CN109702293A - A kind of welding penetration quality real-time control method of view-based access control model detection - Google Patents
A kind of welding penetration quality real-time control method of view-based access control model detection Download PDFInfo
- Publication number
- CN109702293A CN109702293A CN201910057453.0A CN201910057453A CN109702293A CN 109702293 A CN109702293 A CN 109702293A CN 201910057453 A CN201910057453 A CN 201910057453A CN 109702293 A CN109702293 A CN 109702293A
- Authority
- CN
- China
- Prior art keywords
- molten bath
- welding
- back side
- center
- welding bead
- 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
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a kind of welding penetration quality real-time control method of view-based access control model detection, steps are as follows: with the visual sensor continuous acquisition welding bead back side image with infrared high-pass filtering piece in welding process；The arc light information of molten bath heat radiation light and welding bead groove gap transmission is obtained by previous frame image；Molten bath back center region is determined according to heat radiation light information, and welding bead position of center line is calculated according to the arc light information that welding bead groove gap penetrates；Parallel light light source is opened when acquiring a later frame image, the shadow feature convexed to form according to molten bath and back of weld calculates molten bath back center position and molten bath back side characteristic width in conjunction with obtained molten bath back center region；Calculate the position offset between molten bath back center and welding bead center line；Welding parameter is adjusted according to the characteristic width at the molten bath back side and position offset and controls penetration, is repeated the above process until welding terminates, to realize the closed-loop control to welding penetration quality.
Description
Technical field
The present invention relates to a kind of welding penetration quality real-time control methods of view-based access control model detection, belong to Automation of Welding skill
Art field.
Background technique
The drawbacks of Automation of Welding replaces traditional artificial operation with automation equipment, overcomes human weld improves labour life
Efficiency is produced, is of great significance to guarantee welding process stability and weldquality.Welding penetration quality control is that welding is automatic
One of the technical issues of change faces.During actual welding, the fit-up gap of welding workpiece, radiating condition variation, workpiece are wrong
The various uncertain factors such as side will lead to that joint penetration quality is inconsistent, and local non-penetration, weldering leakage or weld seam is caused to misalign.
In order to overcome the influence of the uncertain factor welding quality in operating condition, the reliability of automatic welding, welding system are improved
System should be able to realize the real-time detection function to penetration quality in welding process, to further be reached by adjusting welding parameter in real time
To the control to welding penetration quality.
It finds, to there are many detection means of welding process penetration quality, such as uses by prior art documents
Arc voltage sense molten bath vibration information, using ultrasound examination molten bath the position of solid-liquid interface, using the reflected acoustic signal of welding process
Penetration signal etc..Vision-based detection is because the advantages of its is non-contact, electromagnetism interference, abundant information, is in welding in various detection means
With very extensive application prospect.The vision-based detection of penetration quality is divided into the positive vision-based detection in molten bath and molten bath is carried on the back
The vision-based detection in face.Chinese invention patent CN106624266B describes a kind of penetration quality determining method, passes through visual sensing
Device obtains molten bath positive information, and combines welding current information, calculates weld seam deviation and penetration signal using neural network model.
On the one hand, this method needs lot of experimental data to train a practical neural network model, to be believed according to molten bath front
Breath correctly reflects penetration signal, when operation conditions changes such as welding workpiece thickness, arc length, needs re -training model；Separately
On the one hand, interference of the image that molten bath front obtains by strong arc light, therefore signal-to-noise ratio is low, and characteristics of image is unstable." intelligence
Can manufacture " on the paper " Intelligent Control Research of aluminum alloy TIG weld seam back side width " delivered take to irradiate to back of weld and tie
Structure light obtains structure optical information with visual sensor, identifies the method for back of weld width to carry out welding penetration quality testing.
The back of weld width that this method obtains is the back of weld width of resolidified region, can be drawn to welding quality closed-loop control system
Enter time delay process, while the entire chamfered shape of molten bath and weld seam can not be obtained, can not also obtain weld seam centering situation, be reflected
Weld penetration information it is less.
