CN111968072A - Thick plate T-shaped joint welding position autonomous decision-making method based on Bayesian network - Google Patents
Thick plate T-shaped joint welding position autonomous decision-making method based on Bayesian network Download PDFInfo
- Publication number
- CN111968072A CN111968072A CN202010644425.1A CN202010644425A CN111968072A CN 111968072 A CN111968072 A CN 111968072A CN 202010644425 A CN202010644425 A CN 202010644425A CN 111968072 A CN111968072 A CN 111968072A
- Authority
- CN
- China
- Prior art keywords
- welding
- node
- point
- characteristic
- points
- 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.)
- Granted
Links
- 238000003466 welding Methods 0.000 title claims abstract description 189
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000000007 visual effect Effects 0.000 claims abstract description 23
- 238000005070 sampling Methods 0.000 claims abstract description 13
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 6
- 230000035772 mutation Effects 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000037452 priming Effects 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229910000679 solder Inorganic materials 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
- B23K37/00—Auxiliary devices or processes, not specially adapted to a procedure covered by only one of the preceding main groups
-
- 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
- B23K9/00—Arc welding or cutting
- B23K9/32—Accessories
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
Abstract
The invention provides an autonomous decision-making method for the starting/welding position of a T-shaped joint of a thick plate based on a Bayesian network model by means of the visual conversion characteristics of welding experience and leg requirements. Firstly, detecting characteristic information of the outline of a welding seam to be welded according to visual sensing, converting the requirement of a welding leg into visual description characteristics, and judging the filling state of the welding seam into three stages of bottoming, filling and cover surface welding; secondly, establishing a Bayesian network model by utilizing the weld contour characteristic points detected in real time and the state of weld filling judgment; and finally, by combining an estimated posterior importance sampling reasoning algorithm, carrying out real-time decision making on the welding position based on a maximum posterior probability criterion, and selecting the welding position with the maximum posterior probability from the identified welding seam contour characteristic points. The method aims to realize the automatic decision of the welding position during initial and welding in the multi-pass welding process of the T-shaped joint with the web plate thickness exceeding 30mm through the algorithm model, and improve the welding efficiency.
Description
Technical Field
The invention relates to a Bayesian network-based thick plate T-shaped joint welding position autonomous decision method, and belongs to the technical field of automatic welding based on a visual sensor.
Background
Real-time autonomous decision-making of welding positions of multilayer multi-pass arc welding is one of the main technologies affecting the welding efficiency of thick plates. At present, multilayer multi-pass arc welding is still adopted for thick plate welding. In multilayer and multi-pass automatic welding, a certain position needs to be continuously selected from the welding seam outline of a region to be welded for multiple times as a welding position in starting welding and tracking of the next welding seam, and the selection process is a welding position decision process.
Because different welding positions can generate different welding effects, the effective welding position decision method can not only improve the welding efficiency, but also concern the welding quality. Because effective decision-making of the welding position in multi-pass welding needs to be carried out according to related welding knowledge, current groove detection information and the like, the independent decision-making process of the welding position is very challenging to implement by researching related technologies, algorithm flows and the like. Currently, related researches mostly stay in welding position detection, simulation environment-based welding seam planning, visual sensing-based welding seam primary planning, online correction and the like, and no research is involved in real-time decision-making of a proper welding position. Therefore, it is necessary to design a method for improving welding efficiency while meeting the welding quality of a joint in engineering by comprehensively considering the working condition specificity of thick plate welding.
Disclosure of Invention
Aiming at the defects of the real-time decision method for the welding position of the T-shaped thick plate joint, the invention provides the Bayesian network-based automatic decision method for the welding position of the T-shaped thick plate joint, which realizes the real-time automatic decision of the welding position during multi-pass welding starting/welding by utilizing the model, and improves the welding efficiency while trying to meet the welding quality of the joint. In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a Bayesian network-based thick plate T-shaped joint welding position autonomous decision method comprises the following steps:
step one, detecting characteristic information of the outline of a welding seam to be welded according to visual sensing, converting the requirement of a welding leg into visual description characteristics, and judging the filling state of the welding seam into three stages of bottoming, filling and cover surface welding.
And secondly, establishing a Bayesian network model by utilizing the weld contour characteristic points and the weld filling judgment states detected in real time, and calculating the conditional probability of each sub-node by combining the empirical knowledge of the welding position and the coordinate information of the characteristic points identified in real time.
And thirdly, obtaining the posterior probability of each welding line contour feature point node by utilizing an estimated posterior importance sampling reasoning algorithm, and selecting a corresponding feature point as the welding position of the current sampling based on a maximum posterior probability rule.
Further, in the first step, characteristic information of the contour of the weld to be welded is detected according to visual sensing, and the specific steps are as follows:
the first step is as follows: according to a welding seam image acquired by visual sensing based on structured light, an identified welding seam outline is refined to have only one welding seam data point in the vertical direction, and the slope is calculated by adopting a formula (1):
wherein x (), y () represent the coordinates of the data point; n denotes the number of data points involved in each calculation, j denotes the data point footer, and i denotes the footer for data points adjacent to j.
The second step is that: and a one-dimensional smoothing filter is adopted to further suppress the slope disturbance.
The third step: nonlinear piecewise fitting of the slope data is performed using a polynomial up to 50 times:
f(x)＝a1x50+a2x49+…a50 (2)
wherein the coefficient a1,a2,…a50Determined by the coordinates of the fitted data points.
The fourth step: firstly, the fitting result replaces the original slope data to carry out mutation detection. And secondly, acquiring the slope monotonous interval and the position of the slope monotonous interval by adopting an expression (3), marking the length of each interval according to the slope change rate of each monotonous interval, and sequencing the intervals from large to small.
The fifth step: and finally, introducing the number N of observation result supervision characteristic points, selecting the first N ordered monotonous intervals to determine the positions of the characteristic points, and determining the positions of the characteristic points by the determined monotonous intervals.
Further, in the step one, the requirement of the solder fillet is converted into a visual description characteristic, and the specific steps are as follows:
the first step is as follows: determining an included angle beta between the surface of the welding leg and the web plate according to the required sizes K1 and K2 of the joint welding leg;
the second step is that: identifying characteristic points of the weld contour, determining boundary points A on web plate openings from the identified characteristic points, then linearly fitting data of the web plates and data of the bottom plate respectively to obtain two linear equations, and determining intersection points B of the linear equations;
the third step: and drawing a straight line passing through the point A in the image, so that an included angle between the straight line and a straight line obtained by fitting the web data is beta, and an intersection point of the straight line and a straight line obtained by fitting the bottom plate data is D.
Further, in the step one, the filling state of the welding seam is judged to be three stages of priming, filling and cover surface welding, and the specific steps are as follows:
the first step is as follows: determination
And
whether or not it is established in the formula
The ordinate representing the feature point on the rightmost side,
and the ordinate represents the ith characteristic point, and when the rightmost characteristic point is positioned at the lowest position in the image and the distance difference between the rightmost characteristic point and other characteristic points in the vertical direction is more than or equal to 20 pixels, the current welding state is backing welding.
The second step is that: when the feature point at the rightmost end identified by the weld contour is to the left of the point D and the weld is in the non-backing weld, the current weld is in the fill weld stage.
The third step: and judging the current welding state to be neither a backing welding stage nor a filling welding stage, and judging the current welding state to be a cover surface welding stage.
Further, the conditional probability calculation of the bayesian network for autonomous decision of the welding position and each child node is specifically as follows:
the first step is as follows: the parent node of the Bayesian network is in a welding state, the first-layer child node is the horizontal distance between all the identified other feature points and the rightmost feature point, the second-layer child node is the vertical distance between all the identified other feature points and the rightmost feature point, the third-layer child node is the identified other feature points, and the last layer is the rightmost feature point.
The second step is that: the designated X represents a father node variable 'welding state', which has two states of 'filling welding' and 'cover welding'; specifying Yi(i ═ 1,2, …) represents the node "horizontal distance i", with two states "satisfied" and "not satisfied"; specifying Wj(j ═ 1,2, …) denotes the node "vertical distance j", there are also two states "satisfy" and "Not satisfied "; specifying Zk(k ═ 1,2, …) represents the "feature point k" node, with two states, "select" and "deselect"; the designation O represents the "rightmost feature point" node, and there are also two states, "select" and "deselect".
The third step: the conditional probability of a child node is defined as follows: by using
Represents node YkConditional probability of (c)
Represents a node WkConditional probability of (c)
Representing a node ZkConditional probability of (a) and with pi (Z)k) Representing a node ZkAt a parent node,. pi. (O) denotes the parent node at node O, dijRepresenting the conditional probability at O.
The fourth step: calculating the horizontal distance l and the vertical distance l in the model, wherein the two variables respectively refer to the horizontal distances (H) between the other characteristic points and the rightmost characteristic pointl＝xrm-xl) And a vertical distance (V)l＝yrm-yl)，xrmAnd yrmRespectively, the coordinates of the rightmost feature point in the image.
The fifth step: distance threshold values H are respectively set in the horizontal direction and the vertical direction0And V0When the distances between other characteristic points and the rightmost characteristic point in the horizontal and vertical directions are larger than H0And V0Then, the feature point participates in the welding position decision.
And a sixth step: in the filling welding stage:
wherein x is more than or equal to 3, and k belongs to (0,1), and the value is not unique. Experiments prove that x is 5, and k is 0.7 to meet the decision requirement.
The seventh step: in the cover surface welding stage:
wherein the content of the first and second substances,
and is
Eighth step: definition of
And is
Then
Also, the same applies toDefinition of
And is
Thus, the
Wherein the operation symbol V-shaped represents a logical operation or operation.
Further, the specific steps of implementing the real-time decision of the welding position based on the maximum posterior probability criterion are as follows:
the first step is as follows: and determining real-time evidence of the parent node, namely determining the stage of the current welding.
The second step is that: and obtaining the posterior probability of each weld contour feature point node by utilizing an estimated posterior importance sampling reasoning algorithm.
The third step: and selecting the characteristic point with the maximum posterior probability as the welding position of the sampling.
The invention has the beneficial effects that:
the invention provides an automatic decision-making method for a thick plate T-shaped joint welding position based on a Bayesian network model, which is an innovative reliable algorithm. And then, a Bayesian network model for the T-shaped joint multi-pass welding position decision is provided, the effectiveness of the model in implementing the welding position autonomous decision is verified, and a foundation is laid for subsequently improving the implementation of other typical joint welding position autonomous decisions.
Drawings
FIG. 1 is a flowchart of a method for autonomously deciding a welding position of a T-shaped joint of a thick plate based on a Bayesian network model according to the present invention;
FIG. 2 is an exemplary illustration of imaging analysis of a T-joint weld and visual feature transformation required for a fillet in accordance with the present invention;
FIG. 3 is a diagram of an exemplary structure of a Bayesian network model (taking four feature points as an example) according to the present invention;
FIG. 4 is a schematic diagram of an exemplary filling weld position decision in a multi-pass welding process for the front surface of a T-shaped joint with a web 30mm thick according to the present invention;
FIG. 5 is a diagram illustrating an exemplary weld location decision of a cap weld in a multi-pass weld on the back of a T-joint with a web 30mm thick according to the present invention;
FIG. 6 is a schematic diagram of an exemplary filling weld position decision in a multi-pass welding process for the back of a T-shaped joint with a web 50mm thick according to the present invention;
FIG. 7 is an exemplary diagram illustrating the weld position decision of a cap weld in a multi-pass welding process on the front surface of a T-shaped joint with a web thickness of 50mm according to the present invention.
Detailed Description
The present invention will be further described with reference to the following specific examples, but the present invention is not limited to these examples.
The experimental background related to the invention is as follows: the welding method is MAG welding, and the shielding gas is 20% CO2+ 80% Ar, welding material Q345b, web plate thickness of 30mm and 50mm respectively, filling bevel of K-shaped bevel, and about 7 and 11 welding seams are needed for filling one side respectively. H020 pixels, V0120 pixels. Experiments prove that x is 5, and k is 0.7 to meet the decision requirement.
Example 1: the web thickness was 30 mm. An autonomous decision method for a welding position of a T-shaped joint of a thick plate based on a bayesian network is shown in fig. 1 and includes:
step one, detecting characteristic information of the outline of a welding seam to be welded according to visual sensing, converting the requirement of a welding leg into visual description characteristics, and defining the filling state of the welding seam as three stages of bottoming, filling and cover surface welding.
And secondly, establishing a Bayesian network model by utilizing the weld contour characteristic points and the weld filling judgment states detected in real time, and calculating the conditional probability of each sub-node by combining the empirical knowledge of the welding position and the coordinate information of the characteristic points identified in real time.
And thirdly, obtaining the posterior probability of each welding line contour feature point node by utilizing an estimated posterior importance sampling reasoning algorithm, and selecting a corresponding feature point as the welding position of the current sampling based on a maximum posterior probability rule.
In the first step, characteristic information of the contour of the to-be-welded weld joint is detected according to visual sensing, and the specific steps are as follows:
the first step is as follows: according to the weld image (as shown in fig. 2a) acquired by the visual sensing based on the structured light, the identified weld contour is refined to have only one weld data point in the vertical direction, and the slope is calculated by adopting the formula (1):
wherein x (), y () represent the coordinates of the data point; n represents the number of data points involved in each calculation, and the larger the value of n in the range of 2-10, the better the effect of inhibiting the slope fluctuation, but the larger the calculation consumption, the more n is 5 in the research; j represents a data point subscript, i represents a subscript of a data point adjacent to j;
the second step is that: a one-dimensional smoothing filter is adopted to further restrain slope disturbance, and a filter window is selected to be 1 multiplied by 9;
the third step: the slope data is subjected to nonlinear segmentation (equipartition) fitting (formula 2) by adopting a polynomial up to 50 times, the overall change trend of the slope is effectively approximated, and the influence of local disturbance of the slope on subsequent mutation detection is reduced.
f(x)＝a1x50+a2x49+…a50 (2)
Wherein the coefficient a1,a2,…a50Determining coordinates of the fitted data points;
the fourth step: to detect this mutation, the fitting results are first tested for mutation in place of the raw slope data. Secondly, obtaining a slope monotonous interval and the position of the slope monotonous interval by adopting a formula (3), marking the length of each interval according to the slope change rate of each monotonous interval, and sequencing the intervals from large to small;
the fifth step: and finally, introducing the number N of observation result supervision characteristic points, selecting the first N ordered monotonous intervals to determine the positions of the characteristic points, and determining the positions of the characteristic points by the determined monotonous intervals.
The decision of the welding position in the first step is related to the current filling state of the groove, and the filling state of the thick plate joint is generally divided into backing welding, filling welding and cover surface welding. In order to realize real-time autonomous decision making, the state needs to be automatically judged by using visual characteristics of online detection, and the specific steps are as follows:
the first step is as follows: acquiring a backing weld outline (as shown in fig. 2b) by using visual sensing based on structured light, determining the weld outline belonging to a web plate area according to the identified weld outline characteristic points, obtaining all data points belonging to the web plate outline by taking the related characteristic points A as marks, and performing straight line fitting on the data points to obtain a straight line equation L1;
the second step is that: determining a welding seam contour belonging to a bottom plate area, obtaining all data points belonging to the bottom plate contour by taking the related characteristic points as marks, and performing straight line fitting on the data points to obtain a straight line equation L2;
the third step: the intersection B of the line L1 and L2 is determined (see FIG. 2d), and the intersection angle α of the line L1 and L2 is determined, the distortion angle between the two plates being
The fourth step: determining the fillet surface to web angle β (FIG. 2c) based on the web break upper end A to floor requirement K1 and the floor requirement K2;
the fifth step: drawing a straight line L3 passing through A in the image to enable the included angle between L3 and L1 to be beta;
and a sixth step: determining the intersection point D (shown in FIG. 2D) of L3 and L2, wherein the line segment BD in the image represents the requirement of the solder leg on the bottom plate;
the seventh step: and (3) making a vertical line of the line segment AD by passing the point B, and rotating the line segment BC clockwise by an angle gamma to BC ', wherein BC' represents the 3 rd requirement of the welding leg: the web extends a distance from the intersection of the base plate to the surface of the fillet.
Finally, the current welding state is judged according to the position of the rightmost characteristic point and the position of the point D which are identified in real time, and judgment is carried out firstly
And
whether or not it satisfies, wherein
The ordinate representing the feature point on the rightmost side,
the ordinate of the ith characteristic point is represented, and if the ordinate of the ith characteristic point is satisfied, the current welding state is backing welding; if the welding is not backing welding and the rightmost characteristic point is on the left of the point D, the current welding is filling welding; if the welding is neither backing welding nor filling welding, the current welding is cover surface welding; and if the judged result is backing welding, the welding position is the average coordinate position of the second characteristic point and the third characteristic point.
The conditional probability calculation of the Bayesian network model and the child nodes in the second step is specifically as follows:
a bayesian network model for autonomous decision making of the weld location for the fill and cap weld phases is shown in fig. 3.
Firstly, judging the welding state of the current welding to obtain evidence of a father node;
secondly, except the characteristic point at the leftmost side, the horizontal distance H between other characteristic points and the characteristic point at the rightmost side is calculated in sequencel＝xrm-xlAnd a vertical distance Vl＝yrm-ylWherein x isrmAnd yrmThe coordinates of the rightmost characteristic point in the image are respectively;
thirdly, calculating the conditional probability of each sub-node of the first layer, and in the filling welding stage:
wherein x is more than or equal to 3, and k belongs to (0,1), and the value is not unique. Experiments prove that x is 5, and k is 0.7 to meet the decision requirement.
In the cover surface welding stage:
wherein the content of the first and second substances,
and is
Definition of
And is
Then
Also, define
And is
Thus, the
Wherein the operation symbol V-shaped represents a logical operation or operation.
The method comprises the following steps of:
firstly, after determining prior probability and real-time evidence nodes, the model obtains the posterior probability of each target node (weld contour feature point) by utilizing an estimated posterior importance sampling reasoning algorithm;
and secondly, selecting the characteristic point with the maximum posterior probability as the welding position of the sampling.
Firstly, selecting one welding which is performed on the front side of a joint in a filling welding stage as a test object, wherein the contour appearance of a welding seam collected before welding and in welding is as shown in figure 4a, and according to the obtained characteristic point information (figure 4b) of the welding seam contour, the decision results of the Bayesian network and the analytic hierarchy process provided in the text are both characteristic points 1 which are taken as welding positions, the corresponding decision basis is different from the process (the left part of figure 4 c), and the welding result is as shown in figure 4 d. Secondly, selecting one welding of the back surface of the joint in the cover surface welding stage, wherein visual characteristic information of the welding seam is shown in figures 5a and b, the welding position decided by the two methods is a characteristic point 2, the decision basis is shown in the right part of figure 4c, and the welding result is shown in figure 5 c.
Example 2: the web thickness was 50mm and the weld location decision process was performed similarly to example 1.
Test results show that for the problem of selecting multilayer and multi-pass welding positions with the thickness of the T-shaped joint web plate within 50mm, the Bayesian network decision model provided by the invention can meet autonomous decision, and provides technical support for further realizing intelligent welding of thick plates; the method can also effectively implement the welding position decision by utilizing an analytic hierarchy process based on the maximum probability rule, but has the difficult problem that a judgment matrix cannot be effectively and automatically constructed so as to automatically adapt to the change of the number of characteristic points of the contour of the welding line.
The foregoing merely represents preferred embodiments of the invention, which are described in some detail and detail, and therefore should not be construed as limiting the scope of the invention. It is specifically stated that the above calculation method for the prior probability is not exclusive, but is included in the scope of the present invention.
Claims (6)
1. A Bayesian network-based thick plate T-shaped joint welding position autonomous decision method is characterized by comprising the following steps: the method comprises the following steps:
step one, detecting characteristic information of the outline of a welding seam to be welded according to visual sensing, converting the requirement of a welding leg into visual description characteristics, and judging the filling state of the welding seam into three stages of bottoming, filling and cover surface welding;
secondly, establishing a Bayesian network model by utilizing the weld contour characteristic points and the weld filling judgment states detected in real time, and calculating the conditional probability of each sub-node by combining the empirical knowledge of the welding position and the coordinate information of the characteristic points identified in real time;
and thirdly, obtaining the posterior probability of each welding line contour feature point node by utilizing an estimated posterior importance sampling reasoning algorithm, and selecting a corresponding feature point as the welding position of the current sampling based on a maximum posterior probability rule.
2. The autonomous decision-making method for the welding position of the T-shaped joint of the thick plate according to claim 1, characterized in that: in the first step, characteristic information of the contour of the welding seam to be welded is detected according to visual sensing, and the specific steps are as follows:
the first step is as follows: according to a welding seam image acquired by visual sensing based on structured light, an identified welding seam outline is refined to have only one welding seam data point in the vertical direction, and the slope is calculated by adopting a formula (1):
wherein x (), y () represent the coordinates of the data point; n denotes the number of data points involved in each calculation, j denotes a data point footer, i denotes a footer for data points adjacent to j;
the second step is that: a one-dimensional smoothing filter is adopted to further restrain slope disturbance;
the third step: nonlinear piecewise fitting of the slope data is performed using a polynomial up to 50 times:
f(x)＝a1x50+a2x49+…a50 (2)
wherein the coefficient a1,a2,…a50Determining coordinates of the fitted data points;
the fourth step: firstly, replacing original slope data with a fitting result to carry out mutation detection, then adopting a formula (3) to obtain a slope monotonous interval and the position of the slope monotonous interval, marking the length of each interval according to the slope change rate of each monotonous interval, and sequencing the intervals from large to small;
the fifth step: and finally, introducing the number N of observation result supervision characteristic points, selecting the first N ordered monotonous intervals to determine the positions of the characteristic points, and determining the positions of the characteristic points by the determined monotonous intervals.
3. The autonomous decision-making method for the welding position of the T-shaped joint of the thick plate according to claim 1, characterized in that: in the first step, the leg requirement is converted into a visual description characteristic, and the specific steps are as follows:
the first step is as follows: determining an included angle beta between the surface of the welding leg and the web plate according to the required sizes K1 and K2 of the joint welding leg;
the second step is that: identifying characteristic points of the weld contour, determining boundary points A on web plate openings from the identified characteristic points, then linearly fitting data of the web plates and data of the bottom plate respectively to obtain two linear equations, and determining intersection points B of the linear equations;
the third step: and drawing a straight line passing through the point A in the image, so that an included angle between the straight line and a straight line obtained by fitting the web data is beta, and an intersection point of the straight line and a straight line obtained by fitting the bottom plate data is D.
4. The autonomous decision-making method for the welding position of the T-joint of the thick plate according to claim 3, wherein: in the first step, the filling state of the welding line is judged to be three stages of priming, filling and cover surface welding, and the specific steps are as follows:
the first step is as follows: determination
Whether or not it is established in the formula
The ordinate representing the feature point on the rightmost side,
the ordinate of the ith characteristic point is represented, and when the rightmost characteristic point is at the lowest position in the image and the distance difference between the rightmost characteristic point and other characteristic points in the vertical direction is more than or equal to 20 pixels, the current welding state is backing welding;
the second step is that: when the characteristic point at the rightmost end identified by the welding seam outline is on the left side of the point D and the welding is in non-backing welding, the current welding is in a filling welding stage;
the third step: and judging the current welding state to be neither a backing welding stage nor a filling welding stage, and judging the current welding state to be a cover surface welding stage.
5. The autonomous decision-making method for the welding position of the T-shaped joint of the thick plate according to claim 1, characterized in that: the conditional probability calculation of the bayesian network and each child node in the second step is specifically as follows:
the first step is as follows: the parent node of the Bayesian network is in a welding state, the first layer of child nodes are the horizontal distances between all the identified other feature points and the rightmost feature point, the second layer of child nodes are the vertical distances between all the identified other feature points and the rightmost feature point, the third layer of child nodes are the identified other feature points, and the last layer of child nodes is the rightmost feature point;
the second step is that: the designated X represents a father node variable 'welding state', which has two states of 'filling welding' and 'cover welding'; specifying Yi(i ═ 1,2, …) represents the node "horizontal distance i", with two states "satisfied" and "not satisfied"; specifying Wj(j ═ 1,2, …) represents the node "vertical distance j", there are also two states "satisfied" and "not satisfied"; specifying Zk(k ═ 1,2, …) represents the "feature point k" node, with two states, "select" and "deselect"; the designated O represents the node of the rightmost characteristic point, and has two states of selection and non-selection;
the third step: the conditional probability of a child node is defined as follows: by using
Represents node YkConditional probability of (c)
Represents a node WkConditional probability of (c)
Representing a node ZkConditional probability of (a) and with pi (Z)k) Representing a node ZkAt a parent node,. pi. (O) denotes the parent node at node O, dijRepresents the conditional probability at O;
the fourth step: calculating the horizontal distance l and the vertical distance l in the model, wherein the two variables respectively refer to the horizontal distances (H) between the other characteristic points and the rightmost characteristic pointl＝xrm-xl) And a vertical distance (V)l＝yrm-yl)，xrmAnd yrmThe coordinates of the rightmost characteristic point in the image are respectively;
the fifth step: distance threshold values H are respectively set in the horizontal direction and the vertical direction0And V0When the distances between other characteristic points and the rightmost characteristic point in the horizontal and vertical directions are larger than H0And V0Then, the characteristic point participates in the decision of the welding position;
and a sixth step: in the filling welding stage:
wherein x is more than or equal to 3, and k belongs to (0,1), and the value is not unique;
the seventh step: in the cover surface welding stage:
wherein the content of the first and second substances,
and is
Eighth step: definition of
And is
Then
Also, define
And is
Thus, the
Wherein the operation symbol V-shaped represents a logical operation or operation.
6. The autonomous decision-making method for the welding position of the T-shaped joint of the thick plate according to claim 1, characterized in that: the concrete steps of implementing the real-time decision of the welding position based on the maximum posterior probability criterion in the third step are as follows:
the first step is as follows: determining real-time evidence of a father node, namely determining the current welding stage;
the second step is that: obtaining the posterior probability of each weld contour feature point node by using an estimated posterior importance sampling reasoning algorithm;
the third step: and selecting the characteristic point with the maximum posterior probability as the welding position of the sampling.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010644425.1A CN111968072B (en) | 2020-07-07 | 2020-07-07 | Automatic decision-making method for welding position of thick plate T-shaped joint based on Bayesian network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010644425.1A CN111968072B (en) | 2020-07-07 | 2020-07-07 | Automatic decision-making method for welding position of thick plate T-shaped joint based on Bayesian network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111968072A true CN111968072A (en) | 2020-11-20 |
| CN111968072B CN111968072B (en) | 2024-04-02 |
Family
ID=73360894
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010644425.1A Active CN111968072B (en) | 2020-07-07 | 2020-07-07 | Automatic decision-making method for welding position of thick plate T-shaped joint based on Bayesian network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111968072B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112975190A (en) * | 2021-03-23 | 2021-06-18 | 北京石油化工学院 | Multilayer multi-pass welding method, device, equipment and system based on visual sensing |
| CN113762400A (en) * | 2021-09-13 | 2021-12-07 | 南昌大学 | Weld joint position autonomous extraction method based on naive Bayes classifier |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101526480A (en) * | 2009-04-27 | 2009-09-09 | 清华大学 | Real-time detection method of butt weld of thin plates based on visual sense |
| US20110192825A1 (en) * | 2010-02-08 | 2011-08-11 | Prima Industrie S.P.A. | Method for monitoring the quality of laser-machining processes and corresponding system |
| CN102292187A (en) * | 2008-11-21 | 2011-12-21 | 普雷茨特两合公司 | Method and device for monitoring a laser machining operation to be performed on a workpiece and laser machining head having such a device |
| CN106952281A (en) * | 2017-05-15 | 2017-07-14 | 上海交通大学 | A kind of method that weld profile feature recognition and its welding bead are planned in real time |
| CN109949346A (en) * | 2019-03-04 | 2019-06-28 | 天津科技大学 | Welding seam tracking method based on histograms of oriented gradients particle filter |
-
2020
- 2020-07-07 CN CN202010644425.1A patent/CN111968072B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102292187A (en) * | 2008-11-21 | 2011-12-21 | 普雷茨特两合公司 | Method and device for monitoring a laser machining operation to be performed on a workpiece and laser machining head having such a device |
| CN101526480A (en) * | 2009-04-27 | 2009-09-09 | 清华大学 | Real-time detection method of butt weld of thin plates based on visual sense |
| US20110192825A1 (en) * | 2010-02-08 | 2011-08-11 | Prima Industrie S.P.A. | Method for monitoring the quality of laser-machining processes and corresponding system |
| CN106952281A (en) * | 2017-05-15 | 2017-07-14 | 上海交通大学 | A kind of method that weld profile feature recognition and its welding bead are planned in real time |
| CN109949346A (en) * | 2019-03-04 | 2019-06-28 | 天津科技大学 | Welding seam tracking method based on histograms of oriented gradients particle filter |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112975190A (en) * | 2021-03-23 | 2021-06-18 | 北京石油化工学院 | Multilayer multi-pass welding method, device, equipment and system based on visual sensing |
| CN113762400A (en) * | 2021-09-13 | 2021-12-07 | 南昌大学 | Weld joint position autonomous extraction method based on naive Bayes classifier |
| CN113762400B (en) * | 2021-09-13 | 2023-10-31 | 南昌大学 | Autonomous extraction method for weld joint position based on naive Bayes classifier |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111968072B (en) | 2024-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106952281B (en) | The method that a kind of identification of weld profile feature and its welding bead are planned in real time | |
| CN105118044B (en) | A kind of wheel shape cast article defect automatic testing method | |
| CN111968072A (en) | Thick plate T-shaped joint welding position autonomous decision-making method based on Bayesian network | |
| CN111667462B (en) | Automatic detection method and system for small and weak targets in welding defects | |
| CN109509181B (en) | Cladding pool shape visual detection method based on serial contour search | |
| CN110264457B (en) | Welding seam autonomous identification method based on rotating area candidate network | |
| CN115397595A (en) | Method, device and computer program for determining the performance of a welding method via digital processing of images of a welded workpiece | |
| CN111738369A (en) | Weld penetration state and penetration depth real-time prediction method based on visual characteristics of molten pool | |
| CN110814465A (en) | Universal method for automatically extracting welding seam contour | |
| CN113674206B (en) | Extraction method suitable for characteristic parameters of deep-melting K-TIG welding molten pool and keyhole entrance | |
| CN116597391B (en) | Synchronous on-line monitoring method for weld surface morphology and penetration state | |
| CN113516651A (en) | Welding joint defect detection method and device based on residual error network | |
| CN116493735B (en) | Real-time tracking method for motion splash in Wanwave-level ultra-high power laser welding process | |
| CN113421304A (en) | Intelligent positioning method for industrial radiographic negative image weld bead area | |
| CN105321166A (en) | Annular weld joint edge extraction method based on GAP predictor and self-adaptive genetic algorithm | |
| CN113909689B (en) | Pipeline welding groove feature extraction method of laser light bar | |
| CN110414388B (en) | Hump and fusion penetration online early warning method based on depth prediction network | |
| CN114119504A (en) | Automatic steel part welding line detection method based on cascade convolution neural network | |
| CN113551599A (en) | Welding seam position deviation visual tracking method based on structured light guidance | |
| CN110608684B (en) | Single-layer multi-channel weld accumulation deposition effect detection method and system | |
| CN111438460B (en) | Vision-based thick plate T-shaped joint welding seam forming characteristic online measurement method | |
| Kulkarni et al. | Machine learning based classification of welded components | |
| CN113814528A (en) | Interactive welding line curve extraction and automatic track generation method | |
| Du et al. | Automatic inspection of weld defects with x-ray real-time imaging | |
| Ma et al. | Weld geometry monitoring for metal inert gas welding process with galvanized steel plates using Bayesian network |
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 |