KR100912905B1 - Method for controling welding robot - Google Patents

Abstract

A welding robot control method is disclosed. In a laser vision sensor that recognizes a repair weld at a junction between a reactor head and a control rod driving mechanism (CRDM) nozzle, two laser modules irradiate a laser beam at the junction of the nozzle and an imaging module is located between the two laser modules. The image is imaged by imaging the laser beam reflected from the joining portion of the nozzle. The welding robot control apparatus calculates distance information for controlling the repairing robot by removing image distortion due to diffuse reflection from the image of the junction portion of the nozzle detected using the laser vision sensor. According to the laser vision sensor and the welding robot control apparatus and method using the same, it is possible to quickly measure the shape of the reactor head inclined plane depending on the position of the CRDM nozzle, and robustly to the diffuse reflection of the EDM (Electrical Discharge Machining) machining surface. Accurate welding trajectories can be created.

Laser vision sensor, reactor head, CRDM nozzle, welding

Description

Welding robot control method {Method for controling welding robot}

1 is a cross-sectional view of a conventional reactor head,

2 is an internal configuration of a preferred embodiment of a laser vision sensor according to the present invention;

3 is a perspective view showing an example of a path of a laser beam of a laser vision sensor according to the present invention;

4 is a view showing a method of selecting a laser source of the laser vision sensor according to the present invention according to the inclined surface of the welding portion,

5 is a block diagram showing the configuration of a preferred embodiment of a welding robot control apparatus according to the present invention, and

6 is a flowchart illustrating a process of performing a preferred embodiment of the welding robot control method according to the present invention.

The present invention relates to a welding robot control method using a laser vision sensor that recognizes a junction between a reactor head and a CRDM nozzle, and more particularly, a junction portion of a reactor head and a CRDM nozzle corresponding to a radioactive contamination zone, that is, a J-groove portion. It relates to a welding robot control method that can automatically perform repair welding for.

In general, a control rod drive mechanism (CRDM) of a nuclear reactor controls the nuclear output of a nuclear power plant by controlling nuclear reactions of a nuclear reactor by drawing and inserting control rods by using electromagnetic force.

1 is a cross-sectional view of a conventional reactor head.

Referring to FIG. 1, the control rod drive device 102 is installed on an upper portion of the reactor head 101, and a CRDM nozzle 103 is provided at a lower end of the control rod drive device 102 so that the control rod drive device 102 may be substantially disposed through the CRDM nozzle 103. Withdrawal and insertion of the control rod is made. On the other hand, the junction portion of the CRDM nozzle 103 and the reactor head 101 has a predetermined shape, which is commonly referred to as a J-groove region 104, and the J-groove TIG (Tungsten Inert Gas) welding The CRDM nozzle and the reactor head are joined together.

However, there is a problem in that stress corrosion corrosion (SCC) occurs in the J-groove during long-term operation of a nuclear power plant. Accordingly, repair welding of the J-groove is required, but the repair welding is not easily performed because the inside of the reactor head is a radioactive contamination zone.

Therefore, the reactor head CRDM nozzle joint repair welding must be performed remotely or automatically. This requires a three-dimensional sensor that remotely recognizes the shape of the welding object and generates a welding trajectory. These sensors must be compact and light enough to be able to be mounted on a repair robot and have a shape that allows rotational movement between nozzles. In addition, it is necessary to measure the shape of the reactor head inclined plane depending on the position of the nozzle and to generate a robust and accurate welding trajectory even in the diffuse reflection of the electric discharge machining (EDM) machining surface.

Prior arts relating to repair of reactor head CRDM nozzle damage include US Patent Nos. 6,211,482 and Korean Patent Nos. 10-0598353 and 10-0500973.

Conventional US Patent No. 6,211,482 relates to the removal of damaged parts of the reactor head CRDM nozzle inner surface by discharge machining, and to repair the removed parts by laser welding, and proposes a repair method applicable in the radioactive contamination zone. However, there is a limit that does not present specific details about automatic generation of welding trajectories.

