CN1546268A - Intelligent control system for weld seam tracking and fusion penetration in spiral pipes - Google Patents
Intelligent control system for weld seam tracking and fusion penetration in spiral pipes Download PDFInfo
- Publication number
- CN1546268A CN1546268A CNA2003101227508A CN200310122750A CN1546268A CN 1546268 A CN1546268 A CN 1546268A CN A2003101227508 A CNA2003101227508 A CN A2003101227508A CN 200310122750 A CN200310122750 A CN 200310122750A CN 1546268 A CN1546268 A CN 1546268A
- Authority
- CN
- China
- Prior art keywords
- penetration
- weld
- weld seam
- helix tube
- temperature field
- 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.)
- Pending
Links
Images
Landscapes
- Butt Welding And Welding Of Specific Article (AREA)
Abstract
The invention discloses a kind of integrated intelligent control system for welding seam tracing and penetration in spiral tube, which includes spiral tube submerged arc welding machine, the character lies in: it also includes temperature field sensor, control box, and industrial control computer, there sets welding seam tracing and penetration integrated intelligent control model and interface driving circuit in the control box, there designs a temperature field welding seam and penetration characters identification information processing model in the industrial control computer. The invention only needs a sensor in order to extract the bias and penetration information of welding seam, which uses seam identification space frequency domain method to identify welding seam, uses penetration constant temperature line width method to identify the penetration degree. The invention also creates constant temperature line width which shows the penetration degree indirectly and reverse transmission nerve network for penetration degree. The invention realizes the automatic tracing of welding seam, completes the intelligent control for welding seam tracing and penetration.
Description
Technical field
The present invention relates to welding automatization technology, especially relate to a kind of helix tube inside weld and follow the tracks of and the penetration integrated intelligence control system.
Background technology
Welding be a kind of labour intensity bigger, to the process that workers ' health influence is serious, automatization level is lower, so Automation of Welding is the direction that the solder technology worker endeavours always.In welding process, the binomial key technology that influences automation is weld joint tracking and penetration control.For weld joint tracking, sensing technologies such as laser image, rotating the arc are arranged at present, the former be its price of additional sensor relatively expensive, have leading problem, need groove; The latter has realized that welding gun and sensor unite two into one, and does not have leading problem, but also needs groove.For penetration control, the extraction weld penetration information methods such as detecting weld pool surface shape, the heat radiation of the back side, molten bath that adopts is arranged, realize penetration control.Up to the present, realize simultaneously that the control of weld joint tracking and penetration still needs two to overlap independently sensing, information processing and control system.This uses very inconvenient for industry spot, two cover system price comparisons simultaneously are expensive, reliability is poor.
Therefore, distribute as the helix tube submerged-arc welding manufacture process from the temperature field, how to extract the key character parameter such as the information such as penetration, weld seam deviation of reflection welding quality quickly and accurately, realize weld joint tracking and the integrated control of penetration, farthest improve the automatization level of welding process, remain still unsolved technical barrier.For from the distribution in temperature field, extracting weld seam deviation, the research worker has carried out big quantity research both at home and abroad, mainly concentrate on gradient method, radius method, wavelet transformation etc., the two kinds of technical methods in front can be dealt with problems in desirable temperature field distributes, but for the industry spot applicable cases, still have problems such as the more weak and ratio of precision of antijamming capability is relatively poor, and wavelet transformation is identified in real-time control aspect and can't meets the demands.Therefore from the distribution in temperature field, obtain weld seam deviation information exactly and still need to carry out the research of deep technical research and application experiment from the aspects such as formation of weld seam information.For weld penetration information, there is not good extracting method so far.In order to realize penetration control, need quick, as to stablize, effectively from the temperature field distributes, extract weld penetration information method of research.
Summary of the invention
The object of the present invention is to provide sensor of a kind of need just can accurately extract weld joint tracking and the penetration integrated intelligence control system that identifies weld seam deviation and weld penetration information, can satisfy weldering manufacturing automation requirement in the helix tube simultaneously.
The object of the present invention is achieved like this: the present invention includes the helix tube submerged arc welding machine that is placed in the helix tube, feature is also to comprise the temperature field sensor, control cabinet and industrial control computer, the temperature field sensor that is installed in the helix tube inside weld back side is made up of infrared fileter and image charge coupled apparatus (ICCD), weld joint tracking and penetration integrated intelligent control module and interface driving circuit are installed in control cabinet, image pick-up card and input/output cards (I/O) are installed on the mainboard of industrial control computer, and design has temperature field weld seam and penetration feature identifying information processing module in industrial control computer.
Temperature field weld seam and penetration feature identifying information processing module are made of computer information processing module, weld seam recognition module and penetration identification module, weld joint tracking and penetration integrated intelligent control module are made of penetration deviation comparator, weld seam deviation comparator and multivariable neural network fuzzy control device, and interface driving circuit is made of interface circuit, crosshead shoe and wire feeding motor.
The present invention is made up of the helix tube submerged arc welding machine, temperature field sensor, control cabinet and the industrial control computer that are placed in the helix tube, the weld joint tracking and penetration integrated intelligent control module and the interface driving circuit that are made of weld seam deviation comparator, penetration deviation comparator and multivariable neural network fuzzy control device are installed in control cabinet, image pick-up card and input/output cards (I/O) are installed on the mainboard of industrial control computer, and design has temperature field weld seam and penetration feature identifying information processing module in industrial control computer.The present invention only needs a sensor just can accurately extract simultaneously and identifies weld seam deviation and weld penetration information, adopt weld seam recognition spatial frequency domain method identification weld seam, adopt the thermoisopleth width method identification penetration degree of penetration identification, the present invention has also set up the thermoisopleth width that reflects the penetration degree indirectly and the reverse transfer neutral net of penetration degree, thereby can be by extracting the purpose that the thermoisopleth width information reach identification penetration degree.The present invention has also designed weld joint tracking and the penetration integrated control module that comprises a multivariable neural network fuzzy control device, and this control mould utensil has than the better control performance of the simple fuzzy controller of PID.Weld seam recognition spatial frequency domain method has overcome that weld joint recognition method error ratios such as traditional gradient method, radius method are big, the more weak shortcoming of antijamming capability, and accuracy of identification has reached 0.1mm.
When the helix tube submerged arc welding machine begins welded seam, the temperature field sensor that is installed in the helix tube inside weld back side is penetrated the width of cloth at field signal and is converted to the temperature field signal, obtaining the helix tube temperature field distributes, the temperature field signal is sent into the temperature field weld seam and penetration feature identifying information processing module is carried out information processing, deliver to weld seam recognition module and penetration identification module after the information processing more respectively and carry out weld seam recognition and penetration identification, weld seam information that identifies and weld penetration information are imported the weld seam deviation comparator of weld joint tracking and penetration integrated intelligent control module more respectively, penetration deviation comparator, with given welding gun and position while welding, the thermoisopleth width compares, draw weld seam deviation, the penetration deviation, weld seam deviation, the penetration deviation is sent into multivariable neural network fuzzy control device simultaneously and is handled, result after the processing sends into and passes through interface circuit, the interface circuit output burst length drives crosshead shoe, the output armature voltage drives wire feeding motor, the welding gun travel mechanism and the welding current (wire feed rate) of control helix tube submerged arc welding machine, thereby the position adjustments that realizes welding torch realizes seam tracking, finish the control of weld joint tracking and penetration integrated intelligent, realize the automation of helix tube inside weld manufacture process.
Description of drawings
Fig. 1 is a structural representation of the present invention;
Fig. 2 is a structured flowchart of the present invention;
Fig. 3 is a back of weld temperature field distribution map;
Fig. 4 is gradient method recognition result figure;
Fig. 5 is the weld seam recognition spatial frequency domain method recognition result figure of Fig. 3;
Fig. 6 is the temperature field distribution map;
Fig. 7 is the Isothermal Line Distribution figure of Fig. 6;
Fig. 8 is section of weld joint figure;
Fig. 9 is thermoisopleth width neural network model figure;
Figure 10 is the arc welding process illustraton of model that utilizes the BP neutral net to set up;
Figure 11 is the internal frame diagram of multivariable welding quality neural network fuzzy control device;
Figure 12 is weld joint tracking and the unsteered experiment test specimen of penetration figure;
Figure 13 is the experiment test specimen figure of weld joint tracking and the integrated control of penetration.
The specific embodiment
Below in conjunction with embodiment and contrast accompanying drawing the utility model is described in further detail.
The present invention is by the helix tube submerged arc welding machine 2 that is placed in the helix tube 1, temperature field sensor 4, control cabinet 9 and industrial control computer 8 are formed, the temperature field sensor 4 that is installed in the helix tube 1 inside weld back side is made up of infrared fileter 3 and image charge coupled apparatus 5, weld joint tracking and penetration integrated intelligent control module 10 and interface driving circuit 12 are installed in control cabinet 9, image pick-up card 7 and input/output cards 6 are installed on the mainboard of industrial control computer 8, and design has temperature field weld seam and penetration feature identifying information processing module 11 in industrial control computer 8.
Temperature field weld seam and penetration feature identifying information processing module 11 are made of message processing module 21, weld seam recognition module 20 and penetration identification module 19, weld joint tracking and penetration integrated intelligent control module 10 are made of penetration deviation comparator 13, weld seam deviation comparator 14 and multivariable neural network fuzzy control device 15, and interface driving circuit 12 is made of interface circuit 16, crosshead shoe 17 and wire feeding motor 18.
1, the specific implementation algorithm of the identification of the weld seam recognition spatial frequency domain method in the weld seam recognition module 20 weld seam is:
At first carry out the outstanding weld seam information of horizontal filtering
M--participates in average pixel number in the formula, and Gray (k, l)--k is capable, the gray scale of l row pixel.
Vertical then filtering is to eliminate High-frequency Interference:
N--participates in average pixel number in the formula, b (i)--vertical filtered gray scale vector, a (i)--the output valve of horizontal filtering, i ∈ [0,500].
The 3rd step carried out smoothing processing to b (i) and obtains c (i):
In the formula--the gray scale vector after the smoothing processing, R--participate in average pixel number, and R is much larger than N, i ∈ [0,500].
Obtain weld seam feature distribution curve by handling above three groups of data at last:
d(i)=diff(b(i))*c(i)*abs(diff(b(i)-c(i)) (4)
Fig. 5 obtains the gradient distribution curve after adopting the spatial frequency domain method that Fig. 3 is handled, but processing structure contrast wind with Fig. 4, this method has improved the resolution ratio of identification greatly, especially to the relatively poor situation of picture quality, accuracy of identification reaches 0.1mm (error is within 10 pixels), is better than similar result of study.
2, the identification of the thermoisopleth width method in penetration identification module penetration degree:
With reference to Fig. 7, Fig. 8, Fig. 9, w is the thermoisopleth width, and h is the penetration degree.As seen from the figure, h is big more, and then w is big more.Therefore, in the present invention, adopt reverse transfer neutral net (BP neutral net) to set up h, the mapping relations between w just can be by extracting the purpose that the thermoisopleth width information reaches identification penetration degree.
3, the arc welding process model that utilizes the BP neutral net to set up: with reference to Figure 10, adopt neutral net to set up the model of weld seam deviation and penetration, the actual welds deviation of welding process is considered in the model input, weldingvoltage, speed of welding, weld width, welding current, this arc welding process model has comprised two BP neutral nets, BP network 1 is used for realizing actual deviation and detects non-linear relation between the weld seam deviation, BP network 2 is used for realizing the actual welds deviation, weldingvoltage, speed of welding, weld width, the decoupling zero of welding current and these several variablees of penetration degree, weldingvoltage, speed of welding, weld width is a constant.What BP network 1 adopted is the 1-5-1 structure, promptly imports dimension, the output dimension is 1, and the site is 5; What BP network 2 adopted is the 2-25-1 structure, but because weld width, speed of welding and weldingvoltage are uncontrollable in the actual welding process, so the input dimension is 2, i.e. and welding current and actual welds deviation, exporting dimension is 1, the site is 25.
4, multivariable neural network fuzzy control device: with reference to Figure 11, it includes a plurality of BP neutral nets, wherein BP network 1 is used for finishing the Nonlinear Mapping of object and the decoupling zero between multivariable, BP network 2,3 is used for finishing the fuzzy reasoning of conventional fuzzy controller, and sort controller has than the better control performance of the simple fuzzy controller of PID.
When the helix tube submerged arc welding machine begins welded seam, the temperature field sensor 4 that is installed in the helix tube 1 inside weld back side is penetrated the width of cloth at field signal and is converted to the temperature field signal, obtaining the helix tube temperature field distributes, the message processing module 21 that the temperature field signal is sent in temperature field weld seam and the penetration feature identifying information processing module 11 carries out information processing, deliver to weld seam recognition module 20 and penetration identification module 19 after the information processing more respectively and carry out weld seam recognition and penetration identification, weld seam information that identifies and weld penetration information are imported the weld seam deviation comparator 14 of weld joint tracking and penetration integrated intelligent control module 10 more respectively, penetration deviation comparator 13, with given welding gun and position while welding, the thermoisopleth width compares, draw weld seam deviation, the penetration deviation, weld seam deviation, the penetration deviation is sent into multivariable neural network fuzzy control device 15 simultaneously and is handled, result after the processing sends into interface circuit 16 again, the 16 output burst lengths of interface circuit drive crosshead shoe 17, the output armature voltage drives wire feeding motor 18, the welding gun travel mechanism and the welding current (wire feed rate) of control helix tube submerged arc welding machine, thereby the position adjustments that realizes welding torch realizes seam tracking, finish the control of weld joint tracking and penetration integrated intelligent, realize the automation of helix tube inside weld manufacture process.
Weld joint tracking precision of the present invention is at ± 0.5mm, fusion penetration control evenly, the thermoisopleth width is controlled at ± 1mm.
Claims (5)
1, a kind of helix tube inside weld is followed the tracks of and the penetration integrated intelligence control system, comprise the helix tube submerged arc welding machine (2) that is placed in the helix tube (1), it is characterized in that: also comprise temperature field sensor (4), control cabinet (9) and industrial control computer (8), the temperature field sensor (4) that is installed in helix tube (1) the inside weld back side is made up of infrared fileter (3) and image charge coupled apparatus (5), weld joint tracking and penetration integrated intelligent control module (10) and interface driving circuit (12) are installed in control cabinet (9), image pick-up card (7) and input/output cards (6) are installed on the mainboard of industrial control computer (8), and design has temperature field weld seam and penetration feature identifying information processing module (11) in industrial control computer (8).
2, helix tube inside weld as claimed in claim 1 is followed the tracks of and the penetration integrated intelligence control system, it is characterized in that: temperature field weld seam and penetration feature identifying information processing module (11) are by message processing module (21), weld seam recognition module (20) and penetration identification module (19) constitute, weld joint tracking and penetration integrated intelligent control module (10) are by penetration deviation comparator (13), weld seam deviation comparator (14) and multivariable neural network fuzzy control device (15) constitute, and interface driving circuit (12) is by interface circuit (16), crosshead shoe (17) and wire feeding motor (18) constitute.