Summary of the invention
In view of the deficiencies of the prior art, the present invention proposes that a kind of welding penetration quality of view-based access control model detection is controlled in real time
Method processed.The arc light that the present invention penetrates the molten bath heat radiation of welding back of work, welding bead groove gap carries out visual sensing, with
The shadow feature in the molten bath and back of weld that are constructed using directional light is combined, and molten bath back side characteristic width and molten bath back are obtained
Face center with respect to welding bead center line position offset, so that the pose to welding current and welding gun with respect to welding bead is adjusted in real time
Section realizes the control to penetration quality.The present invention is suitable for the backing welding in the case of welding workpiece closed butt joint, back linerless
It is welded with single layer.
The present invention is achieved by the following technical solutions:
A kind of welding penetration quality real-time control method of view-based access control model detection, steps are as follows:
(1) visual sensor and parallel light light source with infrared high-pass filter are placed at the welding workpiece back side,
Start to weld, visual sensor acquires welding bead back side image；
(2) arc light for obtaining molten bath heat radiation light information when acquiring odd-numbered frame image and being penetrated from welding bead groove gap
Information determines molten bath back center region according to heat radiation light information, the arc light information penetrated according to welding bead groove gap
Calculate welding bead position of center line；Parallel light light source is opened when acquiring even frame image, according to molten bath and back of weld
The shadow feature convexed to form, extract molten bath and weld seam back side profile, in conjunction with obtained by odd-numbered frame image molten bath back
Face center region calculates molten bath back center position and molten bath back side characteristic width；
(3) position offset between molten bath back center and welding bead center line is calculated, the feature according to the molten bath back side is wide
Degree adjusts welding current, and the relative pose of welding gun and welding bead is adjusted according to position offset；Step (2) are returned to, until welding knot
Beam.
In above-mentioned steps (1), it is described the welding workpiece back side place with infrared high-pass filter visual sensor and
Parallel light light source, specifically includes: the visual sensor visual field is directed at the welding workpiece rear surface regions immediately below welding gun, directional light
For the wave band of lighting source in the passband of infrared high-pass filter, illumination region is visual sensor shooting area, the illumination light
Source, visual sensor, the positional relationship of welding workpiece meet optical reflection law, i.e. the directional light of lighting source transmitting is welded
The reflection of workpiece can enter in visual sensor optical path.
Preferably, described that molten bath back center region is determined according to heat radiation light information in above-mentioned steps (2), tool
Body method are as follows: in odd-numbered frame image, the molten bath back side shows as the higher region of brightness due to heat radiation light, utilizes figure
As Threshold segmentation extracts the region, the position of the centroid in the region is calculated, using the neighborhood of the centroid as the molten bath
Back center region, Size of Neighborhood are determined according to experiment.
Preferably, in above-mentioned steps (2), the arc light information penetrated according to welding bead groove gap calculates welding bead center line
Position, method particularly includes: in odd-numbered frame image, the arc light penetrated from welding bead groove gap show as one it is long and narrow highlighted
Region extracts the highlight regions using Threshold segmentation, and carries out morphological erosion and refinement to the highlight regions, utilizes straight line
Hough transformation calculates the center line of the highlight regions, and using the center line as the welding bead center line.
Preferably, in above-mentioned steps (2), the shadow feature convexed to form according to molten bath and back of weld is extracted
The back side profile in molten bath and weld seam, method particularly includes: in even frame image, the lower nonreentrant surface of molten bath and back of weld causes
Its parallel light light reflected largely cannot be introduced into visual sensor, show as the darker area of brightness in the picture
Domain, and other regions at the welding workpiece back side reflect most of illuminating ray in visual sensor, surface is in the picture
Highlight regions extract molten bath and welded seam area using Threshold segmentation from image, traverse the pixel in the region, obtain the region
Profile, using the profile as the back side profile in the molten bath and weld seam.
