CN117444404B - Intelligent positioning method and system for laser welding - Google Patents
Intelligent positioning method and system for laser welding Download PDFInfo
- Publication number
- CN117444404B CN117444404B CN202311544410.8A CN202311544410A CN117444404B CN 117444404 B CN117444404 B CN 117444404B CN 202311544410 A CN202311544410 A CN 202311544410A CN 117444404 B CN117444404 B CN 117444404B
- Authority
- CN
- China
- Prior art keywords
- welding
- laser
- edge line
- suspected
- weld
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003466 welding Methods 0.000 title claims abstract description 356
- 238000000034 method Methods 0.000 title claims abstract description 43
- 239000000779 smoke Substances 0.000 claims abstract description 36
- 238000005286 illumination Methods 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 6
- 238000003708 edge detection Methods 0.000 claims description 6
- 238000009825 accumulation Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 239000003517 fume Substances 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 13
- 238000007781 pre-processing Methods 0.000 abstract description 2
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 6
- 239000000463 material Substances 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 239000011324 bead Substances 0.000 description 3
- 230000004069 differentiation Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/20—Bonding
- B23K26/21—Bonding by welding
- B23K26/24—Seam welding
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/70—Auxiliary operations or equipment
- B23K26/702—Auxiliary equipment
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Plasma & Fusion (AREA)
- Mechanical Engineering (AREA)
- Laser Beam Processing (AREA)
Abstract
The invention relates to the technical field of intelligent welding, and provides an intelligent positioning method and system for laser welding, comprising the following steps: collecting an image of a workpiece to be welded, recording a welding track, preprocessing the image of the workpiece to be welded, acquiring the diameters of a suspected welding seam edge line, a laser welding area and a laser smoke area according to the image of the workpiece to be welded, calculating the similarity of the welding track and the welding track engagement of the suspected welding seam edge line, acquiring the welding seam edge line, calculating the laser interference gradient of the welding seam edge line, determining the actual value of the autoregressive term number, and predicting the position of the welding point in the laser welding area by using a time sequence prediction model to realize intelligent positioning of welding. The invention aims to solve the problem that the positioning of the welding point in the laser welding area is not real-time in the welding process.
Description
Technical Field
The invention relates to the technical field of intelligent welding, in particular to an intelligent positioning method and system for laser welding.
Background
The laser welding is a high-efficiency precise welding method which uses a laser beam with high energy density as a heat source, is one of important aspects of application of laser material processing technology, and is mainly used for welding thin-wall materials and low-speed welding, the welding process belongs to heat conduction type, namely laser radiation is used for heating the surface of a workpiece, surface heat is diffused to the inside through heat conduction, and the workpiece is melted by controlling parameters such as the width, energy, peak power, repetition frequency and the like of laser pulses, so that a specific molten pool is formed. The laser welding occupies an important position in precision welding by virtue of the advantages of no welding pollution and no deformation, and is widely applied to the industries of automobile body steel plate welding, battery welding, sealing relay and the like.
In the laser welding process, the machine vision positioning is utilized to provide positioning parameters, so that accurate positioning is realized, the robot is guided to perform high-precision laser welding, the welding quality is effectively improved, continuous accurate operation is also facilitated, the working efficiency is improved, and the system cost is saved.
However, the workpiece to be welded is thermally deformed during the welding process, and arc light, smoke and the like are generated at the same time, so that the positioning of the welding point in the laser welding area is not real-time.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide an intelligent positioning method and system for laser welding, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for intelligent positioning in laser welding, including the steps of:
collecting an image of a workpiece to be welded, and recording a welding track;
according to the image of the workpiece to be welded, gray scale distinction of pixel points in the image of the workpiece to be welded and suspected weld edge lines are obtained;
calculating the similarity of the welding track of the suspected welding seam edge line according to the suspected welding seam edge line and the welding track;
according to the image of the workpiece to be welded, acquiring illumination components of pixel points in the image of the workpiece to be welded and a laser binary image;
acquiring the diameter of a laser welding area and a laser smoke area according to the laser binary image;
calculating the welding track engagement degree of the suspected welding seam edge line according to the welding track similarity of the suspected welding seam edge line and the diameter of the laser smoke area;
the suspected weld edge line with the largest welding track engagement degree is marked as a weld edge line;
calculating the laser interference gradient of the weld edge line according to the gray scale distinction and illumination components of all pixel points on the weld edge line;
determining the actual value of the autoregressive item number according to the laser interference gradient of the weld edge line;
and according to the welding track, the positions of the pixel points on the edge line of the welding seam, the actual value of the regression term number and the diameter of the laser smoke area, realizing intelligent positioning of welding.