3, helix tube inside weld as claimed in claim 2 is followed the tracks of and the penetration integrated intelligence control system, it is characterized in that: multivariable neural network fuzzy control device includes a plurality of reverse transfer through network, wherein reverse transfer network 1 is used for finishing the Nonlinear Mapping of object and the decoupling zero between multivariable, and reverse transfer network 2,3 is used for finishing the fuzzy reasoning of conventional fuzzy controller.
4, helix tube inside weld as claimed in claim 3 is followed the tracks of and the penetration integrated intelligence control system, it is characterized in that: what reverse transfer network 1 adopted is the 1-5-1 structure, and what reverse transfer network 2 adopted is the 2-25-1 structure.
5, helix tube inside weld as claimed in claim 2 is followed the tracks of and the penetration integrated intelligence control system, it is characterized in that: the weld seam feature distribution curve of the weld seam recognition spatial frequency domain method identification weld seam in the weld seam recognition module (20) is: d (i)=diff (b (i)) * c (i) * abs (diff (b (i)-c (i))).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2003101227508A CN1546268A (en) | 2003-12-17 | 2003-12-17 | Intelligent control system for weld seam tracking and fusion penetration in spiral pipes |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2003101227508A CN1546268A (en) | 2003-12-17 | 2003-12-17 | Intelligent control system for weld seam tracking and fusion penetration in spiral pipes |
Publications (1)
Publication Number | Publication Date |
---|---|
CN1546268A true CN1546268A (en) | 2004-11-17 |
Family
ID=34338740
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2003101227508A Pending CN1546268A (en) | 2003-12-17 | 2003-12-17 | Intelligent control system for weld seam tracking and fusion penetration in spiral pipes |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN1546268A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101793872A (en) * | 2010-02-25 | 2010-08-04 | 鞍山长风无损检测设备有限公司 | Whole spiral tube body flaw detection system capable of tube end flaw detection |
CN101690998B (en) * | 2009-09-20 | 2011-08-03 | 浙江金洲管道工业有限公司 | Pipe weld tracking shifter |
CN101733568B (en) * | 2009-12-09 | 2012-02-01 | 西安交通大学 | Dynamic measurement-based continuous welding method for reducing axial deformation generated by welding large thick-wall pipeline |
CN102554413A (en) * | 2012-01-05 | 2012-07-11 | 机械科学研究院哈尔滨焊接研究所 | Novel tracking method for internal welding of spiral pipe |
CN101676826B (en) * | 2008-09-19 | 2013-04-10 | 北京石油化工学院 | Coordination control system of double metal spiral compound pipe molding and welding |
CN101502906B (en) * | 2008-02-08 | 2013-05-22 | 通用汽车环球科技运作公司 | Weld feature monitoring method and apparatus |
CN104493345A (en) * | 2014-11-19 | 2015-04-08 | 天水锻压机床(集团)有限公司 | Whole-process tracing automatic adjusting system for steel pipe pre-welding machine and control method of whole-process tracing automatic adjusting system |
CN106624266A (en) * | 2016-12-31 | 2017-05-10 | 东莞职业技术学院 | Weld seam deviation and penetration state monitoring method for automobile 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 |
CN110142523A (en) * | 2018-02-13 | 2019-08-20 | 中国石油天然气集团有限公司 | Internal welding device and its control method |
CN113199184A (en) * | 2021-07-05 | 2021-08-03 | 北京航空航天大学 | Weld joint shape prediction method based on improved self-adaptive fuzzy neural network |
-
2003
- 2003-12-17 CN CNA2003101227508A patent/CN1546268A/en active Pending
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101502906B (en) * | 2008-02-08 | 2013-05-22 | 通用汽车环球科技运作公司 | Weld feature monitoring method and apparatus |
CN101676826B (en) * | 2008-09-19 | 2013-04-10 | 北京石油化工学院 | Coordination control system of double metal spiral compound pipe molding and welding |
CN101690998B (en) * | 2009-09-20 | 2011-08-03 | 浙江金洲管道工业有限公司 | Pipe weld tracking shifter |
CN101733568B (en) * | 2009-12-09 | 2012-02-01 | 西安交通大学 | Dynamic measurement-based continuous welding method for reducing axial deformation generated by welding large thick-wall pipeline |
CN101793872A (en) * | 2010-02-25 | 2010-08-04 | 鞍山长风无损检测设备有限公司 | Whole spiral tube body flaw detection system capable of tube end flaw detection |
CN102554413A (en) * | 2012-01-05 | 2012-07-11 | 机械科学研究院哈尔滨焊接研究所 | Novel tracking method for internal welding of spiral pipe |
CN104493345A (en) * | 2014-11-19 | 2015-04-08 | 天水锻压机床(集团)有限公司 | Whole-process tracing automatic adjusting system for steel pipe pre-welding machine and control method of whole-process tracing automatic adjusting system |
CN106624266A (en) * | 2016-12-31 | 2017-05-10 | 东莞职业技术学院 | Weld seam deviation and penetration state monitoring method for automobile welding |
CN106624266B (en) * | 2016-12-31 | 2018-08-07 | 东莞职业技术学院 | A kind of weld seam deviation and penetration signal monitoring method for Automobile 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 |
CN107127432B (en) * | 2017-06-22 | 2019-10-18 | 西南交通大学 | The aluminum alloy MIG welding Fusion Control System and method adjusted based on welder |
CN110142523A (en) * | 2018-02-13 | 2019-08-20 | 中国石油天然气集团有限公司 | Internal welding device and its control method |
CN113199184A (en) * | 2021-07-05 | 2021-08-03 | 北京航空航天大学 | Weld joint shape prediction method based on improved self-adaptive fuzzy neural network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN1546268A (en) | Intelligent control system for weld seam tracking and fusion penetration in spiral pipes | |
CN104992449B (en) | Information identification and surface defect online test method based on machine vision | |
CN111275063B (en) | Robot intelligent grabbing control method and system based on 3D vision | |
CN102699534A (en) | Scanning type laser vision sensing-based narrow-gap deep-groove automatic laser multilayer welding method for thick plate | |
CN108637435A (en) | A kind of three-dimensional seam tracking system and method for view-based access control model and arc voltage sensing | |
CN109693140B (en) | Intelligent flexible production line and working method thereof | |
Xu et al. | Real‐time image capturing and processing of seam and pool during robotic welding process | |
CN107688028A (en) | A kind of laser gain material manufactures overlapping rate on-line monitoring method | |
CN107957245A (en) | Engine link dimension measuring device and its measuring method based on machine vision | |
CN114211168A (en) | Method for correcting plane welding seam track based on image subtraction | |
CN113219971A (en) | Multifunctional intelligent logistics trolley based on convolutional neural network and multithreading parallel control | |
CN111123263A (en) | Tunnel reinforcing steel bar identification and detection system and method based on geological radar detection data | |
CN113909689A (en) | Method for extracting characteristics of pipeline welding groove of laser light strip | |
CN113510472A (en) | Intelligent air blast device and method for die closing and opening of tubular pile die | |
CN207832129U (en) | Engine link dimension measuring device based on machine vision | |
CN104281832A (en) | Visual identity industrial robot | |
CN113551599A (en) | Welding seam position deviation visual tracking method based on structured light guidance | |
CN113828892A (en) | HDR image-based molten pool center identification system and weld joint tracking method | |
CN117232396A (en) | Visual detection system and method for product quality of high-speed production line | |
CN101083877B (en) | Multi-head buried welding integration machine and its method for buried welding | |
Zou et al. | Weld pool image processing algorithm for seam tracking of welding robot | |
CN106530269A (en) | Weld detection method | |
CN113723266A (en) | Brushless motor punching sheet screening system and method | |
CN213196231U (en) | Welding seam tracking robot | |
CN110788858B (en) | Object position correction method based on image, intelligent robot and position correction system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |