Welding seam tracking and self-adaptive control method for welding robot

Abstract

The invention discloses a welding robot welding seam tracking and self-adaptive control method, which belongs to the technical field of welding robots, and comprises the following steps: S1, initial parameter setting: presetting initial welding parameters of a welding robot according to the material, thickness and welding process requirements of a to-be-welded workpiece; s2, welding seam initial positioning, wherein a visual sensor or a laser sensor is used for positioning a welding starting point; s3, welding seam feature extraction, wherein an image or a laser reflection signal of a welding seam area is collected in real time through a sensor; s4, welding seam tracking control, wherein the extracted welding seam geometric features are compared with a preset welding seam model; and S5, self-adaptive control: analyzing the cause of deviation. The welding robot welding seam tracking and self-adaptive control method has the advantages that welding seam tracking precision and stability are improved, welding parameter self-adaptive adjustment is achieved, the image processing and feature extraction process is optimized, an incremental parameter adjustment mode is adopted, and the welding robot welding seam tracking and self-adaptive control method has the welding quality detection and repair welding functions and the like.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of welding robots, in particular to a welding seam tracking and adaptive control method for welding robots. BACKGROUND

[0002] In modern manufacturing, welding as a key connection technology is widely used in automobile manufacturing, shipbuilding, aerospace, machining and many other fields. With the continuous improvement of industrial automation, welding robots have gradually replaced traditional manual welding and become the mainstream equipment of modern welding production due to their advantages of high efficiency, precision and stability. Welding robots can realize continuous and high-speed welding operation, greatly improving production efficiency, reducing labor intensity, and ensuring the stability and consistency of welding quality.

[0003] However, in the actual welding production process, welding robots are affected by many complex factors, which leads to great challenges in welding seam tracking and welding quality control. On the one hand, due to the inevitable errors in the machining and assembly process of workpieces, as well as the thermal deformation generated during the welding process, the actual position, shape and size of the weld may deviate from the preset welding trajectory. If the welding robot cannot accurately track the changes of the weld in time, it will cause the welding torch to deviate from the center of the weld, resulting in defects such as undercut, porosity and incomplete fusion, which will seriously affect the welding quality. On the other hand, the stability of the arc during welding is affected by many factors, such as fluctuations in welding current and voltage, interference from the welding environment, etc. Unstable arc will lead to poor weld formation, such as uneven weld width and inconsistent reinforcement, which will also affect the welding quality and product performance.

[0004] In order to solve the problems of welding robots in welding seam tracking and welding quality control, domestic and foreign scholars and enterprises have conducted a lot of research and proposed some corresponding technologies and methods. At present, the common welding seam tracking technologies mainly include contact tracking and non-contact tracking. Contact tracking obtains weld information by directly contacting the sensor with the weld, but this method is easy to wear the sensor and has poor tracking effect on complex-shaped welds. Non-contact tracking mainly uses visual sensors, laser sensors, etc. to obtain the image or signal of the weld, and then extracts the weld features through image processing or signal analysis algorithm to realize weld tracking. Although non-contact tracking technology has the advantages of not contacting the workpiece and strong adaptability, there are still some problems in actual application, such as poor adaptability to welding environment, easy to be disturbed by arc light, spatter and other factors, resulting in insufficient tracking precision and stability.

[0005] In terms of welding control, the traditional control method usually adopts fixed welding parameters for welding, which cannot dynamically adjust according to the real-time changes of the weld. When the weld deviates, the welding parameters and motion trajectories cannot be adjusted in time and effectively, so as to ensure the stability of the welding quality. Although some advanced control algorithms, such as PID control algorithm, can improve the control performance of the system to some extent, for the welding process with nonlinearity, time-varying and uncertainty, the control effect is still not ideal.

[0006] In summary, the existing welding robot weld tracking and control technology is difficult to meet the high requirements of modern industrial production on welding quality and efficiency, therefore, it is of great practical significance to develop a welding robot weld tracking and adaptive control method with high precision, high stability and self-adaptive ability. SUMMARY

[0007] In view of the deficiencies of the prior art, the present application provides a welding robot weld tracking and adaptive control method, which has the advantages of high precision, high stability and strong self-adaptive ability, and solves the problems raised in the above background art.

