Weld seam tracking method based on underwater welding image

Abstract

The invention relates to a welding seam tracking method based on an underwater welding image, and relates to the technical field of underwater welding. Welding state monitoring operation; and monitoring the welding completion degree. When the device is used, the welding initial state is firstly monitored in an image recognition mode, then the real-time position in the welding process is obtained through the welding state monitoring operation, the real-time position in the welding process is corrected according to the initial state monitoring operation, and therefore the welding precision can be preliminarily guaranteed, and the welding quality is improved. And then monitoring after equipment welding is finished is achieved through welding completion degree monitoring operation, so that after welding is finished, the position of a welding seam is tracked, whether rework needs to be conducted or not is judged according to the welding completion degree, and the welding precision is further guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of underwater welding, and in particular to a weld tracking method based on underwater welding images. Background Art

[0002] Underwater welding is a welding operation performed in an underwater environment. It is mainly used for the installation, maintenance and construction of underwater facilities such as marine structures, ships and pipelines. Due to the presence of water, the underwater welding process is much more complicated than land welding.

[0003] Wet welding in underwater welding is direct welding underwater, and the arc burns underwater. This application is relatively wide, but the safety is poor. In this process, it is necessary to overcome the cooling and pressure effects of water. The arc starting voltage is higher than the arc starting voltage in the atmosphere, and the current demand is also large.

[0004] The above welding method usually adopts manual welding, that is, a manual handheld welding machine is used to directly weld the weld. A large number of bubbles will be generated during the welding process. The bubbles are generated from the weld. When they move vertically upward quickly, they will cause certain obstruction to the weld. Therefore, the end position of the manual handheld welding machine is very easy to be obstructed, resulting in the problem of inaccurate welding. Summary of the invention

[0005] The object of the present invention is to provide a weld tracking method based on underwater welding images.

[0006] The technical problem solved by the present invention is to solve the problem in the prior art that bubbles block the line of sight and affect the welding accuracy during underwater welding.

[0007] The present invention can be implemented by the following technical solution: A weld tracking method based on underwater welding images includes the following operations:

[0008] Initial state monitoring operation, used to monitor the initial state of underwater welding;

[0009] Welding status monitoring operation is used to monitor the welding status during underwater welding, obtain the deviation of welding position, and correct the welding position;

[0010] Welding completion monitoring operation, monitoring the equipment welding trend chart after completion, and judging whether rework is needed based on the welding completion degree.

[0011] A further technical improvement of the present invention is that the initial state monitoring operation includes the following operations:

[0012] Liquid environment temperature sensing operation: by installing a temperature sensor on the welding equipment and using the temperature sensor in contact with the liquid, the temperature of the liquid is sensed in real time to obtain a liquid temperature change table;

[0013] The initial welding state acquisition operation uses a positioning camera to photograph the welding position to obtain a physical image of the welding equipment, and then obtains a map of the area of ​​the position to be welded based on the physical image of the welding equipment.

[0014] A further technical improvement of the present invention is that the welding state monitoring operation includes the following operations:

[0015] The welding position determination operation obtains a map of the area of ​​the position to be welded, and obtains the working depth of the worker according to the first path parameter of welding in the map of the area of ​​the position to be welded;

[0016] The weld position determination operation sets the expected weld position and width according to the first path parameter in the weld position area diagram to obtain a simulated weld area for the weld;

[0017] Weld deviation monitoring operation is used to correct the actual weld range based on the simulated weld area.

[0018] A further technical improvement of the present invention is that when the welding completion monitoring operation is used, it is used to obtain the actual weld position map after the welding is completed, overlap and compare the weld position map and the simulated weld area, determine the deviation range between the two maps, and obtain the weld position deviation ratio value as the welding completion monitoring result.

[0019] A further technical improvement of the present invention is that the welding deviation monitoring operation includes the following steps:

[0020] The analysis operation of the area map of the position to be welded obtains the first path parameter in the area map of the position to be welded, matches the liquid flow direction according to the inclination degree of the first path parameter, guides the bubble offset direction according to the control of the liquid flow direction, and shoots the welding position.

[0021] A further technical improvement of the present invention is that when analyzing the first path parameter, the inclination angle in the first path parameter is obtained for partitioning to obtain a plurality of path segments Ln, where n is a positive integer greater than 0, and each path segment Ln corresponds to a respective slope kn relative to the horizontal plane, and the liquid flow direction in each area is determined according to the slope kn corresponding to each path segment Ln.