Preferably, the obtaining the gray scale distinction of the pixel point in the image of the workpiece to be welded and the suspected weld edge line according to the image of the workpiece to be welded includes:
the average value of gray values of all pixel points in the image of the workpiece to be welded is recorded as a welding gray average value;
the absolute value of the difference value between the gray value of the pixel point in the image of the workpiece to be welded and the welding gray average value is recorded as the gray scale distinction of the pixel point in the image of the workpiece to be welded;
acquiring edge information of an image of a workpiece to be welded by using an edge detection operator to obtain a binary image to be welded;
marking a non-closed edge line with the length larger than the preset minimum weld length in the binary image to be welded as an alternative weld edge line;
performing secondary polynomial fitting on the edge line of the alternative welding seam to obtain the fitting goodness of the edge line of the alternative welding seam;
the sum of gray scale distinguishing indexes of all pixel points on the edge line of the alternative welding seam is recorded as a gray scale distinguishing index of the edge line of the alternative welding seam;
the product of the goodness of fit of the candidate weld edge line and the gray scale distinguishing index is recorded as a first product;
the product of the number of pixel points on the edge line of the alternative welding seam and the first product is recorded as a second product;
the normalized value of the second product is recorded as a weld morphology feature index of the candidate weld edge line;
and marking the alternative weld edge line with the weld morphology characteristic index larger than the preset weld characteristic threshold as a suspected weld edge line.
Preferably, the specific calculation method of the welding track similarity of the suspected welding seam edge line includes:
performing secondary polynomial fitting on the suspected weld edge line to obtain fitting parameters and fitting goodness of the suspected weld edge line, wherein the fitting parameters of the suspected weld edge line comprise a quadratic term coefficient, a first term coefficient and a constant term;
acquiring fitting parameters and fitting goodness of a welding track;
and according to the fitting parameters and the fitting goodness of the suspected weld edge line, the fitting parameters and the fitting goodness of the welding track, the similarity of the welding track of the suspected weld edge line is expressed as follows:
wherein HGS is the similarity of welding tracks of suspected weld edge lines,fitting goodness for suspected weld edge line, +.>For the goodness of fit of the welding track, +.>Is the quadratic coefficient of the suspected weld edge line, < +.>A first order coefficient for a suspected weld edge line, < +.>Constant term for suspected weld edge line, +.>Is the quadratic coefficient of the welding track, +.>For the first order coefficient of the welding track, +.>Is a constant term of the welding track.
Preferably, the obtaining the illumination component of the pixel point in the image of the workpiece to be welded and the laser binary image according to the image of the workpiece to be welded includes:
according to the size of a preset Gaussian kernel, obtaining illumination components of all pixel points in an image of a workpiece to be welded, and obtaining a welding laser image;
and acquiring edge information in the welding laser image by using an edge detection operator to obtain a laser binary image.
Preferably, the obtaining the diameter of the laser welding area and the diameter of the laser smoke area according to the laser binary image includes:
marking a closed edge in the laser binary image as a suspected laser welding area;
marking the center point of the minimum circumscribed rectangle of the suspected laser welding area as the position of the suspected laser welding area;
the suspected laser welding area closest to the tail end of the welding track is marked as a laser welding area;
the diagonal length of the smallest bounding rectangle of the laser welded area is noted as the laser fume area diameter.
Preferably, the calculating the welding track engagement degree of the suspected welding seam edge line according to the welding track similarity of the suspected welding seam edge line and the laser smoke area diameter includes:
the minimum value of the distance between the two endpoints of the suspected weld edge line and the two endpoints of the welding track is recorded as the track neighbor distance of the suspected weld edge line;
the absolute value of the difference value between the track adjacent distance of the suspected weld edge line and the diameter of the laser smoke area is recorded as the laser area coincidence degree of the suspected weld edge line;
and (3) marking the product of the laser region coincidence degree of the suspected weld edge line and the welding track similarity as the welding track engagement degree of the suspected weld edge line.
Preferably, the calculating the laser interference gradient of the weld edge line according to the gray scale differentiation and the illumination component of all the pixel points on the weld edge line includes:
the difference value of gray scale distinction between the pixel point and the previous pixel point on the weld edge line is recorded as the gray scale gradient of the pixel point;
recording the difference value of the illumination components of the pixel point on the weld edge line and the previous pixel point as the laser gradient of the pixel point;
the sum of gray gradient of all pixel points on the edge line of the welding line is recorded as a first accumulation sum;
the sum of the laser gradient of all pixel points on the edge line of the welding line is recorded as a second accumulated sum;
and marking the sum of the first accumulated sum and the second accumulated sum as the laser interference gradient of the weld edge line.
Preferably, the determining the actual value of the autoregressive term according to the laser interference gradient of the weld edge line includes:
marking a normalized value of the laser interference gradient of the weld edge line as a laser interference normalized index;
recording the product of the laser interference normalization index and the adjustment value of the preset autoregressive term number as a third product;
and recording the downward rounded value of the sum of the initial value of the preset autoregressive term number and the third product as the actual value of the autoregressive term number.