Preferably, in above-mentioned steps (2), molten bath back center location that the combination is obtained by odd-numbered frame image
Domain calculates molten bath back center position and molten bath back side characteristic width, method particularly includes: using semiellipse equation to the molten bath
It is fitted with the back side profile of weld seam, the center of semiellipse is enumerated in the region of the molten bath back center, with minimum
Square law finds out the major semiaxis and semi-minor axis parameter of semiellipse, and selection makes the smallest semiellipse of error sum of squares as the molten bath
With the fitting result of back of weld profile, using the center of the semiellipse as the molten bath back center, by the semiellipse
Maximum width as molten bath back side characteristic width.
Preferably, in above-mentioned steps (3), the positional shift calculated between molten bath back center and welding bead center line
Amount, method particularly includes: using by the way of asking and a little arriving the distance of straight line, calculating molten bath back center between welding bead center line away from
From using the distance as the position offset.
The present invention carries out time-sharing multiplex to visual sensor, and synthesis is obtained by molten bath heat radiation, arc light, active illumination light source
Image information, directly acquire the position offset of molten bath back side characteristic width, molten bath back center and welding bead centreline space, and
Thus welding parameter is adjusted in real time controls penetration quality.The present invention quickly positions the molten bath back side using molten bath thermal radiation information and is scheming
Position as in, the ingenious arc light penetrated using welding bead groove gap are determined welding bead position of center line, utilize parallel light light
Source constructs shadow condition at the welding workpiece back side, so that molten bath and weld profile feature are obvious in the image obtained, significantly reduces
Image procossing difficulty improves image processing speed, reaches the requirement of real-time control.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the welding penetration quality real-time control method of view-based access control model detection disclosed by the invention；
Fig. 2 is the schematic device for welding workpiece back side visual sensing used in the embodiment of the present invention；
Fig. 3 is the schematic diagram for the odd-numbered frame image that visual sensor obtains in the embodiment of the present invention；
Fig. 4 is the schematic diagram for the even frame image that visual sensor obtains in the embodiment of the present invention；
Fig. 5 is the position offset δ obtained after image procossing in the embodiment of the present invention, molten bath back side characteristic width w shows
It is intended to；
Appended drawing reference:
1- welding gun；2- welding piece；
21- welding bead；The infrared high-pass filter of 3-；
4- visual sensor；5- parallel light light source.
Specific embodiment
The present invention is further illustrated with reference to the accompanying drawings and examples.
Fig. 1 is a kind of flow chart of the welding penetration quality real-time control method of view-based access control model detection disclosed by the invention,
Method and step is as follows:
(1) visual sensor and parallel light light source with infrared high-pass filter are placed at the welding workpiece back side,
Start to weld, visual sensor acquires welding bead back side image；
(2) arc light for obtaining molten bath heat radiation light information when acquiring odd-numbered frame image and being penetrated from welding bead groove gap
Information determines molten bath back center region according to heat radiation light information, the arc light information penetrated according to welding bead groove gap
Calculate welding bead position of center line；Parallel light light source is opened when acquiring even frame image, according to molten bath and back of weld
The shadow feature convexed to form, extract molten bath and weld seam back side profile, in conjunction with obtained by odd-numbered frame image molten bath back
Face center region calculates molten bath back center position and molten bath back side characteristic width；
(3) position offset between molten bath back center and welding bead center line is calculated, the feature according to the molten bath back side is wide
Degree adjusts welding current, and the relative pose of welding gun and welding bead is adjusted according to position offset；Step (2) are returned to, until welding knot
Beam.
In step (1), the visual sensor visual field is directed at the welding workpiece rear surface regions immediately below welding gun, and the present embodiment uses
Cutoff wavelength is in the high-pass filter of 808nm, and the wave band of parallel light light source is in the passband of infrared high-pass filter, illumination
Region is visual sensor shooting area, and the lighting source, visual sensor, the positional relationship of welding workpiece meet optical reflection
The reflection of the welded workpiece of directional light of law, i.e. lighting source transmitting can enter in visual sensor optical path, as shown in Fig. 2,
Parallel light light source 5, visual sensor 4 are equal with angle α, the β of 1 axis of welding gun, and three's axis is in the same plane, separately
Outside, the width direction of the image obtained in order to facilitate image procossing, visual sensor is perpendicular to welding direction.
Visual sensor picture is wide 300 pixel, high 400 pixel in the present embodiment.For ease of description, in visual sensor
Rectangular coordinate system is established in the image of acquisition, using pixel as length unit, the image upper left corner is origin, picture traverse and height side
To respectively X-axis and Y-axis.
In step (2), the odd-numbered frame image that acquires is as shown in figure 3, since parallel light light source 5 at this time is in
Closed state, the image for the arc light that the image that visual sensor 4 obtains is molten bath heat radiation, welding bead groove gap penetrates.Image
The middle molten bath back side is since heat radiation light is to showing as the higher irregular area of brightness, using carrying out image threshold segmentation, and from face
It is long-pending that the higher irregular area of the brightness can determine that image connectivity domain progress simple screening with two aspect of perimeter, calculate its shape
Position (the x of heart A0,y0), by the neighborhood D of A
D:{(x,y)|x∈[x0-Δx,x0+Δx]},|y∈[y0-Δy,y0+Δy]
As molten bath back center region, wherein Δ x, Δ y are preset value, represent the wave of molten bath back center
Dynamic range, is determined by testing, Δ x=Δ y=15 is taken in the present embodiment.
In step (2), in the odd-numbered frame image that acquires, the arc light penetrated from welding bead groove gap shows as one
Long and narrow highlight regions, the position of arc light represent the position of welding bead groove gap.Therefore the height is extracted using Threshold segmentation
Bright area, and morphological erosion and refinement are carried out to the highlight regions, the highlight regions are calculated using straight line Hough transformation
Center line, and using the center line as the welding bead center line L, as shown in Figure 2.
In step (2), the even frame image that acquires is as shown in figure 4, since parallel light light source at this time is in
The parallel light light that the lower nonreentrant surface of opening state, molten bath and back of weld causes it to reflect largely cannot be introduced into vision
In sensor, the darker region of brightness is shown as in the picture, and other regions at the welding workpiece back side are by most of illumination light
Line reflection enters in visual sensor, and surface is highlight regions in the picture, and molten bath and weldering are extracted from image using Threshold segmentation
Region is stitched, the pixel in the region is traversed, obtains the profile in the region, is taken turns the profile as the molten bath and the back side of weld seam
Exterior feature, as shown in phantom in figure 4.This method utilizes the lower convexity matter in molten bath and back of weld, constructs shadow item with directional light
Part enhances the characteristics of image of molten bath and back of weld.
Utilize semiellipse equation
The back side profile of above-mentioned molten bath and weld seam is fitted, (x ' in formula0,y0) be semiellipse center, also for true
Fixed molten bath back center position B, a, b are the parameter to be asked of semiellipse.(x ' is enumerated in the neighborhood D of above-mentioned A0,y0), with most
Small square law finds out semiellipse parameter a, b, and selection makes the smallest semiellipse of error sum of squares as the molten bath and back of weld
The fitting result of profile, by finally obtained (x '0,y0) it is used as molten bath back center position B, molten bath back side characteristic width w=
2a, as shown in Figure 5.
In conjunction with welding bead position of center line L obtained above and molten bath back center position B, using seeking distance between beeline and dot
Mode calculates position offset δ between the two, such as Fig. 5.
Welding current is adjusted according to the characteristic width w at the molten bath back side in step (3), welding gun is adjusted according to position offset δ
With the relative pose of welding bead, control method used can be the PID control of extensive utilization in industry, be also possible to fuzzy control
Deng.
It should be noted that above-described embodiment is only used to illustrate the technical scheme of the present invention, it is not to be protected to the present invention
The limitation of range.Therefore, although this specification is referring to above embodiment, invention is explained in detail, ability
The those of ordinary skill in domain is it should be understood that can modify to technical solution of the present invention or equivalent replacement, without departing from this
The spirit and scope of inventive technique scheme.
Claims (7)
1. a kind of welding penetration quality real-time control method of view-based access control model detection, which is characterized in that steps are as follows:
(1) visual sensor and parallel light light source with infrared high-pass filter are placed at the welding workpiece back side, started
Welding, visual sensor acquire welding bead back side image；
(2) the arc light letter for obtaining heat radiation light information in molten bath when acquiring odd-numbered frame image and being penetrated from welding bead groove gap
Breath, determines molten bath back center region according to heat radiation light information, the arc light information meter penetrated according to welding bead groove gap
Calculate welding bead position of center line；Parallel light light source is opened when acquiring even frame image, according to molten bath and back of weld
The shadow feature convexed to form extracts the back side profile in molten bath and weld seam, in conjunction with the molten bath back side obtained by odd-numbered frame image
Center region calculates molten bath back center position and molten bath back side characteristic width；
(3) position offset between molten bath back center and welding bead center line is calculated, according to the characteristic width tune at the molten bath back side
Welding current is saved, the relative pose of welding gun and welding bead is adjusted according to position offset；Step (2) are returned to, until welding terminates.
2. a kind of welding penetration quality real-time control method of view-based access control model detection according to claim 1, feature exist
In, it is described that the visual sensor and parallel light light source for having infrared high-pass filter are placed at the welding workpiece back side, specifically
Include:
The visual sensor visual field is directed at the welding workpiece rear surface regions immediately below welding gun, and the wave band of parallel light light source is infrared
In the passband of high-pass filter, illumination region is visual sensor shooting area, the lighting source, visual sensor, Welder
The positional relationship of part meets optical reflection law, i.e. the reflection of the welded workpiece of directional light of lighting source transmitting can enter vision
In sensor optical path.
3. a kind of welding penetration quality real-time control method of view-based access control model detection according to claim 1, feature exist
In, it is described that molten bath back center region is determined according to heat radiation light information, method particularly includes:
In odd-numbered frame image, the molten bath back side shows as the higher region of brightness due to heat radiation light, utilizes image threshold
Value segmentation extracts the region, the position of the centroid in the region is calculated, using the neighborhood of the centroid as the molten bath back side
Center region, Size of Neighborhood are determined according to experiment.
4. a kind of welding penetration quality real-time control method of view-based access control model detection according to claim 1, feature exist
In, the arc light information penetrated according to welding bead groove gap calculates welding bead position of center line, method particularly includes:
In odd-numbered frame image, the arc light penetrated from welding bead groove gap shows as a long and narrow highlight regions, utilizes threshold
Value segmentation extracts the highlight regions, and carries out morphological erosion and refinement to the highlight regions, utilizes straight line Hough transformation meter
The center line of the highlight regions is calculated, and using the center line as the welding bead center line.
5. a kind of welding penetration quality real-time control method of view-based access control model detection according to claim 1, feature exist
In the shadow feature convexed to form according to molten bath and back of weld extracts the back side profile in molten bath and weld seam, specific side
Method are as follows:
In even frame image, the parallel light light that the lower nonreentrant surface of molten bath and back of weld causes it to reflect is most of
It cannot be introduced into visual sensor, show as the darker region of brightness in the picture, and other regions at the welding workpiece back side will
Most of illuminating ray reflects in visual sensor, and surface is highlight regions in the picture, using Threshold segmentation from image
Extract molten bath and welded seam area, traverse the pixel in the region, obtain the profile in the region, using the profile as the molten bath and
The back side profile of weld seam.
6. a kind of welding penetration quality real-time control method of view-based access control model detection according to claim 1, feature exist
In the molten bath back center region that the combination is obtained by odd-numbered frame image calculates molten bath back center position and melts
Pond back side characteristic width, method particularly includes:
It is fitted using back side profile of the semiellipse equation to the molten bath and weld seam, in molten bath back center location
The center that semiellipse is enumerated in domain, the major semiaxis and semi-minor axis parameter of semiellipse are found out with least square method, and selection keeps error flat
Fitting result of the square and the smallest semiellipse as the molten bath and back of weld profile, using the center of the semiellipse as institute
Molten bath back center is stated, using the maximum width of the semiellipse as molten bath back side characteristic width.