[0008] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0009] A welding robot weld tracking and adaptive control method, comprising the following steps:

[0010] S1, initial parameter setting: according to the material, thickness of the workpiece to be welded and the welding process requirements, the initial welding parameters of the welding robot are preset, the initial welding parameters include welding current, welding voltage, welding speed and wire feeding speed; at the same time, the initial search range and tracking accuracy threshold of the weld tracking are set;

[0011] S2, initial positioning of the weld: the starting point of welding is positioned by using a vision sensor or a laser sensor, the three-dimensional coordinate information of the starting point is obtained, and the welding torch of the welding robot is moved to the starting point position;

[0012] S3, weld feature extraction: in the welding process, the image or laser reflection signal of the weld area is collected in real time by the sensor, the image processing algorithm or signal analysis method is used to extract the geometric features of the weld, the geometric features include the weld centerline position, weld width and weld height;

[0013] S4, weld seam tracking control: compare the extracted weld seam geometric features with the preset weld seam model, calculate the deviation of the current position of the welding gun and the ideal weld seam position; according to the deviation value, a fuzzy control algorithm or a PID control algorithm is used to generate an adjustment instruction to adjust the motion trajectory of the welding robot in real time, so that the welding gun always moves along the weld seam center line, and weld seam tracking is realized; when the deviation value is less than the tracking accuracy threshold, the current welding parameters and motion trajectory are maintained; when the deviation value is greater than or equal to the tracking accuracy threshold, the adaptive control step is entered;

[0014] S5, adaptive control: analyze the causes of the deviation, and adjust the corresponding welding parameters according to different causes; if the deviation is caused by workpiece assembly error or thermal deformation, the motion trajectory of the welding robot and the welding speed are adjusted; if the deviation is caused by the change of arc stability during welding, the welding current and welding voltage are adjusted; the adjusted welding parameters are fed back to the welding robot, and the welding and weld seam tracking continue until the welding is completed.

[0015] Further, in the initial parameter setting, a welding parameter database corresponding to different materials and thicknesses of workpieces is established by experimental data, and the initial welding parameters are queried and obtained from the database according to the material and thickness of the workpiece to be welded.

[0016] Further, in the initial positioning of the weld seam, the visual sensor is any one of a CCD camera or a CMOS camera, and the laser sensor is any one of a laser displacement sensor or a laser triangulation sensor.

[0017] Further, in the weld seam feature extraction, the specific process of extracting the weld seam geometric features by using the image processing algorithm is as follows:

[0018] S3-1, pre-process the collected weld seam area image, which includes grayscale, filter denoising and image enhancement;

[0019] S3-2, use an edge detection algorithm to detect the edges of the weld seam;

[0020] S3-3, according to the detected edges, use the least square method to fit the weld seam center line and calculate the weld seam center line position; at the same time, count the number of pixels between the edges, and convert the weld seam width and height according to the image resolution.

[0021] Further, the edge detection algorithm is any one of Canny algorithm or Sobel algorithm.

[0022] Further, in the generation of the adjustment instruction by the fuzzy control algorithm in the weld tracking control, the deviation of the current position of the welding torch from the ideal weld position and the change rate of the deviation are taken as input variables of the fuzzy controller, and the adjustment instruction is taken as an output variable; fuzzy reasoning is performed through a pre-set fuzzy rule base to obtain a fuzzy value of the output variable, and then accurate adjustment instruction is obtained through defuzzification processing.

[0023] Further, the method for analyzing the causes of the deviation in the adaptive control is to establish a multi-physical field coupling model of the welding process, combine the real-time collected welding parameters and weld geometry characteristics, simulate the heat conduction, molten pool flow and arc behavior in the welding process, and thus determine whether the deviation is caused by workpiece assembly error, thermal deformation or arc stability change.

[0024] Further, in the adjustment of the welding parameters in the adaptive control, an incremental adjustment mode is adopted, that is, the adjustment amplitude is a pre-set incremental value each time, the welding parameters are gradually optimized according to the welding effect after adjustment, and the adjustment is stopped until the deviation value is less than the tracking precision threshold.

[0025] Further, the method further comprises a welding quality detection step, in which, after the welding is completed, a non-destructive testing method is used to detect the weld quality; if defects are detected in the weld, the defect position and type are recorded, and the initial welding parameters and the weld tracking control strategy are adjusted according to the defect condition, and the parts with defects are re-welded.

[0026] Further, the non-destructive testing method is any one of ultrasonic testing or ray testing.

[0027] Compared with the prior art, the welding robot weld tracking and adaptive control method provided by the present application has the following beneficial effects:

[0028] 1. The welding robot weld tracking and adaptive control method can accurately obtain the weld center line position, weld width and weld height and other information by using a visual sensor or a laser sensor to collect images or laser reflection signals of the weld area in real time, and using advanced image processing algorithms or signal analysis methods to extract the weld geometry characteristics; the extracted weld geometry characteristics are compared with a pre-set weld model, the deviation of the current position of the welding torch from the ideal weld position is calculated, and an adjustment instruction is generated by using a fuzzy control algorithm or a PID control algorithm according to the deviation value, so that the motion trajectory of the welding robot is adjusted in real time; the influence of workpiece assembly error and thermal deformation and other factors on the weld position is effectively overcome, the accuracy and stability of the weld tracking are greatly improved, the welding torch is ensured to always move along the weld center line, the generation of welding defects is reduced, and the advantages of high accuracy and high stability are achieved.

[0029] 2、The welding robot weld tracking and adaptive control method can analyze the causes of deviation and automatically adjust the corresponding welding parameters according to different causes, can dynamically optimize the welding parameters according to the actual situation of the welding process, can make the welding process always in the best state, thereby guaranteeing the stability and consistency of the welding quality, can better adapt to different welding workpieces and welding environments, improves the versatility and flexibility of the welding robot, and achieves the advantages of strong adaptability.

[0030] 3、The welding robot weld tracking and adaptive control method can effectively remove noise and interference in the image, improve the quality and definition of the image, can provide more accurate and comprehensive information for weld tracking control, and further improves the precision and reliability of weld tracking; by adopting an incremental adjustment mode to adjust the welding parameters, the system can be stabilized to avoid excessive parameter adjustment, the adjustment of the welding parameters is more smooth and stable, and the system can better adapt to the dynamic changes of the welding process, and the control performance and robustness of the system are improved.

[0031] 4、The welding robot weld tracking and adaptive control method uses a non-destructive testing method to detect the weld quality after welding is completed, can timely find and handle welding defects, ensures that the welding quality meets the requirements, reduces the scrap rate, and improves the production efficiency and product quality. BRIEF DESCRIPTION OF DRAWINGS

[0032] Fig. 1 A flowchart of a welding robot weld tracking and adaptive control method according to the present application is shown.

[0033] Fig. 2 A flowchart of an image processing algorithm for extracting weld geometric features in a welding robot weld tracking and adaptive control method according to the present application is shown. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0035] Embodiment 1:

[0036] Please refer to Figs. 1-2 A welding robot weld tracking and adaptive control method according to the present embodiment includes the following steps:

[0037] S1, initial parameter setting: according to the material and thickness of the workpiece to be welded and the welding process requirements, the initial welding parameters of the welding robot are preset, including welding current, welding voltage, welding speed and wire feeding speed; at the same time, the initial search range and tracking accuracy threshold of the weld tracking are set;

[0038] S2, initial positioning of the weld: the starting point of welding is positioned by using a visual sensor or a laser sensor, the three-dimensional coordinate information of the starting point is obtained, and the welding torch of the welding robot is moved to the starting point position;

[0039] S3, weld feature extraction: during welding, the image or laser reflection signal of the weld area is collected in real time by the sensor, and the geometric features of the weld are extracted by using image processing algorithm or signal analysis method, including weld centerline position, weld width and weld height;

[0040] S4, weld tracking control: compare the extracted weld geometric features with the preset weld model, calculate the deviation of the current position of the welding torch from the ideal weld position; according to the deviation value, generate adjustment instructions by using fuzzy control algorithm or PID control algorithm, and adjust the motion trajectory of the welding robot in real time, so that the welding torch always moves along the weld centerline, realizing weld tracking; when the deviation value is less than the tracking accuracy threshold, the current welding parameters and motion trajectory are maintained; when the deviation value is greater than or equal to the tracking accuracy threshold, the adaptive control step is entered;

[0041] S5, adaptive control: analyze the causes of the deviation, and adjust the corresponding welding parameters according to different reasons; if the deviation is caused by workpiece assembly error or thermal deformation, adjust the motion trajectory and welding speed of the welding robot; if the deviation is caused by the change of arc stability during welding, adjust the welding current and welding voltage; the adjusted welding parameters are fed back to the welding robot, and the welding and weld tracking continue until the welding is completed.

[0042] Specifically, in the initial parameter setting, a welding parameter database corresponding to different materials and thicknesses of workpieces is established by experimental data, and the initial welding parameters are obtained from the database according to the material and thickness of the workpiece to be welded.

[0043] Specifically, in the initial positioning of the weld, the visual sensor is any one of CCD camera or CMOS camera, and in the initial positioning of the weld, the laser sensor is any one of laser displacement sensor or laser triangulation sensor.

[0044] Specifically, in the welding seam tracking control, the deviation of the current position of the welding torch from the ideal welding seam position and the rate of change of the deviation are taken as input variables of a fuzzy controller, and the adjustment instruction is taken as an output variable; fuzzy reasoning is performed through a pre-set fuzzy rule base to obtain a fuzzy value of the output variable, and then accurate adjustment instructions are obtained through defuzzification processing.

[0045] Specifically, the method for analyzing the causes of the deviation in the adaptive control is to establish a multi-physical field coupling model of the welding process, combine the real-time collected welding parameters and welding seam geometric features, simulate the heat conduction, molten pool flow and arc behavior in the welding process, and thus determine whether the deviation is caused by workpiece assembly error, thermal deformation or arc stability change.

[0046] Specifically, in the adaptive control, an incremental adjustment method is adopted, that is, the adjustment amplitude is a pre-set incremental value each time, and the welding parameters are gradually optimized according to the welding effect after adjustment until the deviation value is less than the tracking precision threshold.

[0047] In this embodiment, the specific process of extracting the welding seam geometric features by using the image processing algorithm in the welding seam feature extraction is as follows:

[0048] S3-1, pre-process the collected welding seam area image, which includes grayscale, filter denoising and image enhancement;

[0049] S3-2, detect the edge of the welding seam by using an edge detection algorithm.

[0050] S3-3, fit the welding seam center line by using the least square method according to the detected edge, calculate the welding seam center line position, and at the same time, count the number of pixels between the edges, and convert the welding seam width and height by combining the image resolution.

[0051] Specifically, the edge detection algorithm is any one of the Canny algorithm or the Sobel algorithm.

[0052] In this embodiment, it also includes a welding quality detection step, after the welding is completed, the welding quality is detected by using a non-destructive testing method; if defects are detected in the welding seam, the defect position and type are recorded, and the initial welding parameters and the welding seam tracking control strategy are adjusted according to the defect condition, and the parts with defects are re-welded.

[0053] Specifically, the non-destructive testing method is any one of ultrasonic testing or radiographic testing.

[0054] Embodiment 2:

[0055] The welding seam tracking and adaptive control method of the welding robot in this embodiment is applied in the welding of automobile body sheet:

[0056] The automobile body sheet welding system in this embodiment is configured as follows:

[0057] Robot body: FANUC ARC Mate 120iC six-axis welding robot, repeat positioning accuracy ±0.08mm;

[0058] Vision system: Basler ace acA2000-50gm CCD camera + Schott KL1500 LED light source, resolution 2048x1536;

[0059] Laser sensor: SICK OD5000 laser displacement sensor, measurement frequency 1kHz;

[0060] Welding equipment: Panasonic YD-500FR welder, equipped with digital communication interface;

[0061] Control unit: Xilinx Zynq UltraScale+MPSoC FPGA platform.

[0062] In this embodiment, the automobile body sheet welding implementation includes the following steps:

[0063] A1, initial parameter setting:

[0064] A1-1, call welding parameter database (MySQL relational database);

[0065] Among them, the return parameters are current 180A, voltage 22V, speed 45cm / min, and wire feeding speed 8m / min;

[0066] A1-2, set tracking parameters:

[0067] Among them, the search range is ±5mm area on both sides of the weld;

[0068] Among them, the precision threshold is 0.2mm transverse deviation and 0.3mm longitudinal deviation.

[0069] A2, initial positioning of weld:

[0070] A2-1, start laser sensor scanning the starting area (50mmx50mm);

[0071] A2-2, determine the seam center point by peak detection algorithm;

[0072] A2-3, the robot moves to the starting point (X=125.3mm, Y=56.7mm, Z=10.2mm).A3, weld feature extraction:

[0073] A3-1, image preprocessing process:

[0074] Grayscale: Y = 0.299R + 0.587G + 0.114B;

[0075] Filter: 5x5 Gaussian filter (σ = 1.5);

[0076] Enhancement: CLAHE algorithm (clip limit = 2.0);

[0077] A3-2, Canny edge detection:

[0078] Low threshold: 30, high threshold: 90;

[0079] Edge connection distance: 3 pixels;

[0080] A3-3, Geometric feature calculation:

[0081] Centerline: Least square fitting (R 2 > 0.99);

[0082] Width: Edge distance mean ± 3σ to remove outliers.

[0083] A4, Seam tracking control:

[0084] A4-1, Deviation calculation:

[0085] Lateral deviation Δx = 0.15 mm (current value - theoretical value);

[0086] Longitudinal deviation Δz = 0.08 mm;

[0087] A4-2, Fuzzy PID control:

[0088] Input variables: Δx, d(Δx) / dt;

[0089] Output variables: X / Y axis compensation;

[0090] A4-3, Trajectory adjustment:

[0091] Inverse kinematics solution to generate joint angle correction:

[0092] Δθ1 = 0.12°, Δθ2 = -0.08°, Δθ3 = 0.05°.

[0093] A5, Adaptive control:

[0094] A5-1, Multi-physical field simulation (COMSOL Multiphysics): Input real-time parameters: I = 182 A, U = 21.8 V, v = 44 cm / min; Output predicted results: thermal deformation 0.18 mm, weld pool width 4.2 mm; A5-2, Parameter adjustment strategy:

[0095] When Δx persists > 0.3mm:

[0096] Speed adjustment: Δv = -5% x current speed;

[0097] Trajectory compensation: X-axis offset + 0.15mm;

[0098] When melt width > 4.5mm:

[0099] Current adjustment: ΔI = -8A;

[0100] Voltage adjustment: ΔU = -0.5V.

[0101] In this embodiment, the test data is shown in Table 1.

[0102] Table 1

[0103]

[0104] In this embodiment, the economic benefits are shown in Table 2.

[0105] Table 2

[0106]

[0107] Example 3:

[0108] In this embodiment, the application of a welding robot weld tracking and adaptive control method in aerospace aluminum alloy curved surface welding:

[0109] In this embodiment, the difference between the implementation steps of aerospace aluminum alloy curved surface welding and Example 2 is:

[0110] 1. Replace the CCD camera with a blue light laser scanner (LMI Gocator 2350);

[0111] 2. Increase point cloud processing:

[0112] Triangular meshing (MeshLab software);

[0113] Curvature feature extraction (PCA algorithm);

[0114] 3. Control parameter optimization:

[0115] Add surface normal compensation item to fuzzy rule base;

[0116] The thermal deformation compensation coefficient is improved by 30%.

[0117] Example 4:

[0118] In this embodiment, the application of a welding robot weld tracking and adaptive control method in thick plate multi-layer welding:

[0119] In this embodiment, the steps of multi-layer welding of thick plate are different from those of embodiment 2 as follows:

[0120] 1. Welding parameters:

[0121] Base current 280 A, pulse current 350 A;

[0122] Interlayer temperature control at 150±10℃;

[0123] 2. Adaptive strategy:

[0124] Update the heat accumulation model after each layer of welding;

[0125] Optimize the parameters of the next layer using Q-learning algorithm.

[0126] It should be noted that, in this document, the terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply any actual such relationship or order between the entities or operations. Also, the terms "comprises", "comprising", or any other variations thereof are intended to cover non-exclusive inclusions, so that a process, method, article, or apparatus that comprises a list of elements does not only include those elements, but also includes other elements not expressly listed, or other elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the statement "comprises a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0127] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements, and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for weld seam tracking and adaptive control of a welding robot, characterized in that, Includes the following steps: S1. Initial parameter setting: Based on the material, thickness and welding process requirements of the workpiece to be welded, the initial welding parameters of the welding robot are preset, including welding current, welding voltage, welding speed and wire feeding speed; at the same time, the initial search range and tracking accuracy threshold of the weld seam are set. S2. Initial positioning of weld seam: The starting point of welding is located using a vision sensor or laser sensor to obtain the three-dimensional coordinate information of the starting point, and the welding gun of the welding robot is moved to the starting point position. S3. Weld Feature Extraction: During the welding process, images or laser reflection signals of the weld area are collected in real time by sensors. Image processing algorithms or signal analysis methods are used to extract the geometric features of the weld, including the weld centerline position, weld width, and weld height. S4. Weld Seam Tracking Control: The extracted weld seam geometric features are compared with the preset weld seam model to calculate the deviation between the current position of the welding torch and the ideal weld seam position. Based on the deviation value, an adjustment command is generated using a fuzzy control algorithm or a PID control algorithm to adjust the motion trajectory of the welding robot in real time, ensuring that the welding torch always moves along the center line of the weld seam, thus achieving weld seam tracking. When the deviation value is less than the tracking accuracy threshold, the current welding parameters and motion trajectory are maintained. When the deviation value is greater than or equal to the tracking accuracy threshold, the adaptive control step is initiated. S5. Adaptive Control: Analyze the causes of deviations and adjust the corresponding welding parameters according to different causes; if the deviation is due to workpiece assembly errors or thermal deformation causing changes in the weld position, adjust the motion trajectory and welding speed of the welding robot; if the deviation is due to changes in arc stability during welding causing poor weld formation, adjust the welding current and welding voltage; the adjusted welding parameters are fed back to the welding robot to continue welding and weld tracking until welding is completed.

2. The welding robot weld seam tracking and adaptive control method according to claim 1, characterized in that: In the initial parameter setting, a database of welding parameters corresponding to workpieces of different materials and thicknesses is established through experimental data. The initial welding parameters are retrieved from the database based on the material and thickness of the workpiece to be welded.

3. The welding robot weld seam tracking and adaptive control method according to claim 1, characterized in that: The visual sensor used in the initial positioning of the weld seam is either a CCD camera or a CMOS camera, and the laser sensor used in the initial positioning of the weld seam is either a laser displacement sensor or a laser triangulation sensor.

4. The welding robot weld seam tracking and adaptive control method according to claim 1, characterized in that, The specific process of extracting the geometric features of the weld using image processing algorithms is as follows: S3-1. Preprocess the acquired weld area image, including grayscale conversion, filtering and noise reduction, and image enhancement. S3-2. Use an edge detection algorithm to detect the edges of the weld. S3-3. Based on the detected edges, fit the weld centerline using the least squares method and calculate the position of the weld centerline; at the same time, count the number of pixels between the edges and combine the image resolution to obtain the weld width and weld height.

5. The welding robot weld seam tracking and adaptive control method according to claim 4, characterized in that: The edge detection algorithm can be either the Canny algorithm or the Sobel algorithm.

6. The welding robot weld seam tracking and adaptive control method according to claim 1, characterized in that: When generating adjustment commands using a fuzzy control algorithm in the weld seam tracking control, the deviation between the current position of the welding torch and the ideal weld seam position, as well as the rate of change of the deviation, are used as input variables of the fuzzy controller, and the adjustment commands are used as output variables. Fuzzy inference is performed through a pre-set fuzzy rule base to obtain the fuzzy values ​​of the output variables, and then the precise adjustment commands are obtained after defuzzification processing.

7. The welding robot weld seam tracking and adaptive control method according to claim 1, characterized in that: The method for analyzing the causes of deviations in the adaptive control is to establish a multi-physics coupling model of the welding process, combine the welding parameters and weld geometry features acquired in real time, simulate the heat conduction, molten pool flow and arc behavior during the welding process, and thus determine whether the deviation is caused by workpiece assembly error, thermal deformation or changes in arc stability.

8. The welding robot weld seam tracking and adaptive control method according to claim 1, characterized in that: The adaptive control uses an incremental adjustment method when adjusting welding parameters, that is, the adjustment range is a preset increment value each time. The welding parameters are gradually optimized according to the adjusted welding effect until the deviation value is less than the tracking accuracy threshold.

9. The welding robot weld seam tracking and adaptive control method according to claim 1, characterized in that: It also includes a welding quality inspection step. After welding is completed, the quality of the weld is inspected using non-destructive testing methods. If defects are detected in the weld, the location and type of the defects are recorded, and the initial welding parameters and weld tracking control strategy are adjusted according to the defect situation. The defective parts are then repaired by welding.

10. The welding robot weld seam tracking and adaptive control method according to claim 9, characterized in that: The non-destructive testing method can be either ultrasonic testing or radiographic testing.

Images