Preferably, the intelligent positioning of welding is realized according to the welding track, the position of the pixel point on the edge line of the welding seam, the actual value of the regression term number and the diameter of the laser smoke area, and the method comprises the following steps:
the actual value of the autoregressive item number is recorded as a first number;
marking the downward rounding value of the laser smoke area diameter as a second number;
constructing a track prediction sequence according to the sequence from far to near of a first number of pixel points closest to a laser welding area on a welding track;
constructing a weld prediction sequence of a first number of pixel points closest to the laser welding area on the weld edge line according to the sequence from far to near;
according to the abscissa of the pixel points in the track prediction sequence, predicting the abscissa of a second number of subsequent pixel points by using a time sequence prediction model;
predicting the ordinate of the subsequent second number of pixel points by using a time sequence prediction model according to the ordinate of the pixel points in the track prediction sequence;
acquiring the abscissa and the ordinate of a second number of subsequent pixel points of the weld joint prediction sequence;
and according to the track prediction sequence and the weld joint prediction sequence, the abscissa of the welding point in the laser welding area is expressed as follows:
wherein x is j The abscissa of the ith weld point in the laser weld zone,for the abscissa of the p+j-th pixel point in the track prediction sequence, +.>For the abscissa of the p+ (dh-j) th pixel point in the weld prediction sequence, p is a first number, dh is a second number;
acquiring the ordinate of a welding point in a laser welding area;
and moving the laser beams to the welding points in sequence according to the abscissa and the ordinate of all the welding points in the laser welding area, processing and welding the welding points by using the laser beams, and checking the welding points after the welding is finished, so that the welding quality meets the requirements.
In a second aspect, an embodiment of the present invention further provides a laser welding intelligent positioning system, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the steps of any one of the methods described above when executing the computer program.
The invention has at least the following beneficial effects: firstly, according to a to-be-welded workpiece image, obtaining a suspected welding seam edge line, calculating the welding track similarity of the suspected welding seam edge line, and then according to the welding track similarity of the suspected welding seam edge line and the laser smoke area diameter, calculating the welding track engagement of the suspected welding seam edge line, synthesizing the similarity and consistency of the suspected welding seam edge line and the welding track morphological characteristics, obtaining the welding seam edge line, and improving the reliability of obtaining the welding seam edge line; according to the illumination component and gray scale distinction degree of each pixel point on the weld edge line, calculating the laser interference gradient degree of the weld edge line, determining the actual value of the autoregressive item number, and improving the efficiency of positioning the welding point in the laser welding area; according to the actual values of the welding track, the positions of the pixel points on the edge line of the welding line and the autoregressive item number, the ARIMA time sequence prediction model is used for predicting the positions of the welding points in the laser welding area, so that the intelligent positioning of welding is realized, and the problem that the positioning of the welding points in the laser welding area is not real-time in the welding process is solved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of a method for intelligent positioning for laser welding according to an embodiment of the present invention;
fig. 2 is a schematic view of laser welding area prediction.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of specific implementation, structure, characteristics and effects of the intelligent positioning method and system for laser welding according to the invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a laser welding intelligent positioning method and a system specific scheme by combining the drawings.
Referring to fig. 1, a flowchart of steps of a laser welding intelligent positioning method according to an embodiment of the invention is shown, the method includes the following steps:
and S001, collecting an image of the workpiece to be welded, recording a welding track, and preprocessing the image of the workpiece to be welded.
The main purpose of this embodiment is to locate the welding point during the welding process of the workpiece to be welded, so as to assist the laser beam to weld and process the welding point, therefore, it is necessary to collect the image of the workpiece to be welded first and record the welding track. And placing an industrial camera above the station, shooting the workpiece on the whole station in real time while welding the workpiece, acquiring an image of the workpiece to be welded, and recording a welding track. In the process of welding a welding point by using a laser beam, arc light, smoke and the like can be generated, so that the acquired image of the workpiece to be welded is not clear enough, in order to improve the accuracy of subsequent analysis, the image of the workpiece to be welded is subjected to grey-scale treatment, and then the image of the workpiece to be welded is enhanced by using an NLM non-local mean filtering algorithm.
So far, the image of the workpiece to be welded and the welding track are obtained.
Step S002, according to the image of the workpiece to be welded, obtaining a suspected weld edge line, and calculating the gray scale distinction degree of the pixel points in the image of the workpiece to be welded and the welding track similarity of the suspected weld edge line.
It should be noted that welding is a process in which two or more kinds of the same or different materials are integrally connected by bonding and diffusion between atoms or molecules, so that the region to be welded is at least located at the junction of the two materials, and presents a weld seam with different colors from the regions on both sides.
Specifically, the average value of the gray values of all the pixels in the image of the workpiece to be welded is recorded as a welding gray average value, and the absolute value of the difference between the gray values of the pixels in the image of the workpiece to be welded and the welding gray average value is recorded as the gray distinction degree of the pixels in the image of the workpiece to be welded.
Acquiring edge information of an image of a workpiece to be welded by using a canny edge detection operator to obtain a binary image to be welded, and marking a non-closed edge line with the length larger than a preset minimum weld length in the binary image to be welded as an alternative weld edge line, wherein the minimum weld length l is the minimum weld length l min The empirical value is 10, then the least square method is used for carrying out the quadratic polynomial fitting on the edge line of the candidate welding seam, and the fitting curve of the edge line of the candidate welding seam is determined, namely the estimationThe quadratic term coefficient->Coefficient of primary term->And constant item->Wherein x is the abscissa of the pixel point on the edge line of the candidate weld, y is the ordinate of the pixel point on the edge line of the candidate weld, and then the fitting goodness of the edge line of the candidate weld is obtained according to the fitting curve of the edge line of the candidate weld and the coordinates of the pixel point on the edge line of the candidate weld, and the weld morphology feature indexes of the edge line of the candidate weld are expressed as follows according to the gray scale distinction degree and the fitting goodness of the pixel point on the edge line of the candidate weld and the number of the pixel points on the edge line of the candidate weld:
wherein HET is the weld morphology characteristic index of the candidate weld edge line, exp () is an exponential function based on natural constant, and n is a non-closed edgeThe number of pixel points on the edge line, R 2 Is the goodness of fit of the non-closed edge line, df (x i ) Is the gray scale differentiation of the ith pixel point on the non-closed edge line.