7. a kind of welding penetration quality real-time control method of view-based access control model detection according to claim 1, feature exist
In, the position offset calculated between molten bath back center and welding bead center line, method particularly includes:
Using asking a little to by the way of the distance of straight line, molten bath back center is calculated the distance between to welding bead center line, with described
Distance is used as the position offset.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910057453.0A CN109702293B (en) | 2019-01-22 | 2019-01-22 | Welding penetration quality real-time control method based on visual detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910057453.0A CN109702293B (en) | 2019-01-22 | 2019-01-22 | Welding penetration quality real-time control method based on visual detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109702293A true CN109702293A (en) | 2019-05-03 |
| CN109702293B CN109702293B (en) | 2020-01-14 |
Family
ID=66261592
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910057453.0A Active CN109702293B (en) | 2019-01-22 | 2019-01-22 | Welding penetration quality real-time control method based on visual detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109702293B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110490867A (en) * | 2019-08-22 | 2019-11-22 | 四川大学 | Metal increasing material manufacturing forming dimension real-time predicting method based on deep learning |
| CN112453751A (en) * | 2020-11-27 | 2021-03-09 | 上海交通大学 | Pipeline all-position welding back surface online monitoring method based on visual sensing |
| CN113226633A (en) * | 2019-12-04 | 2021-08-06 | 东芝三菱电机产业系统株式会社 | Welding abnormality diagnosis device |
| CN114612457A (en) * | 2022-03-21 | 2022-06-10 | 徐州华宝能源科技有限公司 | Automobile part laser welding adjusting method and system based on computer vision |
| CN115106621A (en) * | 2022-05-20 | 2022-09-27 | 华南理工大学 | All-position robot deep melting K-TIG welding system and control method |
| CN116106980A (en) * | 2023-04-13 | 2023-05-12 | 江苏时代新能源科技有限公司 | Gap detection device and gap detection method |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102021117714A1 (en) | 2021-07-08 | 2023-01-12 | Endress+Hauser SE+Co. KG | Automatic seam detection for a welding process |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07214316A (en) * | 1994-01-28 | 1995-08-15 | Mitsubishi Heavy Ind Ltd | One side automatic welding method and its device |
| JP2006192437A (en) * | 2005-01-11 | 2006-07-27 | Hitachi Ltd | Welding equipment |
| CN101885102A (en) * | 2010-06-29 | 2010-11-17 | 哈尔滨工业大学 | Back surface camera and welding gun follow-up control device and method applied to detection of reverse molten bath for aluminum alloy TIG welding |
| CN107127432A (en) * | 2017-06-22 | 2017-09-05 | 西南交通大学 | The aluminum alloy MIG welding Fusion Control System and method adjusted based on welder |
-
2019
- 2019-01-22 CN CN201910057453.0A patent/CN109702293B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07214316A (en) * | 1994-01-28 | 1995-08-15 | Mitsubishi Heavy Ind Ltd | One side automatic welding method and its device |
| JP2006192437A (en) * | 2005-01-11 | 2006-07-27 | Hitachi Ltd | Welding equipment |
| CN101885102A (en) * | 2010-06-29 | 2010-11-17 | 哈尔滨工业大学 | Back surface camera and welding gun follow-up control device and method applied to detection of reverse molten bath for aluminum alloy TIG welding |
| CN107127432A (en) * | 2017-06-22 | 2017-09-05 | 西南交通大学 | The aluminum alloy MIG welding Fusion Control System and method adjusted based on welder |
Non-Patent Citations (2)
| Title |
|---|
| 田志杰等: "铝合金TIG焊缝背面宽度的智能控制研究", 《智能制造》 * |
| 章妍: "基于铝合金 GTAW 焊三光路视觉传感的熔池特征提取", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110490867A (en) * | 2019-08-22 | 2019-11-22 | 四川大学 | Metal increasing material manufacturing forming dimension real-time predicting method based on deep learning |
