An intelligent welding method based on PLC intelligent control

Abstract

The present invention provides an intelligent welding method based on PLC intelligent control, comprising: using a light-touch positioning device to collect a plurality of first welding points on an object to be welded, collecting a real-time image of the object to be welded, pre-processing the real-time image to obtain a plurality of second welding points, using PLC to analyze the coincidence relationship between the first welding points and the second welding points, obtaining a plurality of target welding points on the object to be welded based on the coincidence relationship, obtaining and executing the welding process corresponding to each target welding point, so as to establish a set of automatic welding technology by using PLC and machine vision, replace manual welding with automatic welding, and improve welding efficiency.

Description

Technical Field

[0001] The invention relates to the field of intelligent control, and in particular to an intelligent welding method based on PLC intelligent control. Background Art

[0002] Welding is an increasingly ubiquitous process in all industries. While such processes can be automated in some cases, there continue to be a large number of applications for manual welding operations, whose success relies heavily on the welding operator using the welding gun or torch correctly. For example, improper torch angle, contact tip to workpiece distance, travel speed, and aim are parameters that can determine weld quality. However, even experienced welding operators often have difficulty monitoring and maintaining these important parameters throughout the welding process, making the creation of an automated welding technology to improve welding efficiency a pressing issue.

[0003] Therefore, the present invention provides an intelligent welding method based on PLC intelligent control. Summary of the invention

[0004] The present invention provides an intelligent welding method based on PLC intelligent control, which is used to establish a set of automatic welding technology by using PLC and machine vision, and replaces manual welding with automatic welding to improve welding efficiency.

[0005] The present invention provides an intelligent welding method based on PLC intelligent control, comprising:

[0006] Step 1: using a light-touch positioning device to collect a plurality of first welding points on the object to be welded;

[0007] Step 2: Collecting a live image of the object to be welded, preprocessing the live image, and obtaining a plurality of second welding points;

[0008] Step 3: Analyze the overlap relationship between the first welding point and the second welding point using PLC, and obtain a plurality of target welding points on the object to be welded based on the overlap relationship;

[0009] Step 4: Obtain the welding process corresponding to each target welding point and execute it.

[0010] In one practicable manner,

[0011] The optical touch positioning device is used to collect a plurality of first welding points on the object to be welded, including:

[0012] Scanning the object to be welded using a first preset light beam to obtain a scanned image;

[0013] Based on the scanned image, draw the appearance structure of the object to be welded, and obtain all key points in the appearance structure;

[0014] Scan each key point respectively using the second preset light to obtain the reflected light corresponding to each key point;

[0015] Analyze the light intensity corresponding to each reflected light, and extract the key points where the light intensity of the reflected light is within a preset light intensity range;

[0016] Each key point is marked on the scanned image and recorded as the first welding point.

[0017] In one practicable manner,

[0018] Collecting a live image of the object to be welded, preprocessing the live image, and obtaining a plurality of second welding points, including:

[0019] Collecting and analyzing the live image of the object to be welded to obtain pixel distribution information of the live image;

[0020] Dividing the live image into a plurality of image regions based on the pixel distribution information, and respectively obtaining image contours of each pair of image regions;

[0021] Using preset welding point samples to train each of the image contours to obtain a plurality of training contours;

[0022] Comparing each training contour with a preset welding point contour respectively, and extracting a training contour consistent with the preset welding point;

[0023] Each training contour is marked on the live image as a second welding point.

[0024] In one practicable manner,

[0025] The PLC is used to analyze the coincidence relationship between the first welding point and the second welding point, including:

[0026] Editing the first welding points by using PLC to obtain a first distribution layout of the first welding points, obtaining a plurality of first distribution points, and making first marks on the first distribution points in a rectangular coordinate system;

[0027] Editing the second welding points by using PLC to obtain a second distribution layout of the second welding points, obtaining a plurality of second distribution points, and performing a second marking on the second distribution points in a rectangular coordinate system;

[0028] An overlap mark between the first mark and the second mark is obtained, and an overlap relationship between the first welding point and the second welding point is established.

[0029] In one practicable manner,

[0030] Obtain the welding process corresponding to each target welding point and execute it, including:

[0031] Acquire the physical parameters of the object to be welded on the live image, obtain the welding scheme of the object to be welded in a preset database, and record the welding points with the same welding process as a process class according to the welding scheme;

[0032] Obtaining process parameters corresponding to each process type respectively, and establishing corresponding welding parameters for each process type respectively in combination with the physical parameters;

[0033] The welding scheme is analyzed to obtain a welding process, a target welding point corresponding to the welding process is obtained, and welding work is performed in combination with corresponding welding parameters.

[0034] In one practicable manner,

[0035] Drawing the appearance structure of the object to be welded based on the scanned image includes:

[0036] Obtaining the pixel value corresponding to each pixel point on the scanned image and establishing a pixel matrix;

[0037] Obtaining a membership value corresponding to each matrix element in the pixel matrix, dividing the matrix elements with consistent membership values ​​into one category, and obtaining a segmentation result;

[0038] Mapping the segmentation result onto the scanned image, dividing the scanned image into a plurality of image regions;

[0039] Obtain the regional grayscale value corresponding to each image region respectively, and obtain the structural height corresponding to each image region;

[0040] The regional features corresponding to each image region are established, and the appearance structure of the object to be welded is drawn in combination with the structural height corresponding to each image region.

[0041] In one practicable manner,

[0042] Editing the first welding points by using PLC to obtain a first distribution layout of the first welding points includes:

[0043] Obtain the reflected light corresponding to each first welding point, obtain the waveform width corresponding to each reflected light, and simultaneously obtain each reflected light, obtain the aperture range generated by each reflected light, and establish a waveform information list;

[0044] Inputting each first welding point into a preset space respectively to obtain an initial welding point model;

[0045] Using PLC to edit the waveform information list, obtaining a light model corresponding to each reflected light;

[0046] Superimposing the light model and the initial welding point model;

[0047] Training the initial welding point model based on the light model to obtain a target welding point model;

[0048] The target welding point model is planarized to obtain a first distribution layout of the first welding points.

[0049] In one practicable manner,

[0050] Training the initial welding point model based on the light model includes:

[0051] Analyze each virtual reflected light in the light model respectively to obtain a light starting point corresponding to each virtual light;

[0052] Marking the starting point of the light on the light model to obtain a plurality of reflection points;

[0053] Analyze each virtual reflected light in the light model respectively to obtain the light beam height corresponding to each virtual light;

[0054] Obtaining a corresponding reflection height based on the light beam height, and establishing a virtual reflector of a corresponding height at the reflection point;

[0055] Analyzing the initial welding point model to obtain a plurality of virtual welding points;

[0056] It is determined whether the virtual welding point and the virtual reflector at the corresponding position are at the same height. If not, the height of the virtual welding point is corrected, all trained virtual welding points are obtained, and a target welding point model is established.

[0057] In one practicable manner,

[0058] Each of the image contours is trained using a preset welding point sample to obtain a number of training contours, including:

[0059] Establishing an image screening standard using the preset welding sample;

[0060] Using the image screening standard to screen the image contour to obtain a first image contour;

[0061] Comparing the first image contour with a preset welding point sample to obtain corresponding contour outliers;

[0062] Performing mutual adaptation training on the contour outliers and the preset welding point samples to obtain a second image contour;

[0063] A second image contour consistent with the preset welding point sample is extracted to obtain a plurality of training contours.

[0064] In one practicable manner,

[0065] Using the preset welding sample, an image screening criteria is established, including:

[0066] Performing corrosion treatment and expansion treatment on the preset welding samples respectively to obtain corresponding corrosion samples and expansion samples;

[0067] Acquire a first image region corresponding to the eroded sample and a second image region corresponding to the expanded sample;

[0068] An image screening criterion is established based on the first image region and the second image region.

[0069] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0070] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0072] Figure 1 A schematic diagram of the working process of an intelligent welding method based on PLC intelligent control in an embodiment of the present invention;

[0073] Figure 2 It is a schematic diagram of a workflow of collecting a plurality of first welding points on an object to be welded by using a light-touch positioning device in an intelligent welding method based on PLC intelligent control in an embodiment of the present invention;

[0074] Figure 3 The present invention is a schematic diagram of a workflow for collecting a real-time image of an object to be welded, preprocessing the real-time image, and obtaining a plurality of second welding points in an intelligent welding method based on PLC intelligent control in an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0076] Embodiment 1:

[0077] This embodiment provides an intelligent welding method based on PLC intelligent control, such as Figure 1As shown, including:

[0078] Step 1: using a light-touch positioning device to collect a plurality of first welding points on the object to be welded;

[0079] Step 2: Collecting a live image of the object to be welded, preprocessing the live image, and obtaining a plurality of second welding points;

[0080] Step 3: Analyze the overlap relationship between the first welding point and the second welding point using PLC, and obtain a plurality of target welding points on the object to be welded based on the overlap relationship;

[0081] Step 4: Obtain the welding process corresponding to each target welding point and execute it.

[0082] In this example, the optical touch positioning device may be a laser scanner;

[0083] In this example, the object to be welded can be any raw material that needs to be welded;

[0084] In this example, the first welding point represents a welding point collected by the optical touch positioning device;

[0085] In this example, the second welding point represents a welding point acquired through image preprocessing.

[0086] The working principle and beneficial effects of the above technical solution: In order to realize automated welding by replacing manual welding with machine welding, the specific position of the welding point needs to be determined before welding, and then the welding work is performed. First, the welding points on the object to be welded are collected by two different methods, and then analyzed by PLC to obtain the exact position of the welding point, and then the welding process required for each welding point is analyzed, and finally welding is performed. In this way, the position of the welding point is determined by two collection methods, which improves the accuracy of welding, avoids the generation of defective products, and improves the efficiency of welding.

[0087] Example 2

[0088] On the basis of Example 1, the intelligent welding method based on PLC intelligent control uses a light-touch positioning device to collect a plurality of first welding points on the object to be welded, such as Figure 2 As shown, including:

[0089] Scanning the object to be welded using a first preset light beam to obtain a scanned image;

[0090] Based on the scanned image, draw the appearance structure of the object to be welded, and obtain all key points in the appearance structure;

[0091] Scan each key point respectively using the second preset light to obtain the reflected light corresponding to each key point;

[0092] Analyze the light intensity corresponding to each reflected light, and extract the key points where the light intensity of the reflected light is within a preset light intensity range;

[0093] Each key point is marked on the scanned image and recorded as the first welding point.

[0094] In this example, the optical touch positioning device can emit a first preset light beam and a second preset light beam;

[0095] In this example, the first preset light beam may be an infrared light beam;

[0096] In this example, the appearance structure represents an expression method for describing the appearance of the object to be welded in a virtual space;

[0097] In this example, key points represent points on non-straight regions in the appearance structure;

[0098] In this example, the second preset light may be ultraviolet light;

[0099] Acquiring all key points in the appearance structure includes: sharpening the appearance structure to obtain a sharpened appearance structure;

[0100] Then the sharpened appearance structure is stretched to extract edge data of the object to be welded;

[0101] Extracting data jump values ​​at different data bits in the edge data, and extracting target data bits whose data jump values ​​are within a preset range;

[0102] The position of the target data bit on the appearance structure is obtained and recorded as a key point.

[0103] The working principle and beneficial effects of the above technical solution are as follows: the position of the welding point is determined by a light-touch positioning device, and the first light beam is used to scan the object to be welded to obtain the approximate position of the welding point as the key point, and then the second light beam is used to scan the key point, and the reflected light of each key point is analyzed to determine whether it is a welding point. Finally, several first welding points can be obtained. In this way, not only the position of the welding point can be quickly obtained, but also the light scanning will not cause damage to the object to be welded.

[0104] Example 3

[0105] On the basis of Example 1, the intelligent welding method based on PLC intelligent control collects the real-time image of the object to be welded, pre-processes the real-time image, and obtains a plurality of second welding points, such as Figure 3 As shown, including:

[0106] Collecting and analyzing the live image of the object to be welded to obtain pixel distribution information of the live image;

[0107] Dividing the live image into a plurality of image regions based on the pixel distribution information, and respectively obtaining image contours of each pair of image regions;

[0108] Using preset welding point samples to train each of the image contours to obtain a plurality of training contours;

[0109] Comparing each training contour with a preset welding point contour respectively, and extracting a training contour consistent with the preset welding point;

[0110] Each training contour is marked on the live image as a second welding point.

[0111] In this example, the pixel distribution information represents the distribution of different pixel values ​​on the live image;

[0112] In this example, the image region refers to the region with the same pixel value on the live image;

[0113] In this example, the image contour represents the contour of an image region;

[0114] In this example, the preset welding point contour represents the contour of a standard welding point.

[0115] The working principle and beneficial effects of the above technical solution are as follows: the real-time image of the object to be welded is collected and analyzed through machine vision, and then the pixel contour on the real-time image is further trained according to the pixel distribution information of the real-time image combined with the preset welding point samples, and finally a few second welding points are obtained on the real-time image.

[0116] Example 4

[0117] On the basis of Example 1, the intelligent welding method based on PLC intelligent control uses PLC to analyze the coincidence relationship between the first welding point and the second welding point, including:

[0118] Editing the first welding points by using PLC to obtain a first distribution layout of the first welding points, obtaining a plurality of first distribution points, and making first marks on the first distribution points in a rectangular coordinate system;

[0119] Editing the second welding points by using PLC to obtain a second distribution layout of the second welding points, obtaining a plurality of second distribution points, and performing a second marking on the second distribution points in a rectangular coordinate system;

[0120] An overlap mark between the first mark and the second mark is obtained, and an overlap relationship between the first welding point and the second welding point is established.

[0121] In this example, the first distribution points correspond one-to-one to the first welding points, and the second distribution points correspond one-to-one to the second welding points;

[0122] In this example, the coincidence mark indicates the position where the first distribution point and the second distribution point coincide;

[0123] In this example, the coincident mark refers to the mark that corresponds when the first mark and the second mark are at the same position;

[0124] In this example, since the first mark corresponds to the first welding point and the second mark corresponds to the second welding point, an overlap relationship between the first welding point and the second welding point that overlap each other is established;

[0125] In this example, first marking the first distribution point in a rectangular coordinate system includes:

[0126] Select a target distribution point at the lower left corner from the first distribution points, and input the target distribution point into the origin position of a rectangular coordinate system;

[0127] Establishing a position vector between each remaining first distribution point and the target distribution point respectively;

[0128] Based on the position vector, the remaining first distribution points are input into the rectangular coordinate system to complete the first marking.

[0129] The working principle and beneficial effects of the above technical solution are as follows: after obtaining the first welding point and the second welding point, they are edited using PLC and then the edited distribution points are marked in a coordinate system, and the overlapping points of the two marks are taken. In this way, the corresponding relationship between the two welding points is obtained, which facilitates the subsequent welding work to be performed according to the corresponding relationship between the two welding points.

[0130] Example 5

[0131] On the basis of Example 1, the intelligent welding method based on PLC intelligent control obtains and executes the welding process corresponding to each target welding point, including:

[0132] Acquire the physical parameters of the object to be welded on the live image, obtain the welding scheme of the object to be welded in a preset database, and record the welding points with the same welding process as a process class according to the welding scheme;

[0133] Obtaining process parameters corresponding to each process type respectively, and establishing corresponding welding parameters for each process type respectively in combination with the physical parameters;

[0134] The welding scheme is analyzed to obtain a welding process, a target welding point corresponding to the welding process is obtained, and welding work is performed in combination with corresponding welding parameters.

[0135] In this example, the physical parameters represent the appearance parameters of the object to be welded, including: length, width, height, perimeter, and area;

[0136] In this example, the welding plan indicates the order in which welding work is performed on each welding point and the welding action performed for each welding point when a welding operation is performed on the object to be welded;

[0137] In this example, the process parameters represent the conditions required when performing the welding process;

[0138] In this example, the welding parameters represent the actual conditions when the welding work is performed under the constraints of the physical parameters of the object to be welded;

[0139] In this example, the process parameters corresponding to each process type are obtained respectively, and the corresponding welding parameters are established for each process type in combination with the physical parameters, including:

[0140] Obtaining process parameters corresponding to each process type, and establishing a process operation range based on the process parameters;

[0141] Based on the physical parameters, establishing a physical range;

[0142] Projecting the process operation range onto the physical object range to obtain an overlapping range;

[0143] The overlap range is analyzed to obtain welding parameters.

[0144] The working principle and beneficial effects of the above technical solution are as follows: in order to better perform welding work, a welding scheme is first matched for the object to be welded according to its actual parameters during welding. In order to facilitate the next step of analysis, welding points with consistent welding processes are recorded as a process class. According to the process parameters of the process class, the corresponding welding parameters are obtained in combination with the actual conditions of the object to be welded. Finally, the welding work can be performed in combination with the welding process, thereby achieving the purpose of welding automation and improving the accuracy of welding.

[0145] Example 6

[0146] On the basis of Example 2, the intelligent welding method based on PLC intelligent control draws the appearance structure of the object to be welded based on the scanned image, including:

[0147] Obtaining the pixel value corresponding to each pixel point on the scanned image and establishing a pixel matrix;

[0148] Obtaining a membership value corresponding to each matrix element in the pixel matrix, dividing the matrix elements with consistent membership values ​​into one category, and obtaining a segmentation result;

[0149] Mapping the segmentation result onto the scanned image, dividing the scanned image into a plurality of image regions;

[0150] Obtain the regional grayscale value corresponding to each image region respectively, and obtain the structural height corresponding to each image region;

[0151] The regional features corresponding to each image region are established, and the appearance structure of the object to be welded is drawn in combination with the structural height corresponding to each image region.

[0152] In this example, the pixel matrix represents a matrix established according to the pixel value corresponding to each point on the scanned image, wherein the matrix elements in the pixel matrix are consistent with the pixel values ​​of the pixel points at the corresponding positions;

[0153] In this example, the membership value represents the degree of membership between different matrix elements in the pixel matrix and the central matrix element. The higher the membership degree, the greater the membership value, and vice versa.

[0154] In this example, the structure height indicates the height of a structure on the object to be welded. Since each image area corresponds to a physical area on the object to be welded, and since the heights of the physical areas are not necessarily consistent, different grayscale values ​​will appear on the scanned image.

[0155] In this example, the regional features include the area and shape of the region;

[0156] In this example, the appearance structure includes the area, shape, and height of a point on the object to be welded.

[0157] Example to verify this example: Get the pixel value of each pixel on the scanned image, and you can build a pixel matrix Then the membership values ​​of each matrix element in the pixel matrix are obtained as follows: 0.21, 0.19, 0.33, 0.55, 1, 0.35, 0.57, 0.47, 0.45, and then the matrix can be divided into The segmentation result is then mapped to the scanned image to obtain 5 image areas, and the regional grayscale values ​​corresponding to each image area are obtained, which are: H1, H2, H3, H4, H5, and the structural heights corresponding to each image area are obtained: G1, G2, G3, G4, G5. Then the regional features of each image area are established, which are: Y1, Y2, Y3, Y4, Y5. In this way, the 5 appearance structures of the object to be welded can be obtained: GY1, GY2, GY3, GY4, GY5.

[0158] The working principle and beneficial effects of the above technical solution: Since the welding work corresponding to different welding points is not the same, it is necessary to analyze the appearance structure of the object to be welded before performing the welding work. First, a pixel matrix is ​​established according to the pixel value corresponding to each pixel point on the scanned image, and then the matrix is ​​segmented according to the membership value of each element in the matrix. The segmentation result is projected onto the scanned image using the mapping principle. In this way, the scanned image is also divided into several image areas. Then, the structural height of each image area is established according to the grayscale value of the area. Finally, combined with the regional characteristics of the area, the appearance structure of the object to be welded is drawn to facilitate subsequent query of welding points.

[0159] Example 7

[0160] On the basis of Embodiment 4, the intelligent welding method based on PLC intelligent control uses PLC to edit the first welding points to obtain a first distribution layout of the first welding points, including:

[0161] Obtain the reflected light corresponding to each first welding point, obtain the waveform width corresponding to each reflected light, and simultaneously obtain each reflected light, obtain the aperture range generated by each reflected light, and establish a waveform information list;

[0162] Inputting each first welding point into a preset space respectively to obtain an initial welding point model;

[0163] Using PLC to edit the waveform information list, obtaining a light model corresponding to each reflected light;

[0164] Superimposing the light model and the initial welding point model;

[0165] Training the initial welding point model based on the light model to obtain a target welding point model;

[0166] The target welding point model is planarized to obtain a first distribution layout of the first welding points.

[0167] In this example, the waveform width indicates the width of the reflected light;

[0168] In this example, the aperture range indicates the range covered by the reflected light;

[0169] In this example, the waveform information list is edited using PLC, including:

[0170] Inputting the waveform information list into a PLC editing space, and editing the waveform information list into a target expression in the PLC editing space;

[0171] Interpreting the target expression to obtain a propagation path of light corresponding to each reflected light ray;

[0172] The propagation path is compiled into digital information.

[0173] The working principle and beneficial effects of the above technical solution are as follows: in analyzing the first distribution layout, the idea of ​​reverse reasoning is used to obtain the reflected light corresponding to each welding point, and then the reflected light is analyzed to determine the source of each reflected light, establish a light model, and at the same time establish an initial welding model based on the welding point. In order to further test and correct the first welding point, the light model is superimposed on the initial welding model, and the position of the welding point is corrected according to the superposition result. Finally, planarization processing is performed to obtain the first distribution layout of the first welding point.

[0174] Example 8

[0175] On the basis of Example 7, the intelligent welding method based on PLC intelligent control, training the initial welding point model based on the light model, includes:

[0176] Analyze each virtual reflected light in the light model respectively to obtain a light starting point corresponding to each virtual light;

[0177] Marking the starting point of the light on the light model to obtain a plurality of reflection points;

[0178] Analyze each virtual reflected light in the light model respectively to obtain the light beam height corresponding to each virtual light;

[0179] Obtaining a corresponding reflection height based on the light beam height, and establishing a virtual reflector of a corresponding height at the reflection point;

[0180] Analyzing the initial welding point model to obtain a plurality of virtual welding points;

[0181] It is determined whether the virtual welding point and the virtual reflector at the corresponding position are at the same height. If not, the height of the virtual welding point is corrected, all trained virtual welding points are obtained, and a target welding point model is established.

[0182] In this example, the reflection points correspond one-to-one with the starting points of the light rays;

[0183] In this example, the beam height is inversely proportional to the reflection height;

[0184] Take an example to verify this instance: analyze each reflected light in the light model, obtain the light starting point Dn corresponding to each virtual light, then obtain several corresponding reflection points Fn on the light model, and then analyze the beam height Kn of each virtual light, and compare Kn with the height of the virtual welding point on the initial welding model. If the two are not the same height, correct the height of the virtual welding point to Kn, and finally establish the target welding point model.

[0185] The working principle and beneficial effects of the above technical solution are as follows: by analyzing each reflected light in the light model, the starting point of each virtual light is obtained, the position of the virtual welding point is analyzed from the side through the starting point of the light, and then compared with the existing position of the virtual welding point in the initial welding model to determine whether it needs to be corrected, and then perform corresponding operations to provide accurate data for subsequent welding work.

[0186] Example 9

[0187] On the basis of Example 3, the intelligent welding method based on PLC intelligent control uses preset welding point samples to train each image contour to obtain a plurality of training contours, including:

[0188] Establishing an image screening standard using the preset welding sample;

[0189] Using the image screening standard to screen the image contour to obtain a first image contour;

[0190] Comparing the first image contour with a preset welding point sample to obtain corresponding contour outliers;

[0191] Performing mutual adaptation training on the contour outliers and the preset welding point samples to obtain a second image contour;

[0192] A second image contour consistent with the preset welding point sample is extracted to obtain a plurality of training contours.

[0193] In this example, the image screening criteria represent a screening mechanism for screening image profiles similar to a preset welding sample;

[0194] In this example, the contour outlier points represent contour points where the contour of the first image is inconsistent with the preset welding points;

[0195] In this example, the first image profile represents an image profile similar to a preset welding sample;

[0196] In this example, the second image contour represents the first image contour after being trained with preset welding point samples.

[0197] The working principle and beneficial effects of the above technical solution are as follows: In order to further strengthen the establishment of the training contour and shorten the establishment time, an image screening standard is first established using a preset welding sample to screen the image contour, and the image contour retained by the screening is then trained to adapt to the preset welding point. Finally, a qualified image contour is selected based on the preset contour sample, thereby obtaining the training contour, shortening the training time, and improving the welding efficiency.

[0198] Example 10

[0199] On the basis of Example 9, the intelligent welding method based on PLC intelligent control uses the preset welding sample to establish an image screening standard, including:

[0200] Performing corrosion treatment and expansion treatment on the preset welding samples respectively to obtain corresponding corrosion samples and expansion samples;

[0201] Acquire a first image region corresponding to the eroded sample and a second image region corresponding to the expanded sample;

[0202] An image screening criterion is established based on the first image region and the second image region.

[0203] The working principle and beneficial effects of the above technical solution are as follows: by performing corrosion treatment and expansion treatment on the preset welding sample, the sample morphology of the preset welding sample under different conditions can be obtained, and then an image screening standard is established to improve the screening accuracy.

[0204] Embodiment 11

[0205] On the basis of Example 3, the intelligent welding method based on PLC intelligent control divides the live image into a plurality of image areas based on the pixel distribution information, including:

[0206] Analyzing the pixel distribution information to obtain several types of pixel information contained in the live image;

[0207] The number of pixel points corresponding to each pixel information is obtained respectively, and the texture feature corresponding to each pixel information is calculated according to formula (1);

[0208]

[0209] Where W represents the texture feature corresponding to the pixel information, p i represents the pixel value corresponding to the i-th pixel in the pixel information, p z represents the pixel value corresponding to the central pixel in the pixel information, p i-1 represents the pixel value corresponding to the i-1th pixel in the pixel information, and n represents the number of pixels in the pixel information;

[0210] According to the calculation result of formula (1), the pixel information of the texture feature within the preset feature range is recorded as the first pixel information, and the pixel information of the texture feature outside the preset feature range is recorded as the second pixel information;

[0211] Calculate the pixel distribution characteristics corresponding to each pixel information according to formula (2);

[0212]

[0213] Wherein, F represents the pixel distribution feature corresponding to the pixel information, and h represents the preset feature range;

[0214] According to the calculation result of formula (2), the pixel points consistent with the pixel distribution characteristics are marked in the live image, and divided to obtain a plurality of image regions.

[0215] The working principle and beneficial effects of the above technical solution are as follows: In order to make the acquired second welding point closer to the actual welding point, the pixel distribution information of the live image is analyzed, and the texture features of the live image are first obtained using the formula, and the pixel information is divided into two categories according to the texture, and the next step is to analyze the pixel distribution features. Finally, the corresponding image area can be obtained on the live image to facilitate subsequent welding.

[0216] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent welding method based on PLC intelligent control, characterized in that: include: Step 1: using a light-touch positioning device to collect a plurality of first welding points on the object to be welded; Step 2: Collecting a live image of the object to be welded, preprocessing the live image, and obtaining a plurality of second welding points; Step 3: Analyze the overlap relationship between the first welding point and the second welding point using PLC, and obtain a plurality of target welding points on the object to be welded based on the overlap relationship; Step 4: Obtain the welding process corresponding to each target welding point and execute it; The PLC is used to analyze the coincidence relationship between the first welding point and the second welding point, including: Editing the first welding points by using PLC to obtain a first distribution layout of the first welding points, obtaining a plurality of first distribution points, and making first marks on the first distribution points in a rectangular coordinate system; Editing the second welding points by using PLC to obtain a second distribution layout of the second welding points, obtaining a plurality of second distribution points, and performing a second marking on the second distribution points in a rectangular coordinate system; Acquire a coincidence mark between the first mark and the second mark, and establish a coincidence relationship between the first welding point and the second welding point; Editing the first welding points by using PLC to obtain a first distribution layout of the first welding points includes: Obtain the reflected light corresponding to each first welding point, obtain the waveform width corresponding to each reflected light, and simultaneously obtain each reflected light, obtain the aperture range generated by each reflected light, and establish a waveform information list; Inputting each first welding point into a preset space respectively to obtain an initial welding point model; Using PLC to edit the waveform information list, obtaining a light model corresponding to each reflected light; Superimposing the light model and the initial welding point model; Training the initial welding point model based on the light model to obtain a target welding point model; Planarizing the target welding point model to obtain a first distribution layout of the first welding points; Training the initial welding point model based on the light model includes: Analyze each virtual reflected light in the light model respectively to obtain a light starting point corresponding to each virtual light; Marking the starting point of the light on the light model to obtain a plurality of reflection points; Analyze each virtual reflected light in the light model respectively to obtain the light beam height corresponding to each virtual light; Obtaining a corresponding reflection height based on the light beam height, and establishing a virtual reflector of a corresponding height at the reflection point; Analyzing the initial welding point model to obtain a plurality of virtual welding points; It is determined whether the virtual welding point and the virtual reflector at the corresponding position are at the same height. If not, the height of the virtual welding point is corrected, all trained virtual welding points are obtained, and a target welding point model is established.

2. The intelligent welding method based on PLC intelligent control as claimed in claim 1, characterized in that: The optical touch positioning device is used to collect a plurality of first welding points on the object to be welded, including: Scanning the object to be welded using a first preset light beam to obtain a scanned image; Based on the scanned image, draw the appearance structure of the object to be welded, and obtain all key points in the appearance structure; Scan each key point respectively using the second preset light to obtain the reflected light corresponding to each key point; Analyze the light intensity corresponding to each reflected light, and extract the key points where the light intensity of the reflected light is within a preset light intensity range; Each key point is marked on the scanned image and recorded as the first welding point.

3. The intelligent welding method based on PLC intelligent control as claimed in claim 1, characterized in that: Collecting a live image of the object to be welded, preprocessing the live image, and obtaining a plurality of second welding points, including: Collecting and analyzing the live image of the object to be welded to obtain pixel distribution information of the live image; Dividing the live image into a plurality of image regions based on the pixel distribution information, and respectively obtaining image contours of each pair of image regions; Using preset welding point samples to train each of the image contours to obtain a plurality of training contours; Comparing each training contour with a preset welding point contour respectively, and extracting a training contour consistent with the preset welding point; Each training contour is marked on the live image as a second welding point.

4. The intelligent welding method based on PLC intelligent control as claimed in claim 1, characterized in that: Obtain the welding process corresponding to each target welding point and execute it, including: Acquire the physical parameters of the object to be welded on the live image, obtain the welding scheme of the object to be welded in a preset database, and record the welding points with the same welding process as a process class according to the welding scheme; Obtaining process parameters corresponding to each process type respectively, and establishing corresponding welding parameters for each process type respectively in combination with the physical parameters; The welding scheme is analyzed to obtain a welding process, a target welding point corresponding to the welding process is obtained, and welding work is performed in combination with corresponding welding parameters.

5. The intelligent welding method based on PLC intelligent control as claimed in claim 2, characterized in that: Drawing the appearance structure of the object to be welded based on the scanned image includes: Obtaining the pixel value corresponding to each pixel point on the scanned image and establishing a pixel matrix; Obtaining a membership value corresponding to each matrix element in the pixel matrix, dividing the matrix elements with consistent membership values ​​into one category, and obtaining a segmentation result; Mapping the segmentation result onto the scanned image, dividing the scanned image into a plurality of image regions; Obtain the regional grayscale value corresponding to each image region respectively, and obtain the structural height corresponding to each image region; The regional features corresponding to each image region are established, and the appearance structure of the object to be welded is drawn in combination with the structural height corresponding to each image region.

6. The intelligent welding method based on PLC intelligent control as claimed in claim 3 is characterized in that: Each of the image contours is trained using a preset welding point sample to obtain a number of training contours, including: Establishing an image screening standard using the preset welding sample; Using the image screening standard to screen the image contour to obtain a first image contour; Comparing the first image contour with a preset welding point sample to obtain corresponding contour outliers; Performing mutual adaptation training on the contour outliers and the preset welding point samples to obtain a second image contour; A second image contour consistent with the preset welding point sample is extracted to obtain a plurality of training contours.

7. The intelligent welding method based on PLC intelligent control as claimed in claim 6, characterized in that: Using the preset welding sample, an image screening criteria is established, including: Performing corrosion treatment and expansion treatment on the preset welding samples respectively to obtain corresponding corrosion samples and expansion samples; Acquire a first image region corresponding to the eroded sample and a second image region corresponding to the expanded sample; An image screening criterion is established based on the first image region and the second image region.

Images