It should be noted that, the gray scale distinguishing degree of the pixel points represents the difference degree of the gray scale values of the pixel points and the image of the workpiece to be welded, when the gray scale value of the pixel points on the edge line of the candidate welding line is larger than the gray scale value of the image of the workpiece to be welded, the more likely the pixel points are the welding line which does not belong to the same material as the workpiece to be welded, and the larger the welding line morphology feature index value is; the number of pixel points on the edge line of the alternative welding seam represents the length of the edge line of the alternative welding seam, and when the length of the edge line of the alternative welding seam is longer, the more likely the welding seam is, and the larger the index value of the morphology feature of the welding seam is; the goodness of fit refers to the goodness of fit of a regression curve to an observed value, the maximum value is 1, and when the goodness of fit of the edge line of the alternative welding seam is closer to 1, the better the goodness of fit of the quadratic polynomial to the edge line is, the higher the smoothness degree of the edge line of the alternative welding seam is, the more likely the welding seam is, and the greater the characteristic index of the welding seam morphology is.
It should be further noted that, during the welding process, intense arc light, smoke and the like may be generated, which results in inaccurate positioning of the welding position, but the welding seam is required to be smooth during welding, and welding defects such as air hole slag inclusion and the like must not exist, the same welding seam should be smooth and consecutive, and the welding seam area is selected from the suspected welding seam edge line according to the consistency characteristics of the welding seam, so as to position the welding seam.
Further, a weld characteristic threshold t is set f The weld joint morphology characteristic index is larger than the weld joint characteristic threshold t f The candidate weld edge line of (1) is marked as a suspected weld edge line, and the weld characteristic threshold t f The empirical value was 0.7. Performing quadratic polynomial fitting on the welding track by using a least square method, determining a fitting curve of the welding track, and expressing the similarity HGS of the welding track of the suspected welding seam edge line as follows according to the fitting curve and the fitting goodness of the suspected welding seam edge line and the welding track:
wherein HGS is the similarity of welding tracks of suspected weld edge lines,fitting goodness for suspected weld edge line, +.>For the goodness of fit of the welding track, +.>Is the quadratic coefficient of the suspected weld edge line, < +.>A first order coefficient for a suspected weld edge line, < +.>Constant term for suspected weld edge line, +.>Is the quadratic coefficient of the welding track, +.>For the first order coefficient of the welding track, +.>Is a constant term of the welding track.
It should be noted that, when the goodness of fit, the quadratic term coefficient, the first term coefficient, and the constant term of the fitted curve of the suspected bead edge line and the welding track are closer, the more likely they are on one bead, the greater the welding track similarity value of the suspected bead edge line.
And obtaining gray scale distinction of pixel points in the image of the workpiece to be welded and the similarity of welding tracks of the suspected weld edge lines.
Step S003, obtaining the diameters of a laser welding area and a laser smoke area according to the image of the workpiece to be welded, and calculating the illumination component of the pixel point in the image of the workpiece to be welded and the welding track engagement degree of the edge line of the suspected weld according to the welding track similarity of the edge line of the suspected weld and the diameter of the laser smoke area.
It should be noted that, because arc light, smoke and the like may be generated when a workpiece to be welded is welded by using a laser beam, the positioning of the welding point may be adversely affected, based on the Retinex theory, the illumination component may be approximately represented by convolution of the original image and the gaussian kernel, and the area affected by the laser beam is obtained according to the illumination component of all pixel points in the image of the workpiece to be welded.
Specifically, setting the size of a Gaussian kernel to be 5×5, calculating the illumination component of each pixel point in an image of a workpiece to be welded based on the Retinex theory, obtaining a welding laser image, obtaining edge information in the welding laser image by using a canny edge detection operator, obtaining a laser binary image, marking a closed edge in the laser binary image as a suspected laser welding area, marking the center point of the minimum circumscribed rectangle of the suspected laser welding area as the position of the suspected laser welding area, marking the suspected laser welding area closest to the tail end of a welding track as the laser welding area, and marking the diagonal length of the minimum circumscribed rectangle of the laser welding area as the diameter of the laser smoke area.
It should be further noted that, there may be multiple suspected weld edge lines in the image of the workpiece to be welded, during the welding process, the same weld should be welded along the same track as much as possible, and when the weld is positioned, the suspected weld edge lines with high similarity of the welding track and located around the laser welding area should be selected first, and then the position of the welding point in the laser welding area should be obtained.