[0022] A further technical improvement of the present invention is that when the slope kn of the path segment Ln is 0 degrees, during the welding process, a number of density area movement speed diagrams of the bubbles are obtained by taking pictures and tracking them, and the welding trend diagram is matched according to the multiple density area movement speed diagrams. According to the welding trend diagram, a weld simulation circuit diagram about the bubbles is obtained, and the weld simulation circuit diagram is compared with the simulated weld area to determine whether there is any deviation in the weld.

[0023] A further technical improvement of the present invention lies in that when the slope kn of the path segment Ln is 90 degrees, during the welding process, a liquid with pressure is applied in the welding direction of the path segment Ln, and the angle at which the liquid is applied is b, where b is a set value. By monitoring the bubbles at the set position, it is judged whether the moving position of the density region corresponding to the bubbles is impacted to the set position within the specified time. If not, it is judged that the welding may deviate.

[0024] A further technical improvement of the present invention lies in that when the slope kn range of the path segment Ln is: 0 degrees < kn < 90, the liquid application angle is controlled degrees, which is used as the liquid application angle, where b is a set value. By monitoring the bubbles at the set position, it is judged whether the moving position of the density region corresponding to the bubbles is impacted to the set position within the specified time. If not, it is judged that the welding may deviate.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] 1. When the present application is in use, by using the method of image recognition, first, the initial welding state is monitored, and then the real-time position during the welding process is obtained by using the welding state monitoring operation. The real-time position during the welding process is corrected according to the initial state monitoring operation, so as to initially ensure the welding accuracy. Subsequently, by using the welding completion degree monitoring operation, the monitoring after the welding of the equipment is realized to ensure the tracking of the weld position after the welding is completed. Whether rework is required is judged according to the welding completion degree, further ensuring the welding accuracy.

[0027] 2. The present application analyzes the first path parameter obtained in the welding position determination operation, divides the first path parameter into several path segments, and analyzes each path segment separately. According to the slope corresponding to the path segment, during the welding of the path segment, by monitoring the position change of the bubbles, the recognition of the welding path can be realized, so as to judge whether the welding path is accurate. Therefore, in the present application, by converting the bubble interference into bubble monitoring, the detection of the welding path is realized, ensuring the tracking of the welding path, and thus ensuring the welding accuracy.

[0028] 3. By applying a water flow with a certain pressure and angle, the present application can enable the bubbles generated during the welding process to be quickly guided under the action of the water flow, further ensuring the speed of bubble movement, reducing the influence of bubbles on the line of sight during the welding process, and further ensuring the welding quality. Therefore, the path traceability during the welding process can be ensured in this operation, facilitating the correction of the welding position during the welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings.

[0030] Figure 1 is a flow chart of the method of the present invention;

[0031] Figure 2 It is a schematic diagram of the liquid application direction of the present invention. DETAILED DESCRIPTION

[0032] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0033] See also Figure 1-2 As shown, the weld tracking method based on underwater welding images includes the following steps:

[0034] Step 1: Initial state monitoring operation, which is used to monitor the initial state of underwater welding, including monitoring the temperature of the liquid environment during welding and the position to be welded during initial welding, and adjusting the welding conditions according to the initial state during welding, so as to ensure that the state of underwater welding is controllable;

[0035] Step 2: welding state monitoring operation is used to monitor the welding state during underwater welding, including welding position determination, weld position determination and welding deviation monitoring. The deviation of the welding position can be obtained according to the weld position determination and welding position determination, so that the welding deviation can be corrected to ensure the welding accuracy.

[0036] Step 3: The welding completion monitoring operation is used to monitor the welding trend chart of the equipment after completion, so as to judge the welding completion according to the result, that is, to judge whether the welding needs to be re-welded according to the welding completion to ensure the quality of the finished welding product.

[0037] As a further embodiment of the present application, step one includes the following operations:

[0038] Liquid temperature environment sensing operation, by installing a temperature sensor on the side of the handheld part of the welding equipment, using the temperature sensor to contact the liquid, sensing the temperature of the liquid, and sending the temperature of the liquid to the central processor for preliminary recording. The liquid temperature shows a certain change curve, and a liquid temperature change table is obtained;

[0039] The initial welding state acquisition operation uses a positioning camera to photograph the welding position to obtain a physical image of the welding equipment, and then obtains an area map of the position to be welded based on the physical image of the welding equipment. The image is reprocessed based on the area map of the position to be welded to obtain the specific shape and coordinate parameters of the position to be welded.