| CN113226633A (en) * | 2019-12-04 | 2021-08-06 | 东芝三菱电机产业系统株式会社 | Welding abnormality diagnosis device |
| CN113226633B (en) * | 2019-12-04 | 2023-02-17 | 东芝三菱电机产业系统株式会社 | Welding abnormality diagnosis device |
| CN112453751A (en) * | 2020-11-27 | 2021-03-09 | 上海交通大学 | Pipeline all-position welding back surface online monitoring method based on visual sensing |
| CN112453751B (en) * | 2020-11-27 | 2022-07-15 | 上海交通大学 | Pipeline all-position welding back surface online monitoring method based on visual sensing |
| CN114612457A (en) * | 2022-03-21 | 2022-06-10 | 徐州华宝能源科技有限公司 | Automobile part laser welding adjusting method and system based on computer vision |
| CN115106621A (en) * | 2022-05-20 | 2022-09-27 | 华南理工大学 | All-position robot deep melting K-TIG welding system and control method |
| CN115106621B (en) * | 2022-05-20 | 2023-05-23 | 华南理工大学 | Full-position robot deep-melting K-TIG welding system and control method |
| CN116106980A (en) * | 2023-04-13 | 2023-05-12 | 江苏时代新能源科技有限公司 | Gap detection device and gap detection method |
| CN116106980B (en) * | 2023-04-13 | 2024-02-20 | 江苏时代新能源科技有限公司 | Gap detection device and gap detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109702293B (en) | 2020-01-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109702293A (en) | A kind of welding penetration quality real-time control method of view-based access control model detection | |
| CN105458462B (en) | A kind of trapezoidal weld seam multi-parameter synchronizing visual detecting and tracking method of Varied clearance | |
| CN106825914B (en) | A kind of integrated laser welding gun with welding line tracking function | |
| Lü et al. | Feature extraction of welding seam image based on laser vision | |
| CN104942404B (en) | Dual wavelength binocular vision welding seam tracking method and tracking system | |
| CN107824940A (en) | Welding seam traking system and method based on laser structure light | |
| CN108637435A (en) | A kind of three-dimensional seam tracking system and method for view-based access control model and arc voltage sensing | |
| WO2022188482A1 (en) | Active and passive vision combination-based welding deviation detection system and detection method | |
| CN109604777A (en) | Welding seam traking system and method based on laser structure light | |
| CN109658456A (en) | Tank body inside fillet laser visual vision positioning method | |
| WO1988000508A1 (en) | Method of detecting position data in arc welding | |
| CN108032011B (en) | Initial point guiding device and method are stitched based on laser structure flush weld | |
| CN108856978A (en) | Corner joint penetration control method based near infrared binocular visual identity | |
| Zhang et al. | Narrow-seam identification and deviation detection in keyhole deep-penetration TIG welding | |
| CN107020449A (en) | Weld seam tracking sensor and its welding seam tracking method | |
| WO2020258712A1 (en) | Weld seam formation control apparatus and method | |
| CN105689859A (en) | Connecting method and device for various metals | |
| CN113909689B (en) | Pipeline welding groove feature extraction method of laser light bar | |
| CN109158790A (en) | A kind of Weld pool dynamic process real-time monitoring device and its application method | |
| Zheng et al. | Control for weld penetration in VPPAW of aluminum alloys using the front weld pool image signal | |
| Wang et al. | Tracking using pattern matching of keyhole in visual robotic plasma welding | |
| CN108788544B (en) | Welding seam initial point detection method based on structured light vision sensor | |
| CN206952357U (en) | Weld seam tracking sensor | |
| WO2020039948A1 (en) | Welding control device, display control device, welding system, welding control method, and program | |
| CN208132231U (en) | A kind of weld seam recognition tracking device and automatic welding process equipment |
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 |