Specifically, two end points of the edge line of the suspected weld are respectively marked as S 1 、E 1 The two end points of the welding track are respectively marked as S 2 、E 2 According to the end points of the suspected weld edge line and the welding track, the similarity of the welding track of the suspected weld edge line and the laser smoke area diameter, the suspected weld edge is obtainedThe weld trace engagement HXG of the wire is shown below:
HXG＝HGS×|DOm-ds|
DOm＝min{d(S 1 ，S 2 )，d(S 1 ，E 2 )，d(E 1 ，S 2 )，d(E 1 ，E 2 )}
wherein HXG is the welding track engagement degree of the suspected weld edge line, HGS is the welding track similarity degree of the suspected weld edge line, DOm is the track adjacent distance of the suspected weld edge line, ds is the laser smoke area diameter, min { } is the minimum function, and d (,) represents the Euclidean distance between the two end points.
When the track close distance of the suspected weld edge line is smaller than the diameter difference of the laser smoke area, the suspected weld edge line is more likely to be positioned on the same weld which is interrupted by the influence of laser and smoke, and the welding track engagement value is larger; when the similarity of the welding track of the suspected welding seam edge line is higher, the suspected welding seam edge line is more likely to be positioned on the same welding seam as the welding track, and the welding track engagement value is larger.
So far, the illumination component of the pixel point in the image of the workpiece to be welded and the welding track engagement degree of the suspected welding seam edge line are obtained.
Step S004, obtaining the weld edge line according to the welding track connection degree of all the suspected weld edge lines, calculating the laser interference gradient degree of the weld edge line according to the gray scale distinction degree and the illumination component of the pixel points in the image of the workpiece to be welded, and determining the actual value of the autoregressive term number.
It should be noted that, when the degree of influence of arc light, smoke, or the like generated when welding a workpiece to be welded using a laser beam is higher, the adverse effect of positioning the welding point is greater, and more known information is required as a basis when acquiring the position of the welding point in the laser welding region using a predictive algorithm.
Specifically, a suspected weld edge line with the largest welding track engagement degree is marked as a weld edge line, and the laser interference gradient of the weld edge line is expressed as follows according to the light component and gray scale distinction of each pixel point on the weld edge line:
wherein JRG is the laser interference gradient of the weld edge line, df (x) k ) For the gradation discrimination of the kth pixel point on the weld edge line, df (x k+1 ) L (x) is the gray scale differentiation of the (k+1) th pixel point on the weld edge line k ) L (x) is the illumination component of the kth pixel point on the weld edge line k+1 ) N is the illumination component of the (k+1) th pixel point on the edge line of the welding line f The number of pixel points on the edge line of the welding line is the number of the pixel points.
When the illumination component of the pixel point on the weld edge line is gradually reduced and the gray scale distinction degree is gradually increased, the higher the influence degree of laser is, the larger the laser interference gradient value of the weld edge line is; when the illumination component and the gray scale distinction degree of the pixel points on the weld edge line are negligent, the lower the influence degree of laser is, the smaller the laser interference gradient value of the weld edge line is.
When the ARIMA time sequence prediction model is used for predicting the welding point positions of the laser welding area, the autoregressive term number p represents the number of the welding point positions participating in prediction, and the greater the autoregressive term number is, the higher the reliability is when the prediction is performed according to more welding point positions; the smaller the autoregressive term value is, the more prediction is performed according to the less historical data, and the higher the prediction efficiency is.
Specifically, an initial value p of the autoregressive term number is set 0 Adjustment value p r The actual value p of the autoregressive term number can be expressed as follows:
p＝Floor(p 0 +[1-exp(-JRG)]×p r )
wherein p is the autoregressive term number of the ARIMA time sequence prediction model, floor () is a downward rounding function, exp () is an exponential function based on a natural constant, and JRG is the laser interference gradient of a weld edge line.
When the laser interference gradient of the weld edge line is higher, the influence degree of laser is higher, the autoregressive term number p is larger, prediction is carried out according to more welding point positions, and the accuracy of positioning the welding point in the laser welding area is improved.
Thus, the actual value of the autoregressive term number is obtained.
And S005, predicting the position of the welding point in the laser welding area by using an ARIMA time sequence prediction model according to the welding track, the positions of the pixel points on the edge line of the welding line, the actual value of the autoregressive item number and the diameter of the laser smoke area, so as to realize intelligent positioning of welding.
Specifically, the empirical value of the differential order d is 2, the empirical value of the moving average term q is 3, p pixel points closest to the laser welding area on the welding track are constructed into a track prediction sequence according to the sequence from far to near, p pixel points closest to the laser welding area on the welding edge line are constructed into a welding seam prediction sequence according to the sequence from far to near, wherein p is the actual value of the autoregressive term, the prediction schematic diagram of the laser welding area is shown in fig. 2, the track prediction sequence is E, D, C, B, A, and the welding seam prediction sequence is J, I, H, G, F.
