous stability of subsequent welding; the guiding module 30 can guide the moving module 10, enabling the moving module 10 to move along the predetermined path inside the workpiece and accurately reach the position where welding is required, effectively improving the accuracy and reliability of subsequent welding.

[0099] Furthermore, in this embodiment, the guiding module 30 includes two relatively arranged moving guiding mechanisms 301. The two moving guiding mechanisms 301 are respectively arranged close to both sides of the moving module 10, such that the two moving guiding mechanisms 301 are respectively in contact with the two side walls inside the workpiece inside the workpiece.

[0100] Each moving guiding mechanism 301 includes a support plate 3011, a first rotating plate 3012, a second rotating plate 3013, a first guiding ball 3014, and a second guiding ball 3015.

[0101] The middle of the support plate 3011 is connected to the T-shaped support column 60 through a connecting rod, and both ends of the support plate 3011 extend towards the moving front end and the moving rear end of the moving module 10 respectively; one end of the connecting rod is connected to the middle of the lower surface of the support plate 3011, and the other end of the connecting rod is connected to the upper surface of the T-shaped support column 60.

[0102] One end of the first rotating plate 3012 and one end of the second rotating plate 3013 are respectively rotatably connected to both ends of the support plate 3011, and the first rotating plate 3012 is close to the moving front end of the moving module 10, and the second rotating plate 3013 is close to the moving rear end of the moving module 10. Both the first rotating plate 3042 and the second rotating plate 3043 rotate horizontally around the support plate 3041.

[0103] The end of the first rotating plate 3012 far from the support plate 3011 is rotatably connected to the first guiding ball 3014, and the end of the second rotating plate 3013 far from the support plate 3011 is rotatably connected to the second guiding ball 3015.

[0104] During use, the moving guiding mechanism 301 located on the moving module 10 adjusts the position of the moving guiding mechanism 301 by the horizontal rotation of the first rotating plate 3042 and the second rotating plate 3043 around the support plate 3041 to adapt to the width of the workpiece.

[0105] A return spring is provided at the rotational connection between the first rotating plate 3012 and the support plate 3011, and at the rotational connection between the second rotating plate 3013 and the support plate 3011. Both the first rotating plate 3012 and the second rotating plate 3013 can rotate back to their initial positions through the resilience of the return spring, that is, rotate away from the moving module 10, so that when inside the workpiece, the first guiding ball 3014 and the second guiding ball 3015 on the same moving guiding mechanism 301 are in contact with the inner sidewall of the workpiece and have a certain pressure. The pressure is the force generated by the inner sidewall of the workpiece on the first guiding ball 3014 and the second guiding ball 3015. When guiding the moving module 10, if the pressure suddenly changes or becomes 0, it indicates that the movement of the moving module 10 is offset or abnormal.

[0106] During the movement of the moving module 10, the first guiding ball 3014 and the second guiding ball 3015 will rotate, preventing the first guiding ball 3014 and the second guiding ball 3015 from not moving and generating frictional force with the inner sidewall of the workpiece to hinder the movement of the moving module 10.

[0107] Further, in this embodiment, the control module controls the moving module 10, the cleaning module 20, the guiding module 30, the welding module 40, and the wire feeding module 50 respectively according to the welding control strategy to complete the welding process.

[0108] The welding control strategy includes the following steps:

[0109] S1. Select welding process parameters according to the parameter information of the workpiece to be welded and set them.

[0110] S2. Set an image acquisition component on the moving module 10, and collect image data of the internal path of the workpiece in real time through the image acquisition component. Select the moving speed of the moving module 10 according to the impurity density on the internal path of the workpiece. The image acquisition component can select multiple industrial cameras to be set on the moving module 10, respectively used to capture the path image at the front end of the moving module 10 and the weld image at the subsequent welding position required.

[0111] S3. Guide the movement of the moving module 10 according to the guiding module 30.

[0112] S4. Build a weld recognition model. When the moving module 10 moves to the required welding position, dynamically adjust the welding current and welding speed according to the weld type.

[0113] Further, in this embodiment, step S1 specifically includes:

[0114] The parameter information includes the workpiece thickness and the workpiece material type coefficient.

[0115] The welding process parameters include the initial welding current and the initial welding speed.

[0116] Obtain the workpiece thickness D and the workpiece material type coefficient K of the workpiece ω , and calculate the initial welding current I of the workpiece 0 and the initial welding speed V 0 :

[0117] I 0 = I L * K ω * D * α

[0118]

[0119] Wherein, I L represents the theoretical welding current value; α represents the current adjustment coefficient; V L represents the theoretical welding speed value; K V represents the comprehensive influence coefficient of speed; β represents the speed adjustment coefficient.

[0120] Furthermore, in this embodiment, step S1 further includes:

[0121] The value of the workpiece material type coefficient K ω is selected according to the workpiece material type:

[0122] If the workpiece material type is carbon steel, the workpiece material type coefficient K ω takes a value of 1.0 to 1.3; when the workpiece material type is carbon steel, K ω has an optimal value of 1.2;

[0123] If the workpiece material type is stainless steel, the workpiece material type coefficient K ω takes a value of 1.4 to 1.7; when the workpiece material type is stainless steel, K ω has an optimal value of 1.5;

[0124] If the workpiece material type is alloy material, the workpiece material type coefficient K ω takes a value of 1.8 to 2.0; when the workpiece material type is alloy material, K ω has an optimal value of 1.9;

[0125] An example of step S1 is as follows:

[0126] The workpiece thickness D = 20 mm, the theoretical welding current value I L = 10 A, the current adjustment coefficient α = 0.05 A / mm, the theoretical welding speed value V L = 20 mm / min, the comprehensive influence coefficient of speed K V = 10, the speed adjustment coefficient β = 2.5 mm / min;

[0127] The workpiece material type is carbon steel, then the workpiece material type coefficient Kω Select K ω = 1.2;

[0128] The initial current I is calculated to be 0 = 10 * 0.05 * 20 * 1.2 = 12 A;

[0129] Initial welding speed

[0130] Furthermore, in this embodiment, step S2 specifically includes:

[0131] The moving module 10 makes a forward movement towards the inside of the workpiece with the moving front end, and the image acquisition component performs real-time image acquisition on the path of the forward movement of the moving module 10 to obtain a path image;

[0132] The path image is transmitted to the edge computing gateway, and the path image data is grayscaled by the weighted average method to obtain a grayscale image;

[0133] The median filtering algorithm is used to denoise the grayscale image to obtain a first path image;

[0134] The histogram equalization technique is used to enhance the first path image to obtain a second path image;

[0135] The second path image is successively subjected to image segmentation processing and morphological processing to obtain a third path image. The number of pixel points in the impurity part of the third path image is counted, and the impurity area S is calculated by using an image analysis algorithm. The impurity density M is calculated according to the following formula:

[0136]

[0137] where the unit of the number of pixel points N is piece; the unit of the impurity area S is square pixel; the unit of the impurity density M is piece / square pixel;

[0138] Examples are as follows:

[0139] The image acquisition component can select an industrial camera. The moving module 10 moves from the workpiece port towards the inside of the workpiece at the theoretical initial moving speed, and the industrial camera acquires images of the forward path of the moving module 10 in real time at 30 frames per second and transmits the real-time acquired path images to the edge computing gateway;

[0140] After receiving the real-time transmitted path image, the edge computing gateway performs grayscaling by the weighted average method. For example, for a certain frame of the real-time acquired color image, the RGB value of one pixel point is (190, 160, 130). According to the weighting coefficient W R = 0.299, W G = 0.587, W B= 0.114 to calculate the grayscale value of this pixel point Gray = 0.299 * 190 + 0.587 * 160 + 0.114 * 130 = 164.23. Perform the same calculation for all pixel points of the entire frame of the image to obtain a grayscale image;

[0141] Apply the median filtering algorithm to the grayscale image to remove noise. For example, select a 3×3 filtering window. If the pixel values within the window are 150, 155, 145, 140, 160, 165, 135, 170, 175, sort these values from smallest to largest as 135, 140, 145, 150, 155, 160, 165, 170, 175, and take the median value 155 to replace the original value of the pixel at the center of the window. Traverse the entire frame of the grayscale image to obtain the first path image after denoising;

[0142] Use histogram equalization technology to enhance the first path image. Count the number of pixels at each grayscale level of this frame of the image, calculate the cumulative distribution function, and redistribute the pixel grayscale values. For example, a pixel with an original grayscale level of 120 becomes 140 after processing. After processing, obtain the second path image to make the impurity details clearer in the image;

[0143] Perform real-time image segmentation on the second path image. Use the U-Net model based on machine learning to quickly and accurately distinguish the impurity area and the background area. Then perform morphological processing to further optimize the contour of the impurity area through erosion and dilation operations. Then, in real-time, count the number of pixel points N in the impurity part and calculate the area S of the impurity area. For example, in real-time, count the number of pixel points N in the impurity part = 300, and calculate the area S of the impurity area = 120 square pixels. Then the impurity density M is:

[0144]

[0145] So the impurity density M = 2.4 pieces / square pixel.

[0146] Furthermore, in this embodiment, step S2 further includes:

[0147] The moving speed of the moving module 10 selects the moving speed V according to the impurity density M Y ;

[0148] According to the pre-established mapping relationship between the impurity density and the moving speed range, select the moving speed adjustment coefficient C, and dynamically adjust the moving speed V according to the moving speed adjustment coefficient C Y ;

[0149] If 0 ≤ M < 2, the value of the moving speed adjustment coefficient C is:

[0150]

[0151] If 2 ≤ M < 10, the value range of the movement speed adjustment coefficient C is:

[0152]

[0153] If M ≥ 10, the value range of the movement speed adjustment coefficient C is:

[0154]

[0155] Movement speed V Y Is calculated according to the following formula:

[0156] V Y = V YL * C

[0157] Wherein, V YL Represents the theoretical initial movement speed;

[0158] Examples are as follows:

[0159] According to the above example, the impurity density M = 2.4 per square pixel, then 2 ≤ M < 10, so the value range of the movement speed adjustment coefficient C is:

[0160]

[0161] The theoretical initial movement speed V YL = 2 mm / s, then the movement speed V Y = 2 * 1.1 = 2.2 mm / s.

[0162] Furthermore, in this embodiment, step S3 specifically includes:

[0163] Pressure sensors are arranged at the positions where the two movement guiding mechanisms 301 are in contact with the inner side wall of the workpiece, and the pressure values of the two movement guiding mechanisms 301 are monitored in real time to obtain the real-time guiding pressure value F DX1 And F DX2 ; According to the structure of the above movement guiding mechanism 301, the collected pressure value is the sum of the pressures of the first guiding ball 3014 and the second guiding ball 3015 on the same movement guiding mechanism 301 and the inner side wall of the workpiece;

[0164] If F DX1 = [F Dmin , F Dmax and F DX2 = [F Dmin , F Dmax , it is determined that the movement trajectory of the movement module 10 is normal, and the original movement state is maintained;

[0165] If F DX1 ≠ [F Dmin , F Dmaxand / or F DX2 ≠ F Dmin ,F Dmax ,then it is determined that the movement trajectory of the moving module 10 is abnormal, an alarm message is sent, and a movement adjustment operation is performed;

[0166] Among them, F Dmin represents the minimum value of the preset guiding pressure; F Dmax represents the maximum value of the preset guiding pressure;

[0167] The examples are as follows:

[0168] According to the historical welding data and test data, the minimum value of the preset guiding pressure F Dmin = 5N, and the maximum value of the preset guiding pressure F Dmax = 10N;

[0169] Since the two moving guiding mechanisms 301 are respectively arranged close to both sides of the moving module 10, that is, the left side and the right side of the moving module 10, F DX1 corresponds to the pressure value of the moving guiding mechanism 301 on the left side of the moving module 10, and F DX2 corresponds to the pressure value of the moving guiding mechanism 301 on the right side of the moving module 10;

[0170] The pressure values of the two moving guiding mechanisms 301 are monitored in real time, and at a certain moment, F DX1 = 7N, F DX2 = 8N, satisfying F DX1 = [F Dmin ,F Dmax and F DX2 = [F Dmin ,F Dmax , then it is determined that the movement trajectory of the moving module 10 is normal, and the original movement state is maintained;

[0171] If at a subsequent moment, it is monitored in real time that F DX1 = 12N, F DX2 = 3N, satisfying F DX1 ≠ [F Dmin ,F Dmax and F DX2 ≠ [F Dmin ,F Dmax , then it is determined that the movement trajectory of the moving module 10 is abnormal, an alarm message is sent, and the movement trajectory of the moving module 10 is adjusted through the control module so that it satisfies F DX1 = [F Dmin ,F Dmax and F DX2 = [F Dmin ,F Dmax .

[0172] Further, in this embodiment, step S4 specifically includes:

[0173] Obtain historical weld image data and corresponding weld type label data, where the weld type labels of the weld type label data include butt welds, fillet welds, T-welds, lap welds, plug welds, and groove welds;

[0174] Perform grayscale processing, denoising processing, and image enhancement processing on the historical weld image data to obtain the processed first historical weld image data;

[0175] Use the Canny algorithm and the gray-level co-occurrence matrix method to extract the edge features, geometric features, and texture features of the first historical weld image data to obtain a weld feature vector;

[0176] Use a support vector machine, with the weld feature vector as the input and the weld type label as the output, to perform model training and testing on the historical weld image data and the weld type label data to obtain a trained weld recognition model;

[0177] The moving module 10 moves to the required welding location, the image acquisition component acquires the weld image at the required welding location, and inputs the weld image into the weld recognition model. The weld type at the required welding location is identified, and the welding current and welding speed are adjusted according to the weld type;

[0178] Examples are as follows:

[0179] Obtain 1000 pieces of historical weld image data from the existing welding database. At the same time, the corresponding weld type label data includes 300 butt welds, 250 fillet welds, 200 T-welds, 150 lap welds, 50 plug welds, and 50 groove welds;

[0180] For a color image of a historical butt weld, the RGB value of its pixel is (200, 180, 160). According to the weighting coefficients W R = 0.299, W G = 0.587, W B = 0.114, calculate the gray value of this pixel Gray = 0.299 * 200 + 0.587 * 180 + 0.114 * 160 = 178.28. Perform the same operation on all pixels of these 1000 historical weld images to achieve grayscale of each image;

[0181] The median filtering algorithm is adopted. For the grayscale image, a 3×3 filtering window is taken. For example, the pixel values within the window are 160, 165, 155, 150, 170, 175, 145, 180, 185. After sorting, the median value 165 is taken to replace the pixel value at the center of the window. The entire image is traversed to complete denoising, and the histogram equalization technique is used to count the number of pixels at each gray level of the image, redistribute the pixel gray values, enhance the image contrast, and obtain the first historical weld image data;

[0182] The Canny algorithm is used, with the low threshold set to 50 and the high threshold set to 150. A processed T-shaped weld image is processed to detect the weld edge;

[0183] Geometric feature extraction includes analyzing the geometric information of the shape, length, and width of the weld. For example, for a lap weld, the length and width of the lap part are measured;

[0184] Texture feature extraction is carried out by using the gray-level co-occurrence matrix method. The contrast and entropy texture parameters are calculated to obtain the texture features, and the weld feature vectors of all images are comprehensively obtained;

[0185] Taking the weld feature vector as the input and the weld type label as the output, a support vector machine is used for model training. After multiple rounds of training and parameter adjustment, 800 image data are used for training and 200 image data are used for testing to obtain the trained weld recognition model;

[0186] The moving module 10 moves to the required welding position. The image acquisition component acquires the weld image of the required welding position and inputs the weld image into the weld recognition model. For example, when reaching the first required welding position, if the acquired image is recognized as a butt weld by the weld recognition model, the welding current and welding speed are adjusted according to the butt weld;

[0187] Furthermore, in this embodiment, step S4 further includes:

[0188] Select the welding adjustment coefficient P according to the weld type, and adjust the welding initial current I 0 and the welding initial speed V 0 to obtain the adjusted actual welding current I S and the actual welding speed V S , and the adjustment method is as follows:

[0189] I S =I 0 *P

[0190] V S =V 0 *P

[0191] If the weld type is butt weld, the welding adjustment coefficient P takes values from 1.0 to 1.2; when the weld type is butt weld, the optimal value of the welding adjustment coefficient P is 1.1;

[0192] If the weld type is fillet weld, the welding adjustment coefficient P takes values from 0.9 to 1.1; when the weld type is fillet weld, the optimal value of the welding adjustment coefficient P is 1.0;

[0193] If the weld type is T-joint weld, the welding adjustment coefficient P takes values from 1.2 to 1.4; when the weld type is T-joint weld, the optimal value of the welding adjustment coefficient P is 1.3;

[0194] If the weld type is lap weld, the welding adjustment coefficient P takes values from 0.8 to 1.0; when the weld type is lap weld, the optimal value of the welding adjustment coefficient P is 0.9;

[0195] If the weld type is plug weld, the welding adjustment coefficient P takes values from 1.3 to 1.5; when the weld type is plug weld, the optimal value of the welding adjustment coefficient P is 1.4;

[0196] If the weld type is slot weld, the welding adjustment coefficient P takes values from 1.1 to 1.3; when the weld type is slot weld, the optimal value of the welding adjustment coefficient P is 1.2;

[0197] Examples are as follows:

[0198] If it is identified as a butt weld through the weld recognition model according to the above example, the welding adjustment coefficient P takes values from 1.0 to 1.2, and P = 1.1 is selected;

[0199] Welding initial current I 0 = 12A, welding initial speed V 0 = 21mm / min;

[0200] Then the actual welding current I S = 12 * 1.1 = 13.2A;

[0201] Actual welding speed V S = 21 * 1.1 = 23.1mm / min.

[0202] In the description of this application, it should be understood that the orientation or positional relationship indicated by orientation words such as "front, back, up, down, left, right", "horizontal, vertical, level" and "top, bottom", etc. is usually based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing this application and simplifying the description. Without contrary explanations, these orientation words do not indicate and imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, so it cannot be understood as a limitation on the protection scope of this application.

[0203] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all such changes and deformations should fall within the protection scope of the claims of this application.

Claims

1. An internal welding control system for welding the inside of a workpiece, characterized in that: include: A moving module (10), wherein the moving module (10) is used to move horizontally inside the workpiece; A cleaning module (20), the cleaning module (20) being arranged at one end of the moving module (10) and being used for cleaning the moving path of the moving module (10), the moving module (10) having the end provided with the cleaning module (20) as a moving front end and the other end as a moving rear end; A guide module (30), the guide module (30) being arranged on the moving module (10) and being used for guiding the movement of the moving module (10) inside the workpiece; A welding module (40), the welding module (40) being arranged at the rear end of the movable module (10) and being used for performing welding processing on the inside of a workpiece; A wire feeding module (50), wherein the wire feeding module (50) is arranged on the moving module (10), and the wire feeding module (50) is connected to the welding module (40) and is used to feed welding wire to the welding module (40); The control module, the moving module (10), the cleaning module (20), the guiding module (30), the welding module (40) and the wire feeding module (50) are all connected to the control module by signals.

2. The internal welding control system according to claim 1, characterized in that: The guide module (30) comprises two oppositely arranged movable guide mechanisms (301), and the two movable guide mechanisms (301) are respectively arranged close to two sides of the movable module (10), so that the two movable guide mechanisms (301) are respectively in contact with two side walls inside the workpiece.

3. The internal welding control system according to claim 2, characterized in that: The control module controls the moving module (10), the cleaning module (20), the guiding module (30), the welding module (40) and the wire feeding module (50) respectively according to a welding control strategy to complete the welding process; The welding control strategy includes the following steps: S1. Select welding process parameters and set them according to the parameter information of the required welding workpiece; S2, arranging an image acquisition component on the mobile module (10), collecting image data of the internal path of the workpiece in real time through the image acquisition component, and selecting the moving speed of the mobile module (10) according to the impurity density on the internal path of the workpiece; S3, guiding the movement of the moving module (10) according to the guiding module (30); S4, constructing a weld recognition model, the mobile module (10) moves to the desired welding location, and dynamically adjusts the welding current and welding speed according to the weld type.

4. The internal welding control system according to claim 3, characterized in that: The step S1 specifically includes: The parameter information includes workpiece thickness and workpiece material type coefficient; The welding process parameters include initial welding current and initial welding speed; Obtain the workpiece thickness D and workpiece material type coefficient K of the workpiece ω , and calculate the initial welding current I0 and initial welding speed V0 of the workpiece: I0=I L *K ω *D*α Among them, I L represents the theoretical welding current value; α represents the current adjustment coefficient; V L Indicates the theoretical welding speed value; K V represents the comprehensive speed influence coefficient; β represents the speed adjustment coefficient.

5. The internal welding control system according to claim 4, characterized in that: The step S1 further comprises: The workpiece material type coefficient K ω The value of is selected according to the workpiece material type: If the workpiece material type is carbon steel, then the workpiece material type coefficient K ω The value ranges from 1.0 to 1.3; If the workpiece material type is stainless steel, then the workpiece material type coefficient K ω The value ranges from 1.4 to 1.7; If the workpiece material type is alloy material, then the workpiece material type coefficient K ω The value ranges from 1.8 to 2.

0.

6. The internal welding control system according to claim 3, characterized in that: The step S2 specifically includes: The mobile module (10) moves forward with the mobile front end toward the inside of the workpiece, and the image acquisition component acquires images of the path along which the mobile module (10) moves in real time to obtain a path image; The path image is transmitted to the edge computing gateway, and the path image data is grayed by a weighted average method to obtain a grayscale image; Using a median filter algorithm to perform denoising on the grayscale image to obtain a first path image; Performing image enhancement processing on the first path image by using a histogram equalization technique to obtain a second path image; The second path image is subjected to image segmentation processing and morphological processing in sequence to obtain a third path image, the number of pixels of the impurity part in the third path image is counted, and the impurity area S is calculated using an image analysis algorithm, and the impurity density M is calculated according to the following formula: The unit of the number of pixels N is pieces; the unit of the area of ​​the impurity region S is square pixels; and the unit of the impurity density M is pieces / square pixel.

7. The internal welding control system according to claim 6, characterized in that: The step S2 further comprises: The moving speed of the moving module (10) is selected according to the impurity density M. Y ; According to the pre-established mapping relationship between impurity density and moving speed range, a moving speed adjustment coefficient C is selected, and the moving speed V is dynamically adjusted according to the moving speed adjustment coefficient C. Y ; If 0≤M<2, the value of the moving speed adjustment coefficient C is: If 2≤M<10, the value of the moving speed adjustment coefficient C is: If M≥10, the value of the moving speed adjustment coefficient C is: The moving speed V Y Calculate according to the following formula: In Y =V YL *C Among them, V YL Indicates the theoretical initial moving speed.

8. The internal welding control system according to claim 3, characterized in that: The step S3 specifically includes: Pressure sensors are provided at the positions where the two movable guide mechanisms (301) contact the inner side wall of the workpiece, and the pressure values ​​of the two movable guide mechanisms (301) are monitored in real time to obtain a real-time guide pressure value F DX1 and F DX2 ; If F DX1 =[F Dmin , F Dmax ] and F DX2 =[F Dmin , F Dmax ], it is determined that the motion trajectory of the mobile module (10) is normal and the original motion state is maintained; If F DX1 ≠[F Dmin , F Dmax ] and / or F DX2 ≠[F Dmin , F Dmax ], it is determined that the motion trajectory of the mobile module (10) is abnormal, an alarm message is issued, and a motion adjustment operation is performed; Among them, F Dmin Indicates the minimum preset pilot pressure; F Dmax Indicates the preset maximum pilot pressure.

9. The internal welding control system according to claim 5, characterized in that: The step S4 specifically includes: Acquire historical weld image data and corresponding weld type label data, wherein the weld type labels of the weld type label data include butt welds, fillet welds, T-welds, lap welds, plug welds, and slot welds; Performing grayscale processing, denoising processing, and image enhancement processing on the historical weld image data to obtain processed first historical weld image data; The Canny algorithm and the gray-level co-occurrence matrix method are used to extract edge features, geometric features and texture features of the first historical weld image data to obtain a weld feature vector; Using a support vector machine, with the weld feature vector as input and the weld type label as output, model training and testing are performed on the historical weld image data and the weld type label data to obtain a trained weld recognition model; The mobile module (10) moves to a desired welding location, the image acquisition component acquires a weld image of the desired welding location, and inputs the weld image into the weld recognition model to identify the weld type of the desired welding location, and adjusts the welding current and the welding speed according to the weld type.

10. The internal welding control system according to claim 9, characterized in that: The step S4 further comprises: The welding adjustment coefficient P is selected according to the weld type, and the welding initial current I0 and the welding initial speed V0 are adjusted according to the welding adjustment coefficient P to obtain the adjusted actual welding current I S and actual welding speed V S , the adjustment method is as follows: I S =I0*P V S =V0*P If the weld type is a butt weld, the welding adjustment coefficient P is 1.0 to 1.2; If the weld type is a fillet weld, the welding adjustment coefficient P is 0.9 to 1.1; If the weld type is a T-shaped weld, the welding adjustment coefficient P is 1.2 to 1.4; If the weld type is a lap weld, the welding adjustment coefficient P is 0.8 to 1.0; If the weld type is a plug weld, the welding adjustment coefficient P is 1.3 to 1.5; If the weld type is a slot weld, the welding adjustment coefficient P is 1.1 to 1.3.

Images