[0040] During the welding initial state acquisition operation, when the welding position area map is processed, the welding position of the two plates to be welded is marked, and a welding path is simulated according to the welding position. The welding path is recorded in the welding position area map and recorded as the first path parameter.

[0041] In the above technical solution, the liquid temperature environment perception operation and the welding initial state acquisition operation are recorded as original parameters to facilitate the later tracing of the welding state. The original parameters in the process include the liquid temperature change table and the area map of the position to be welded.

[0042] As a further embodiment of the present application, step 2 includes the following operations:

[0043] The welding position determination operation obtains a map of the area of ​​the position to be welded, matches the water depth of the worker according to the first path parameter of the map of the area of ​​the position to be welded, and then matches the corresponding welding equipment according to the water depth of the worker.

[0044] The weld position determination operation simulates and matches the weld position and width according to the first path parameter in the weld position determination operation, and records the simulated weld area as a standard value, so as to facilitate tracing and recording of deviations of the actual weld.

[0045] The welding deviation monitoring operation is used to determine the operation according to the weld position and make real-time corrections to the position during the welding process to ensure the stability of the welding process and reduce the possibility of deviation during the welding process.

[0046] As a further embodiment of the present application, step three includes the following operations:

[0047] The welding completion monitoring operation is used to obtain the actual weld position and width after welding is completed, overlap and compare the actual weld image with the area map of the position to be welded, determine the deviation range between the two images, and thus obtain the weld position deviation ratio value.

[0048] Furthermore, the welding deviation monitoring operation is as follows:

[0049] The analysis operation of the area map of the position to be welded is performed to obtain the first path parameter in the area map of the position to be welded, and the liquid flow direction is matched according to the inclination degree of the first path parameter. Therefore, according to the control of the liquid flow direction, the bubble offset direction can be guided, thereby reducing the influence of the bubble, directly photographing the welding position, and tracking the welding position.

[0050] Specifically, first obtain the first path parameter, obtain the inclination angle in the first path parameter for partitioning, and obtain a number of path segments Ln, where n is a positive integer greater than 0, and each path segment Ln corresponds to a slope kn relative to the horizontal plane. The liquid flow direction of each area is determined according to the slope kn corresponding to each path segment Ln. In the present application, the direction of bubble generation is guided according to the liquid flow direction, so that the bubbles are quickly moved away from the welding position, ensuring that the weld can be quickly exposed, which is convenient for real-time monitoring of the weld position.

[0051] In order to guide the flow direction of the liquid, a simulation algorithm is used to perform simulation calculations. First, the slope kn of each path segment Ln and the height level hn of each path segment Ln (where the height level hn is the average height in the segment) are obtained, and the liquid temperature change table under the current state is obtained. The liquid temperature T in a unit time period is taken, and then it is simulated. According to the height level hn and the liquid temperature T of each path segment Ln, the overall floating speed of the bubbles is obtained.

[0052] Specifically, first, it is necessary to simulate and judge the overall floating speed of bubbles, that is, to use welding to process the welding points of each height, control the temperature T of the liquid to be consistent, and perform welding at a height h in the liquid. By collecting data with a camera, the density area where bubbles are generated is judged, and the moving speed v of the density area is obtained. By changing the liquid height h, several groups of moving speeds of density areas are obtained. The above are recorded as a data set, and simulated and calculated by a neural algorithm, that is, several groups of height h and density area moving speed v are obtained as training sets, and several groups of height h and density area moving speed v are obtained as verification sets. Under the premise of the current temperature T, the corresponding relationship between the height h and the density area moving speed v is obtained, that is, h=Z*v, and Z is the relationship parameter obtained by the neural algorithm;

[0053] Subsequently, a neural algorithm is used for simulation calculation to calculate the corresponding relationship between the height h, the density area moving speed v and the temperature T when the temperature T changes, that is, several groups of height h and density area moving speed v corresponding to the temperature T are obtained as training sets, and several groups of height h and density area moving speed v corresponding to the temperature T are obtained as verification sets, which are input into the neural algorithm simulation for simulation calculation to obtain h=X*(Z*v)*T; X is the relationship parameter obtained by the neural algorithm, thereby obtaining the relationship between the welding height, welding temperature and bubble moving speed.

[0054] In the above technical solution, the movement speed of the bubble is obtained based on the camera's timed shooting and recognition method, that is, through multiple continuous shooting methods, the average movement speed of the bubble within the recognition time range is determined, so that the welding process can be traced back.

[0055] Through the above technical solution, the relationship between welding height, welding temperature and bubble movement speed h=X*(Z*v)*T is obtained, and this relationship is applied to the actual welding process to monitor the welding in real time.

[0056] In order to obtain the relationship between the height level hn and liquid temperature T of each path segment Ln, as well as the overall floating speed of the bubbles, the present application inputs the parameters corresponding to each path segment into the formula h=X*(Z*v)*T, so as to obtain the numerical value of the floating speed vn corresponding to each path segment; then, based on the numerical relationship of the floating speed vn corresponding to each path segment Ln, it is determined whether it is necessary to adaptively adjust the speed of the bubbles to reduce the shielding effect of the bubbles on the welding part.

[0057] Specifically, the slope kn corresponding to each path segment Ln is obtained. When kn=0, there is no need to control the direction of the liquid at this time, and the bubbles can also move upward evenly. At this time, the floating speed of the bubbles is directly monitored to obtain a dynamic density area movement speed diagram. The multi-stem group density area movement speed diagram corresponding to the path segment Ln is obtained by fixed-point sampling to simulate the position of the weld in the area and obtain a simulated circuit diagram for the weld. The simulated circuit diagram is compared with the simulated weld area in the weld position determination operation to determine the possibility of deviation. If there is a large deviation, the terminal that directly responds to the deviation will alert the operator to stop welding and wait for the bubbles to be completely discharged. The weld is directly observed to determine the welding effect. In this process, it is convenient to monitor the welding of equipment under the horizontal welding path and reduce the possibility of large deviations, thereby transferring the influence of bubbles to bubble monitoring and preliminarily realizing the monitoring of the welding state.

[0058] Further, obtain the slope kn corresponding to each path segment Ln. When kn = 90 degrees, it is necessary to control the direction of the liquid at this time, that is, apply an inclined force to the bottom of the welding part to impact the welding part, and externally guide the bubbles generated in the welding part to reduce the visual impact caused by the bubbles. That is, during use, a liquid stamping device is installed at the bottom of the welding equipment, that is, the liquid is applied to the welding area with a certain pressure and angle b. At this time, the generated bubbles tilt toward the side of the path segment Ln under the action of the external force until the external force ends, and then float vertically. During this process, monitor the bubbles at the set position to judge whether the moving position of the density area corresponding to the bubbles is impacted to the set position within the specified time. If not, it is judged that the welding may deviate, and then directly respond to the deviation on the terminal to alert the operator to stop welding, wait for the bubbles to be completely discharged, directly observe the weld, and judge the welding effect.

[0059] Further, obtain the slope kn corresponding to each path segment Ln. When 0 degrees < kn < 90 degrees, control the application angle of the liquid according to the angle of kn. In this application, control the application angle of the liquid degrees as the application angle of the liquid, that is, apply the liquid in a certain direction externally to make the bubbles away from the path segment Ln and reduce the impact of the bubbles on the path segment Ln. And during this process, the generated bubbles tilt toward the side of the path segment Ln under the action of the external force until the external force ends, and then float vertically. During this process, monitor the bubbles at the set position to judge whether the moving position of the density area corresponding to the bubbles is impacted to the set position within the specified time. If not, it is judged that the welding may deviate, and then directly respond to the deviation on the terminal to alert the operator to stop welding, wait for the bubbles to be completely discharged, directly observe the weld, and judge the welding effect.

[0060] In the above technical solution, it is possible to initially perform real-time tracking and monitoring on the weld, so as to ensure the position accuracy of the weld.

[0061] When the bubbles are completely discharged, directly photograph the welding part to obtain a welding process diagram, compare the welding process diagram with the simulated welding area, and judge whether the weld position in the welding process diagram is within the welding range of the simulated welding area. If so, judge the welding stability. If not, judge that the welding position is abnormal and secondary treatment is required to ensure the position accuracy of the weld and reduce the possible leakage problem of the equipment.

[0062] As a further embodiment of the present application, if the welding accuracy can be guaranteed, the above-mentioned real-time monitoring can be cancelled. It is only necessary to control the angle of applying the liquid according to the slope kn corresponding to the path segment Ln so that the generated bubbles can be quickly removed to avoid the generated bubbles acting on the welding position, which causes the problem of poor visual effect and directly affects the observation of the welding part by the staff after welding. Therefore, in this process, the angle of liquid application is y, which is stable b, and this value can ensure the angle of bubble guidance.

[0063] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A weld tracking method based on underwater welding images, characterized in that: The following operations are included: Initial state monitoring operation, used to monitor the initial state of underwater welding; Welding status monitoring operation is used to monitor the welding status during underwater welding, obtain the deviation of welding position, and correct the welding position; Welding completion monitoring operation, monitoring the equipment welding trend chart after completion, and judging whether rework is needed based on the welding completion degree.

2. The weld seam tracking method based on underwater welding images according to claim 1, characterized in that: The initial status monitoring operation includes the following operations: Liquid environment temperature sensing operation: by installing a temperature sensor on the welding equipment and using the temperature sensor in contact with the liquid, the temperature of the liquid is sensed in real time to obtain a liquid temperature change table; The initial welding state acquisition operation uses a positioning camera to photograph the welding position to obtain a physical image of the welding equipment, and then obtains a map of the area of ​​the position to be welded based on the physical image of the welding equipment.

3. The weld seam tracking method based on underwater welding images according to claim 2 is characterized in that: The welding status monitoring operation includes the following operations: The welding position determination operation obtains a map of the area of ​​the position to be welded, and obtains the working depth of the worker according to the first path parameter of welding in the map of the area of ​​the position to be welded; The weld position determination operation sets the expected weld position and width according to the first path parameter in the weld position area diagram to obtain a simulated weld area for the weld; Weld deviation monitoring operation is used to correct the actual welding range based on the simulated weld area.

4. The weld seam tracking method based on underwater welding images according to claim 3 is characterized in that: When used, the welding completion monitoring operation is used to obtain the actual weld position map after welding is completed, overlap and compare the weld position map and the simulated weld area, determine the deviation range between the two maps, and obtain the weld position deviation ratio value as the welding completion monitoring result.

5. The weld seam tracking method based on underwater welding images according to claim 3 is characterized in that: The welding deviation monitoring operation includes the following steps: The analysis operation of the area map of the position to be welded obtains the first path parameter in the area map of the position to be welded, matches the liquid flow direction according to the inclination degree of the first path parameter, guides the bubble offset direction according to the control of the liquid flow direction, and shoots the welding position.

6. The weld seam tracking method based on underwater welding images according to claim 5 is characterized in that: When analyzing the first path parameter, the inclination angle in the first path parameter is obtained for partitioning to obtain a number of path segments Ln, where n is a positive integer greater than 0, and each path segment Ln corresponds to a slope kn relative to the horizontal plane. The liquid flow direction in each area is determined based on the slope kn corresponding to each path segment Ln.

7. The weld seam tracking method based on underwater welding images according to claim 6 is characterized in that: When the slope kn of the path segment Ln is 0 degrees, during the welding process, several density area movement speed graphs of the bubbles are obtained by taking pictures and tracking them. The welding trend graph is matched according to the multiple density area movement speed graphs. A weld simulation circuit diagram of the bubbles is obtained according to the welding trend diagram. The weld simulation circuit diagram is compared with the simulated weld area to determine whether there is a deviation in the weld.

8. The weld seam tracking method based on underwater welding images according to claim 6, characterized in that: When the slope kn of the path segment Ln is 90 degrees, during the welding process, a pressurized liquid is applied to the welding direction of the path segment Ln, and the angle at which the liquid is applied is b, where b is a set value. By monitoring the bubbles at the set position, it is determined whether the moving position of the density area corresponding to the bubble is impacted to the set position within the specified time. If not, it is determined that the welding may deviate.

9. The weld seam tracking method based on underwater welding images according to claim 6, characterized in that: When the slope kn of the path segment Ln ranges from: 0 degrees < kn < 90 degrees, control the liquid application angle degrees, as the liquid application angle, b is a set value. By monitoring the bubbles at the set position, judge whether the moving position of the density area corresponding to the bubbles is impacted to the set position within the specified time. If not, it is judged that the welding may deviate.

Images