Further, the downward rounding value of the diameter of the laser smoke area is recorded as the number of welding points of the laser smoke area, according to the abscissa of the pixel points in the track prediction sequence, the abscissa of the subsequent dh pixel points is predicted by using an ARIMA time sequence prediction model, and according to the ordinate of the subsequent dh pixel points in the track prediction sequence and the abscissa and the ordinate of the subsequent dh pixel points in the weld prediction sequence in the same method, dh is the number of welding points of the laser smoke area, and the average value of the two sequence prediction results is the abscissa and the ordinate of the welding point in the finally predicted laser welding area, namely the abscissa of the jth welding point in the laser welding area Ordinate->Wherein (1)>For the abscissa of the p+j-th pixel point in the track prediction sequence, +.>For the abscissa of the p+ (dh-j) th pixel in the weld prediction sequence, { circumflex }>For the ordinate of the p+j pixel point in the track prediction sequence, +.>And (3) taking the ordinate of the p+ (dh-j) th pixel point in the weld joint prediction sequence, wherein p is the actual value of the autoregressive term number.
Further, according to the abscissa and the ordinate of all welding points in the laser welding area, the laser beams are sequentially moved to the welding points, the welding points are processed and welded by the laser beams, and after the welding is finished, the welding points are inspected, so that the welding quality meets the requirements.
So far, the positions of all welding points in the laser welding area are obtained, and intelligent positioning of welding is realized.
Based on the same inventive concept as the above method, the present embodiment further provides a laser welding intelligent positioning system, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor executes the computer program to implement the steps of any one of the above-mentioned laser welding intelligent positioning methods.
In summary, the problem that the positioning of the welding point in the laser welding area does not have real-time performance in the welding process is solved, the method includes the steps of analyzing an image of a workpiece to be welded to obtain a suspected welding line, calculating the similarity of welding tracks of the suspected welding line, obtaining the diameters of a laser welding area and a laser smoke area, calculating the welding track connection degree of the suspected welding line according to the welding track similarity of the suspected welding line and the diameters of the laser smoke area, obtaining the welding line according to the welding track connection degree of all the suspected welding line, calculating the laser interference gradient of the welding line, determining the actual value of an autoregressive term, and predicting the position of the welding point in the laser welding area by using an ARIMA time sequence prediction model according to the actual value of the welding track, the positions of pixels on the welding line and the autoregressive term number.
It should be noted that: the sequence of the embodiment is only for description, and does not represent the advantages and disadvantages of the embodiment. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. The intelligent positioning method for laser welding is characterized by comprising the following steps of:
collecting an image of a workpiece to be welded, and recording a welding track;
according to the image of the workpiece to be welded, gray scale distinction of pixel points in the image of the workpiece to be welded and suspected weld edge lines are obtained;
calculating the similarity of the welding track of the suspected welding seam edge line according to the suspected welding seam edge line and the welding track;
according to the image of the workpiece to be welded, acquiring illumination components of pixel points in the image of the workpiece to be welded and a laser binary image;
acquiring the diameter of a laser welding area and a laser smoke area according to the laser binary image;
calculating the welding track engagement degree of the suspected welding seam edge line according to the welding track similarity of the suspected welding seam edge line and the diameter of the laser smoke area;
the suspected weld edge line with the largest welding track engagement degree is marked as a weld edge line;
calculating the laser interference gradient of the weld edge line according to the gray scale distinction and illumination components of all pixel points on the weld edge line;
determining the actual value of the autoregressive item number according to the laser interference gradient of the weld edge line;
and according to the welding track, the positions of the pixel points on the edge line of the welding seam, the actual value of the autoregressive item number and the diameter of the laser smoke area, realizing intelligent positioning of welding.
2. The intelligent positioning method for laser welding according to claim 1, wherein the step of obtaining the gray scale division of the pixel point in the image of the workpiece to be welded and the suspected weld edge line according to the image of the workpiece to be welded comprises the steps of:
the average value of gray values of all pixel points in the image of the workpiece to be welded is recorded as a welding gray average value;
the absolute value of the difference value between the gray value of the pixel point in the image of the workpiece to be welded and the welding gray average value is recorded as the gray scale distinction of the pixel point in the image of the workpiece to be welded;
acquiring edge information of an image of a workpiece to be welded by using an edge detection operator to obtain a binary image to be welded;
marking a non-closed edge line with the length larger than the preset minimum weld length in the binary image to be welded as an alternative weld edge line;
performing secondary polynomial fitting on the edge line of the alternative welding seam to obtain the fitting goodness of the edge line of the alternative welding seam;
the sum of gray scale distinguishing indexes of all pixel points on the edge line of the alternative welding seam is recorded as a gray scale distinguishing index of the edge line of the alternative welding seam;
the product of the goodness of fit of the candidate weld edge line and the gray scale distinguishing index is recorded as a first product;
the product of the number of pixel points on the edge line of the alternative welding seam and the first product is recorded as a second product;
the normalized value of the second product is recorded as a weld morphology feature index of the candidate weld edge line;
and marking the alternative weld edge line with the weld morphology characteristic index larger than the preset weld characteristic threshold as a suspected weld edge line.