In addition, Korean Patent No. 10-0598353 discloses a three-dimensional vision sensor that recognizes a joint part as a technology related to an automated method and equipment for repairing a joint part of a reactor head and a controller driving device nozzle. However, there is a limit in which the specific method and shape of the disclosed 3D vision sensor are not presented.

In addition, Korean Patent No. 10-0500973 is a technology related to welding seam tracking using a scanning type laser vision sensor, which sequentially reflects a laser beam generated from a laser through a laser reflector and a scanning reflector, and then welds it along a laser spot. The laser vision sensor which irradiates a member is disclosed. However, the conventional Korean Patent No. 10-0500973 is structurally complicated because the mirror driving unit is added, and thus there is a problem in that it is not applicable when the working space is narrow due to the increased volume and weight of the laser vision sensor.

SUMMARY OF THE INVENTION The present invention has been made in an effort to provide a welding robot control method for rapidly measuring a shape of a reactor head inclined surface that varies depending on a position of a CRDM nozzle.

delete

delete

delete

Another technical problem to be achieved by the present invention is to provide a welding robot control method for generating a strongly accurate welding trajectory even in the diffuse reflection of the EDM (Electrical Discharge Machining) processing surface.

In order to achieve the above technical problem, a welding robot control device includes a laser source for generating a laser beam in a control device for controlling welding of a welded joint between a reactor head and a control rod driving mechanism (CRDM) nozzle, and generating the laser beam. A focusing lens for focusing the focused laser beam, a structured light lens for structuring the focused laser beam in a line shape, and a reflecting mirror for adjusting the focus so that the structured laser beam is irradiated to a junction portion of the nozzle. A laser vision sensor positioned between two laser modules and the two laser modules and configured to image an image by imaging an image of a laser beam reflected from a junction portion of a nozzle to detect an image of a repair welding portion; And a controller for controlling the laser beam irradiation of the laser vision sensor, calculating distance information from the detected repair welding portion image, and controlling a repair robot based on the calculated distance information.

delete

delete

delete

In addition, the welding robot control method for achieving the above technical problem, the reactor head and the CRDM (Control Rod Driving Mechanism) by using a laser vision sensor for detecting the repair welding site image by irradiating two laser beams from different directions A control method for controlling welding of a welding portion of a nozzle, comprising: a laser beam irradiation step of controlling rotation of the laser vision sensor and selectively irradiating the two laser beams to the bonding portion of the nozzle; Imaging the laser beam reflected from the junction portion of the nozzle to form an image; A distance information calculating step of calculating distance information from the formed image; And a robot control step of controlling the repairing robot based on the calculated distance information.

delete

Hereinafter, a preferred embodiment of a laser vision sensor and a welding robot control method using the same according to the present invention will be described in detail with reference to the accompanying drawings.

Figure 2 is an internal configuration diagram of an embodiment of a laser vision sensor applied to the present invention.

Referring to FIG. 2, the laser vision sensor 200 includes a case 201, two laser modules 202 and 203, an imaging module 206, and a protective glass 207.

 The case 201 is structured so that two laser modules 202 and 203 may be installed on both sides, respectively, and the imaging module 206 is designed to be installed in a central portion between the two laser modules 202 and 203. It is. Here, each of the laser modules 202 and 203 and the imaging module 206 is modularized into a single shape, and is designed to be used immediately if it is installed in the case 201.

The protective glass 207 is provided below the imaging module 206 in the case 201 and protects the imaging module 206 from dust, welding gas, etc. generated during welding. For example, the protective glass 207 may be a protective lead glass.

3 is a perspective view showing an example of a path of a laser beam of the laser vision sensor according to the present invention.

Referring to FIG. 3, the laser modules 202 and 203 irradiate a laser beam to the bonding portion 340 of the nozzle. The laser module 202, 203 includes a laser source 310, a focus lens 320, a structured light lens 330, and reflective mirrors 204, 205.

The laser source 310 generates a laser beam and transmits the generated laser beam to the focus lens 320. The focus lens 320 focuses the transmitted laser beam and irradiates the structured light lens 330. The structured light lens 330 changes the cross-sectional shape of the irradiated laser beam from a point to a line to structure and enter the reflective mirrors 204 and 205. The reflecting mirrors 204 and 205 irradiate the bonding portion 340 of the nozzle by adjusting the focus of the incident laser beam.

The imaging module 206 forms an image by imaging the laser beam reflected from the bonding portion 340 of the nozzle. The imaging module 206 includes a band pass filter 350, an imaging lens 360, and an imaging device 370.

The band pass filter 350 band-passes the laser beam reflected from the junction portion 340 of the nozzle to irradiate the imaging lens 360 with the laser beam having the same wavelength band as the laser beam generated by the laser source 310. The imaging lens 360 focuses and irradiates the irradiated laser beam onto the imaging device 370. The imaging device 370 captures the transmitted laser beam and forms an image. For example, a focusing lens for a CCD (Charged Coupled Device) may be used as the imaging lens 360, and a CCD (Charged Coupled Device) may be used as the imaging device 370.

The laser vision sensor 200 according to the present invention uses two laser modules 202 and 203. Since the reactor head has a hemispherical shape, the nozzle junction portion, that is, the measurement surface of the laser vision sensor 200 inevitably has an inclined surface, so that the laser of the laser vision sensor 200 always has an appropriate angle of incidence at the inclined surface. Laser modules 202 and 203 are needed.

4 is a view showing a method of selecting a laser source of the laser vision sensor according to the present invention according to the inclined surface of the welding site.

Referring to FIG. 4, when the laser vision sensor 200 measures the inclined surface 401, the right laser module 403 may be used to form an image for obtaining normal distance information. On the contrary, when the laser vision sensor 200 measures the opposite inclined surface 402, the left laser module 404 may be used to form an image for obtaining normal distance information. Otherwise, the image may not be accurately formed on the imaging device 370, which may cause incorrect distance information.

Figure 5 is a block diagram showing the configuration of a preferred embodiment of a welding robot control apparatus to which the present invention is applied.

Referring to FIG. 5, the welding robot control apparatus 500 according to the present invention includes a controller 510 and a laser vision sensor 520.

The laser vision sensor 520 selectively irradiates two laser beams to the bonding portion 104 of the nozzle in different directions, and captures the repaired welding portion image by imaging the laser beam reflected from the bonding portion 104 of the nozzle. . As a preferred embodiment of the laser vision sensor 520, the laser vision sensor 200 according to the present invention may be used.

The controller 510 controls the laser beam irradiation of the laser vision sensor 520, calculates distance information from the repair welding portion image detected by the laser vision sensor 520, and repairs the robot based on the calculated distance information ( 590). The controller 510 includes a laser vision sensor controller 512, a distance information calculator 514, and a robot controller 516.

The laser vision sensor controller 512 controls the laser vision sensor 200 to selectively irradiate the laser beam from another direction. The laser vision sensor controller 512 is selectively irradiated through the right laser module 403 and the left laser module 404 of the laser vision sensor 200 according to the horizontal rotation angle and the vertical rotation angle of the laser vision sensor 200. do. That is, the laser vision sensor controller 512 selects the laser modules 403 and 404 to irradiate the laser beam by using information on the inclined plane previously determined according to the horizontal rotation angle and the vertical rotation angle. In this way, the laser vision sensor controller 512 selects the laser modules 403 and 404 using information on the inclined planes previously determined according to the horizontal rotation angle and the vertical rotation angle, thereby rotating the laser vision sensor 200 by 360 degrees. It is possible to detect the repair welding site image for calculating the distance information by rotating within 180 degrees without the need. Accordingly, the welding robot control apparatus 500 according to the present invention can quickly measure the shape of the reactor head inclined surface depending on the position of the nozzle.

The distance information calculator 514 removes the image distortion due to the diffuse reflection of the laser beam from the repair welding portion image detected by the laser vision sensor 520 and calculates the distance information. In general, in order to repair the CRDM nozzle 103 of the reactor head, a portion where a defect occurs in advance of the repair engagement is removed. EDM method is generally used for this removal process and laser light is very irregularly reflected on the surface of Alloy600. If the diffuse reflection cannot be processed, accurate distance information cannot be calculated. Since the distance information calculation unit 514 removes the image distortion due to the diffuse reflection of the laser beam from the repair welding part image and calculates the distance information, the welding robot control apparatus 500 according to the present invention is also distanced to the diffuse reflection of the EDM processing surface. It is possible to accurately generate the welding trajectory which is information.

The robot controller 516 controls the repair robot 590 based on the distance information calculated by the distance information calculator 514. The repair robot 590 performs welding to the repair welding site based on the distance information calculated by the distance information calculator 514.

6 is a flowchart illustrating a process of performing a preferred embodiment of the welding robot control method according to the present invention.

Referring to FIG. 6, the laser vision sensor controller 512 controls the rotation of the laser vision sensor 520 and selectively irradiates two laser beams to a junction portion of the nozzle (S600). Here, the laser vision sensor controller 512 may shorten the rotation time of the laser vision sensor 520 within a preset angle range to shorten the time for the laser vision sensor 520 to irradiate the laser beam. The laser vision sensor 520 forms an image by imaging the laser beam reflected from the junction portion of the nozzle (S610). The distance information calculator 514 calculates distance information for calculating distance information from an image formed by the laser vision sensor 520, and the distance information calculator 514 generates noise from an image formed by the laser vision sensor 520. By removing the boundary line is detected (S620). The distance information calculator 514 extracts initial distance information on the laser image coordinate system from the detected boundary line (S630). The distance information calculator 514 corrects the initial distance information by performing profiling on the extracted initial distance information to remove the distorted information due to the diffuse reflection of the laser beam (S640). The distance information calculator 514 performs profiling using the shortest distance algorithm. The distance information calculator 514 calculates distance information on the actual coordinate system from the corrected initial distance information (S650). The robot controller 516 controls the repair robot 590 based on the distance information calculated by the distance information calculator 514 (S660).

Although the preferred embodiments of the present invention have been shown and described above, the present invention is not limited to the above-described specific preferred embodiments, and the present invention belongs to the present invention without departing from the gist of the present invention as claimed in the claims. Various modifications can be made by those skilled in the art, and such changes are within the scope of the claims.

According to the present invention, it is possible to measure the shape of the reactor head inclined plane depending on the position of the CRDM nozzle, and to produce a laser vision sensor having a shape that is compact and light and capable of rotating movement between nozzles so that it can be mounted on a repair robot. In addition, the laser vision sensor can be fabricated by modularizing the inner core to easily repair internal electronic components.

In addition, the welding robot control method according to the present invention can quickly measure the shape of the reactor head inclined plane depending on the position of the CRDM nozzle using such a laser vision sensor, and also for diffuse reflection of the EDM (Electrical Discharge Machining) machining surface. Robustly accurate weld trajectories can be generated.

Claims (9)

delete delete delete delete delete delete In a control method for controlling repair welding at the junction of a reactor head and a control rod driving mechanism (CRDM) nozzle by using a laser vision sensor that detects repair welding site images by irradiating two laser beams from different directions, A laser beam irradiation step of controlling rotation of the laser vision sensor and selectively irradiating the two laser beams to a junction portion of the nozzle; Imaging the laser beam reflected from the junction portion of the nozzle to form an image; A distance information calculating step of calculating distance information from the formed image; And And a robot control step of controlling the repairing robot based on the calculated distance information. The method of claim 7, wherein In the laser beam irradiation step, Welding robot control method, characterized in that to reduce the time to irradiate the laser beam by limiting the rotation of the laser vision sensor within a predetermined angle range. The method of claim 7, wherein The distance information calculating step, A preprocessing step of detecting boundary lines by removing noise from the formed image; An initial distance information extraction step of extracting initial distance information on a laser image coordinate system from the detected boundary line; A distance information correction step of correcting the initial distance information by performing profiling on the extracted initial distance information to remove distorted information due to diffuse reflection of a laser beam; And And a distance information calculating step of calculating distance information on an actual coordinate system from the corrected initial distance information.

Images