3. The intelligent positioning method for laser welding according to claim 1, wherein the specific calculation method for the welding track similarity of the suspected weld edge line comprises the following steps:
performing secondary polynomial fitting on the suspected weld edge line to obtain fitting parameters and fitting goodness of the suspected weld edge line, wherein the fitting parameters of the suspected weld edge line comprise a quadratic term coefficient, a first term coefficient and a constant term;
acquiring fitting parameters and fitting goodness of a welding track;
and according to the fitting parameters and the fitting goodness of the suspected weld edge line, the fitting parameters and the fitting goodness of the welding track, the similarity of the welding track of the suspected weld edge line is expressed as follows:
wherein HGS is the similarity of welding tracks of suspected weld edge lines,fitting goodness for suspected weld edge line, +.>For the goodness of fit of the welding track, +.>Is the quadratic coefficient of the suspected weld edge line, < +.>A first order coefficient for a suspected weld edge line, < +.>Constant term for suspected weld edge line, +.>Is the quadratic coefficient of the welding track, +.>For the first order coefficient of the welding track, +.>Is a constant term of the welding track.
4. The intelligent positioning method for laser welding according to claim 1, wherein the obtaining the illumination component of the pixel point in the image of the workpiece to be welded and the laser binary image according to the image of the workpiece to be welded comprises:
according to the size of a preset Gaussian kernel, obtaining illumination components of all pixel points in an image of a workpiece to be welded, and obtaining a welding laser image;
and acquiring edge information in the welding laser image by using an edge detection operator to obtain a laser binary image.
5. The intelligent positioning method for laser welding according to claim 1, wherein the step of obtaining the laser welding area and the laser smoke area diameter according to the laser binary image comprises the following steps:
marking a closed edge in the laser binary image as a suspected laser welding area;
marking the center point of the minimum circumscribed rectangle of the suspected laser welding area as the position of the suspected laser welding area;
the suspected laser welding area closest to the tail end of the welding track is marked as a laser welding area;
the diagonal length of the smallest bounding rectangle of the laser welded area is noted as the laser fume area diameter.
6. The intelligent positioning method for laser welding according to claim 1, wherein the calculating the welding track engagement degree of the suspected welding seam edge line according to the welding track similarity of the suspected welding seam edge line and the laser smoke area diameter comprises:
the minimum value of the distance between the two endpoints of the suspected weld edge line and the two endpoints of the welding track is recorded as the track neighbor distance of the suspected weld edge line;
the absolute value of the difference value between the track adjacent distance of the suspected weld edge line and the diameter of the laser smoke area is recorded as the laser area coincidence degree of the suspected weld edge line;
and (3) marking the product of the laser region coincidence degree of the suspected weld edge line and the welding track similarity as the welding track engagement degree of the suspected weld edge line.
7. The intelligent positioning method for laser welding according to claim 1, wherein the calculating the laser interference gradient of the weld edge line according to the gray scale distinction and the illumination component of all the pixel points on the weld edge line comprises:
the difference value of gray scale distinction between the pixel point and the previous pixel point on the weld edge line is recorded as the gray scale gradient of the pixel point;
recording the difference value of the illumination components of the pixel point on the weld edge line and the previous pixel point as the laser gradient of the pixel point;
the sum of gray gradient of all pixel points on the edge line of the welding line is recorded as a first accumulation sum;
the sum of the laser gradient of all pixel points on the edge line of the welding line is recorded as a second accumulated sum;
and marking the sum of the first accumulated sum and the second accumulated sum as the laser interference gradient of the weld edge line.
8. The intelligent positioning method for laser welding according to claim 1, wherein the determining the actual value of the autoregressive term according to the laser interference gradient of the weld edge line comprises:
marking a normalized value of the laser interference gradient of the weld edge line as a laser interference normalized index;
recording the product of the laser interference normalization index and the adjustment value of the preset autoregressive term number as a third product;
and recording the downward rounded value of the sum of the initial value of the preset autoregressive term number and the third product as the actual value of the autoregressive term number.
9. The intelligent positioning method for laser welding according to claim 1, wherein the realizing the intelligent positioning for welding according to the welding track, the position of the pixel point on the edge line of the welding seam, the actual value of the autoregressive item number and the diameter of the laser smoke area comprises the following steps:
the actual value of the autoregressive item number is recorded as a first number;
marking the downward rounding value of the laser smoke area diameter as a second number;
constructing a track prediction sequence according to the sequence from far to near of a first number of pixel points closest to a laser welding area on a welding track;
constructing a weld prediction sequence of a first number of pixel points closest to the laser welding area on the weld edge line according to the sequence from far to near;
according to the abscissa of the pixel points in the track prediction sequence, predicting the abscissa of a second number of subsequent pixel points by using a time sequence prediction model;
predicting the ordinate of the subsequent second number of pixel points by using a time sequence prediction model according to the ordinate of the pixel points in the track prediction sequence;
acquiring the abscissa and the ordinate of a second number of subsequent pixel points of the weld joint prediction sequence;
and according to the track prediction sequence and the weld joint prediction sequence, the abscissa of the welding point in the laser welding area is expressed as follows:
wherein x is j The abscissa of the ith weld point in the laser weld zone,for the abscissa of the p+j-th pixel point in the track prediction sequence, +.>For the abscissa of the p+ (dh-j) th pixel point in the weld prediction sequence, p is a first number, dh is a second number;
acquiring the ordinate of a welding point in a laser welding area;
and moving the laser beams to the welding points in sequence according to the abscissa and the ordinate of all the welding points in the laser welding area, processing and welding the welding points by using the laser beams, and checking the welding points after the welding is finished, so that the welding quality meets the requirements.
10. A laser welding intelligent positioning system comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1-9 when the computer program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311544410.8A CN117444404B (en) | 2023-11-20 | 2023-11-20 | Intelligent positioning method and system for laser welding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311544410.8A CN117444404B (en) | 2023-11-20 | 2023-11-20 | Intelligent positioning method and system for laser welding |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117444404A CN117444404A (en) | 2024-01-26 |
| CN117444404B true CN117444404B (en) | 2024-03-29 |
Family
ID=89592851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311544410.8A Active CN117444404B (en) | 2023-11-20 | 2023-11-20 | Intelligent positioning method and system for laser welding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117444404B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103822594A (en) * | 2014-02-28 | 2014-05-28 | 华南理工大学 | Workpiece scanning imaging method based on laser sensor and robot |
| CN105678776A (en) * | 2016-01-11 | 2016-06-15 | 南京工业大学 | Laser vision sensor based weld image feature point extraction method |
| WO2020020113A1 (en) * | 2018-07-25 | 2020-01-30 | 同高先进制造科技（太仓）有限公司 | Active laser vision weld tracking system and weld position detecting method |
| CN112381783A (en) * | 2020-11-11 | 2021-02-19 | 湖南大学 | Weld track extraction method based on red line laser |
| CN114140439A (en) * | 2021-12-03 | 2022-03-04 | 厦门大学 | Laser welding seam feature point identification method and device based on deep learning |
| CN114523203A (en) * | 2022-03-13 | 2022-05-24 | 扬州沃盛车业制造有限公司 | Intelligent laser welding method and system |
-
2023
- 2023-11-20 CN CN202311544410.8A patent/CN117444404B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103822594A (en) * | 2014-02-28 | 2014-05-28 | 华南理工大学 | Workpiece scanning imaging method based on laser sensor and robot |
| CN105678776A (en) * | 2016-01-11 | 2016-06-15 | 南京工业大学 | Laser vision sensor based weld image feature point extraction method |
| WO2020020113A1 (en) * | 2018-07-25 | 2020-01-30 | 同高先进制造科技（太仓）有限公司 | Active laser vision weld tracking system and weld position detecting method |
| CN112381783A (en) * | 2020-11-11 | 2021-02-19 | 湖南大学 | Weld track extraction method based on red line laser |
| CN114140439A (en) * | 2021-12-03 | 2022-03-04 | 厦门大学 | Laser welding seam feature point identification method and device based on deep learning |
| CN114523203A (en) * | 2022-03-13 | 2022-05-24 | 扬州沃盛车业制造有限公司 | Intelligent laser welding method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117444404A (en) | 2024-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shah et al. | Butt welding joints recognition and location identification by using local thresholding | |
| WO2020038389A1 (en) | Welding seam negative defect recognition method | |
| Xu et al. | Visual sensing technologies in robotic welding: Recent research developments and future interests | |
| CN110717872B (en) | Method and system for extracting characteristic points of V-shaped welding seam image under laser-assisted positioning | |
| CN114821114B (en) | Groove cutting robot image processing method based on vision system | |
| CN108624880B (en) | A kind of Laser Cladding Quality intelligence control system and its intelligent control method | |
| CN108637435A (en) | A kind of three-dimensional seam tracking system and method for view-based access control model and arc voltage sensing | |
| CN112529858A (en) | Welding seam image processing method based on machine vision | |
| Wang | Unsupervised recognition and characterization of the reflected laser lines for robotic gas metal arc welding | |
| Cai et al. | Real-time identification of molten pool and keyhole using a deep learning-based semantic segmentation approach in penetration status monitoring | |
| CN117557571A (en) | Visual detection method and system for alloy resistance welding defects based on image enhancement | |
| Fang et al. | Weld pool image segmentation of hump formation based on fuzzy C-means and Chan-Vese model | |
| CN115272204A (en) | Bearing surface scratch detection method based on machine vision | |
| CN113554587A (en) | Molten pool image geometric feature extraction method and system based on deep learning | |
| CN113781458A (en) | Artificial intelligence based identification method | |
| Zhu et al. | Multi-classification recognition and quantitative characterization of surface defects in belt grinding based on YOLOv7 | |
| He et al. | Discerning weld seam profiles from strong arc background for the robotic automated welding process via visual attention features | |
| CN117444404B (en) | Intelligent positioning method and system for laser welding | |
| Wang et al. | Online detection of weld surface defects based on improved incremental learning approach | |
| Lin et al. | Intelligent seam tracking of an ultranarrow gap during K-TIG welding: a hybrid CNN and adaptive ROI operation algorithm | |
| CN115018813A (en) | Method for robot to autonomously identify and accurately position welding line | |
| Tang et al. | A new method to assess fiber laser welding quality of stainless steel 304 based on machine vision and hidden Markov models | |
| CN116228798B (en) | Intelligent iron tower cutting detection method based on machine vision | |
| CN115213563B (en) | Laser intelligent welding method and system | |
| He et al. | Parameter self-optimizing clustering for autonomous extraction of the weld seam based on orientation saliency in robotic MAG welding |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |