Intelligent wire feeding equipment control method for welding robot

Abstract

The invention relates to the technical field of control systems, in particular to an intelligent wire feeding equipment control method for a welding robot. The curvature of each point in the actual welding track is calculated, and a curvature sequence of the actual welding track is obtained; performing threshold segmentation on the curvature sequence, and determining a plurality of curvature segments, actual track segments corresponding to the curvature segments and curvature representative values of the actual track segments; determining an influence value corresponding to the curvature representative value in the target historical data set; the wire feeding speed allowable range of each actual track section is determined; the welding direction and the optimal wire feeding speed during actual welding are determined; and intelligent wire feeding equipment of the welding robot is controlled to complete welding. According to the method, through a prediction-compensation thought driven by historical data, the wire feeding speed adjustment amount is given at the track curvature change position in advance, the adjustment delay time can be shortened, the wire feeding action and the welding motion are kept synchronous, welding seam forming is more uniform, and splashing is obviously reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of control systems, and in particular to a control method for intelligent wire feeding equipment of a welding robot. Background Art

[0002] In trajectory operations on complex curved surfaces such as pipeline girth welding, aerospace blades, and spiral surfaces, the trajectory route is relatively complex, and changes in the welding trajectory will significantly affect the wire feeding stability.

[0003] Existing problems: Traditional methods usually rely on real-time feedback to adjust the wire feeding speed. However, due to the dynamic changes in the curvature, angle and direction of the welding path, there is a lag in real-time adjustment, which makes it difficult to accurately match the wire feeding speed with the welding movement, thus affecting the stability of the welding quality. Summary of the Invention

[0004] The present invention provides a control method for an intelligent wire feeding device of a welding robot to solve the existing problems.

[0005] The present invention provides a control method for an intelligent wire feeding device for a welding robot using the following technical solutions: An embodiment of the present invention provides a method for controlling an intelligent wire feeding device for a welding robot, the method comprising: obtaining an actual welding trajectory based on point cloud data of a welding area; calculating the curvature of each point in the actual welding trajectory to obtain a curvature sequence of the actual welding trajectory; performing threshold segmentation on the curvature sequence to determine a plurality of curvature segments, actual trajectory segments corresponding to the curvature segments, and curvature representative values ​​of each of the actual trajectory segments; determining an influence value corresponding to the curvature representative value in a target historical data set; wherein the target historical data set is a historical data set corresponding to the type of welding wire used in actual welding; the influence value is used to characterize the degree of influence of the wire feeding speed of the actual trajectory segment on the welding quality; based on the influence value, determining an allowable range of wire feeding speed for each of the actual trajectory segments; based on the allowable range of wire feeding speed and the influence value, determining the welding direction and the optimal wire feeding speed during actual welding; and based on the welding direction and the optimal wire feeding speed, controlling the intelligent wire feeding device of the welding robot to complete welding.

[0006] Furthermore, the method also includes: dividing the historical welding data into multiple historical data sets based on the type of welding wire; determining the curvature sequence of the historical welding trajectory for each historical welding trajectory in each historical data set; performing threshold segmentation on the curvature sequence to determine multiple curvature segments, historical trajectory segments corresponding to the curvature segments, curvature representative values ​​of the curvature segments, and wire feeding speed representative values ​​of each historical trajectory segment; determining the quality score of the historical weld image corresponding to each historical trajectory segment; and determining the mapping relationship between the curvature representative value, the influence value, and the allowable range of the wire feeding speed based on the curvature representative value, the wire feeding speed representative value, and the quality score.

[0007] Furthermore, the method further comprises: obtaining a training data set for a scoring model; wherein the training data set comprises weld images and their corresponding quality scores; training the scoring model using a mean square error loss function to obtain the trained scoring model; Determining the quality score of the historical weld image corresponding to each historical trajectory segment includes: inputting the historical weld image corresponding to the historical trajectory segment into the trained scoring model, and obtaining the quality score output by the scoring model.

[0008] Furthermore, the mapping relationship between the curvature representative value, the wire feeding speed representative value and the quality score is determined based on the curvature representative value, the influence value and the allowable range of the wire feeding speed, including: for each curvature representative value, determining the coordinates of the binary group consisting of the wire feeding speed representative value and the quality score in the target coordinate system, and obtaining a scatter plot; wherein the target coordinate system is used to represent the horizontal axis as the wire feeding speed representative value and the vertical axis as the coordinate axis of the quality score; based on the scatter plot, calculating a first parameter and a second parameter; wherein the first parameter is used to characterize the sensitivity of the welding quality to changes in the wire feeding speed; the second parameter is used to characterize the linear significance of the relationship between the wire feeding speed and the welding quality; based on the first parameter and the second parameter, determining the influence value; based on the maximum and minimum values ​​of the wire feeding speed representative value in the scatter plot, determining the allowable range of the wire feeding speed; obtaining the mapping relationship between the curvature representative value, the influence value and the allowable range of the wire feeding speed.

[0009] Furthermore, the calculation of the first parameter and the second parameter based on the scatter plot includes: performing principal component analysis on the scatter points in the scatter plot to obtain multiple two-dimensional projection vectors and their corresponding projection values; calculating a first angle value between the projection direction of the two-dimensional projection vector corresponding to the maximum projection value and the horizontal axis, and normalizing the first angle value to obtain the first parameter; calculating a second angle value between the projection direction of the two-dimensional projection vector corresponding to the minimum projection value and the horizontal axis, and normalizing the second angle value to obtain the second parameter.

[0010] Furthermore, the determining of the welding direction and the optimal wire feeding speed during actual welding based on the allowable range of wire feeding speed and the influence value includes: obtaining a speed segment corresponding to each of the actual trajectory segments based on the allowable range of wire feeding speed; performing speed segment merging iterations on the target speed segment until all the speed segments are merged to obtain the updated speed segment; determining the welding direction and the optimal wire feeding speed during actual welding based on the allowable range of wire feeding speed and the influence value of the updated speed segment; Among them, the Nth speed segment merging iteration includes: determining a target speed segment; the target speed segment is the speed segment with a ranking value of N when the influence values ​​in the current speed segment are sorted from large to small; determining an intersection speed segment of the target speed segment; wherein the intersection speed segment is the speed segment in which the wire feeding speed allowable ranges in the previous speed segment and the next speed segment of the target speed segment intersect with the wire feeding speed allowable range of the target speed segment; merging the target speed segment with the intersection speed segment to obtain a merged speed segment; wherein the wire feeding speed allowable range of the merged speed segment is the intersection of the wire feeding speed allowable range of the target speed segment and the wire feeding speed allowable range of the intersection speed segment; the influence value of the merged speed segment is the larger value of the influence value of the target speed segment and the influence value of the intersection speed segment; and the speed segment is updated using the merged speed segment.

[0011] Furthermore, the welding direction during actual welding is determined based on the allowed range of wire feeding speed and the influence value of the updated speed segment, including: determining a speed conversion segment based on the updated speed segment; wherein the speed conversion segment is used to characterize the speed segment that requires wire feeding speed adjustment during actual welding; for each speed turning point in each welding direction, calculating the absolute value of the difference between the influence value of the speed conversion segment and the influence value of the next speed segment; determining the conversion risk value of each speed turning point based on the normalized absolute value of the difference and the influence value of the speed conversion segment; wherein the conversion risk is used to characterize the probability of welding quality defects caused by adjusting the wire feeding speed at the speed turning point; determining the total conversion risk value of all the speed turning points in each welding direction; determining the target welding direction; wherein the target welding direction is the welding direction with the smallest total conversion risk value.

[0012] Furthermore, based on the updated allowable range of wire feeding speed of the speed segment and the influence value, the optimal wire feeding speed during actual welding is determined, including: taking the minimum speed change between adjacent speed segments as the optimization goal, and determining the optimal wire feeding speed of each speed segment in turn along the welding direction.

[0013] Furthermore, based on the welding direction and the optimal wire feeding speed, the intelligent wire feeding device of the welding robot is controlled to complete welding, including: obtaining in real time the current welding gun position of the intelligent wire feeding device in the welding robot and the actual trajectory segment that the welding robot is about to enter; determining the optimal wire feeding speed corresponding to the actual trajectory segment; and dynamically fine-tuning the speed of the wire feeding motor in the intelligent wire feeding device through a closed-loop feedback mechanism to match the actual wire feeding speed with the optimal wire feeding speed.

[0014] Furthermore, the obtaining of the actual welding trajectory based on the welding area point cloud data includes: extracting the center line or characteristic edge of the weld in the welding area point cloud data; and fitting the center line or the characteristic edge into the actual welding trajectory.

[0015] The beneficial effects of the technical solution of the present invention are: In an embodiment of the present invention, an actual welding trajectory is obtained based on point cloud data of the welding area; the curvature of each point in the actual welding trajectory is calculated to obtain a curvature sequence of the actual welding trajectory; the curvature sequence is threshold segmented to determine multiple curvature segments, actual trajectory segments corresponding to the curvature segments, and curvature representative values ​​of each actual trajectory segment; in a target historical data set, an influence value corresponding to the curvature representative value is determined; based on the influence value, an allowable range of wire feeding speed for each actual trajectory segment is determined; based on the allowable range of wire feeding speed and the influence value, the welding direction and the optimal wire feeding speed during actual welding are determined; based on the welding direction and the optimal wire feeding speed, the intelligent wire feeding device of the welding robot is controlled to complete welding. At this point, the present invention uses a prediction-compensation idea driven by historical data to give the wire feeding speed adjustment amount in advance at the point where the trajectory curvature changes, which is beneficial to reducing the adjustment lag time, making the wire feeding action synchronized with the welding movement, making the weld formation more uniform, and significantly reducing spatter; on the other hand, the Otsu multi-threshold segmentation and PCA quantified wire feeding speed-quality scoring relationship are used to automatically identify trajectory segments that are sensitive to speed, and prioritize locking the speed of key segments based on k-value sorting, reducing unnecessary frequent acceleration and deceleration, reducing mechanical impact, and improving the welding quality of the entire weld; on the other hand, risk value assessment and two-way path comparison are introduced when determining the final speed sequence, and the welding direction with the smoothest change is selected through dynamic programming or greedy algorithm, so as to achieve a gentle wire feeding transition at the speed turning point, and further suppress defects such as biting and depression caused by sudden speed changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 A schematic structural diagram of an intelligent wire feeding device provided in an embodiment of the present invention; Figure 2 A schematic flow chart of a method for controlling an intelligent wire feeding device for a welding robot provided in an embodiment of the present invention; Figure 3 A flowchart of a method for obtaining a historical data set according to an embodiment of the present invention; Figure 4 A schematic diagram of scattered point distribution in a wire feeding speed-quality scoring coordinate system provided by an embodiment of the present invention; Figure 5 A schematic diagram of speed segment division when the welding direction is the first direction provided by an embodiment of the present invention; Figure 6A schematic diagram of speed segment division when the welding direction is the second direction provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a method for controlling an intelligent wire feeding device for a welding robot according to the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0020] Before introducing the above-mentioned control method of the intelligent wire feeding device for the welding robot, the welding robot and the intelligent wire feeding device of the welding robot involved in the method are first introduced: Welding robots: A welding robot is a mechatronic device designed for industrial welding applications that can automatically complete welding operations according to preset programs or real-time instructions. It typically consists of core components such as a robotic arm, a control cabinet, a welding gun (or laser head), a wire feeder, a sensor system, and a welding power supply. The robotic arm features multi-axis linkage, allowing for flexible position adjustment in three dimensions. The control cabinet parses the program and drives the motors on each axis. The welding gun or laser head directly melts the workpiece and welding wire. During wire feeding, the high energy of the welding gun or laser head causes the wire to melt rapidly, allowing droplets of molten metal to precisely drip onto the weld, thereby improving weld quality and joint strength. The wire feeder stably delivers the wire according to process requirements. A sensor system (such as a laser scanner, visual camera, and current and voltage sensor) monitors weld position, weld pool status, and arc characteristics in real time. The welding power supply provides the required current and voltage. Under the control of software algorithms and hardware closed-loop, the entire system can operate continuously along complex trajectories to ensure consistent weld formation and stable penetration depth, thereby replacing manual labor to complete welding tasks that require high repeatability, high strength, or extremely high precision. It is widely used in automotive manufacturing, shipbuilding, aerospace, pipeline engineering and other fields.

[0021] Intelligent wire feeding equipment: please refer to Figure 1The operating principle of intelligent wire feeding equipment in a certain application scenario is as follows: the welding wire in the wire barrel is fed to the main wire feeder via the main control pulley wire feed tube. During the welding process, the wire end sensor monitors the end position of the welding wire in real time to ensure that the welding wire is correctly delivered to the welding area. The main control unit coordinates the work of each component, controls the operation of the main wire feeder according to the welding parameters, and receives signals from the main wire feeder through the signal amplifier line to adjust the feeding speed and position of the welding wire. The robot control cabinet is responsible for controlling the movement of the robot, enabling the laser head to weld according to the preset trajectory, while maintaining communication with the main control unit to ensure the coordination and synchronization of the welding process. This intelligent wire feeding equipment can automatically adjust the feeding of welding wire according to the requirements of the welding task, ensuring the consistency and stability of welding quality.

[0022] The specific scheme of the control method of the intelligent wire feeding device for the welding robot provided by the present invention is described in detail below with reference to the accompanying drawings.

[0023] See also Figure 2 , which shows a method for controlling an intelligent wire feeding device for a welding robot provided by one embodiment of the present invention, comprising: Step S110: Acquire the actual welding trajectory based on the welding area point cloud data.

[0024] Preferably, in one embodiment of the present invention, the above step S110 includes: extracting the center line or characteristic edge of the weld in the point cloud data of the welding area; and fitting the center line or characteristic edge into the actual welding trajectory.

[0025] It should be noted that the above-mentioned welding area point cloud data can be obtained through equipment such as a laser line scanner or a structured light projector. The scheme of extracting the weld center line or feature edge through point cloud processing and the scheme of fitting the center line or feature edge to the actual welding trajectory are both relatively mature and well-known technologies. For specific implementation methods, please refer to the relevant technology, and the embodiments of the present invention will not be repeated here.

[0026] Step S120: Calculate the curvature of each point in the actual welding trajectory to obtain the curvature sequence of the actual welding trajectory.

[0027] It should be noted that the above-mentioned curvature is a measure of the degree of bending severity of the trajectory curve at a certain point. The curvature of a straight line is zero, and the smaller the radius of the arc, the greater the curvature. The curvature value at a sharp turn or corner is larger. For a welding robot, the greater the curvature of the trajectory curve, the more the welding gun needs to change direction or posture quickly within a very short distance, which will directly affect the flow of the molten pool, the stability of the arc, and the wire feeding resistance. Therefore, the change in curvature becomes an important basis for adjusting process parameters such as wire feeding speed, welding speed, and laser power. For example: the welding speed in the high curvature area is reduced, and the wire feeding amount needs to be reduced to avoid accumulation of deposited metal; while the welding speed in the low curvature area is increased, and the wire feeding amount needs to be increased to prevent weld depression. If not adjusted in time, it will lead to defects such as uneven weld formation and spatter. It can be understood that curvature is a relatively mature and well-known technology. For its specific implementation method, please refer to the relevant technology, and the embodiments of the present invention will not be repeated.

[0028] Step S130: performing threshold segmentation on the curvature sequence to determine a plurality of curvature segments, actual trajectory segments corresponding to the curvature segments, and curvature representative values ​​of each actual trajectory segment.

[0029] It should be noted that step S130 can employ the Otsu multi-threshold segmentation method to perform threshold segmentation on the curvature sequence. The Otsu multi-threshold segmentation method is a generalization of the classic Otsu algorithm, primarily used to automatically divide a continuous data stream (e.g., a curvature sequence) into several "segments" such that the values ​​within a segment are as close as possible and the values ​​between different segments are as different as possible. It is understood that the Otsu multi-threshold segmentation method is a relatively mature and well-known technique. For its specific implementation, please refer to the related art and will not be further described in detail in the present embodiment.

[0030] It should be further noted that in practical applications, an empirically reasonable range can be set for the number of segments (e.g., 3 to 5 segments) based on the complexity of the welding trajectory to avoid over-segmentation. It is understood that the final segmentation result (the final number of segments and the boundaries of each segment) is automatically calculated by the Otsu multi-threshold segmentation method based on the distribution characteristics of the input curvature sequence, rather than relying on fixed parameters. This approach ensures the universality and adaptability of the Otsu multi-threshold segmentation method for welding trajectories of different shapes.

[0031] Step S140: Determine the influence value corresponding to the curvature representative value in the target historical data set; wherein the target historical data set is a historical data set corresponding to the type of welding wire used in actual welding; the influence value is used to characterize the degree of influence of the wire feeding speed of the actual trajectory segment on the welding quality.

[0032] The following describes the solution for obtaining historical data sets: See also Figure 3Preferably, in one embodiment of the present invention, the above-mentioned intelligent wire feeding device control method for a welding robot further includes: Step S210: Divide the historical welding data into multiple historical data sets based on the type of welding wire.

[0033] It should be noted that each historical data set contains multiple data items, each of which contains: Track image: the image of the welding area obtained by the industrial camera after welding is completed.

[0034] Trajectory Curve: Trajectory data is a curve formed by the welding trajectory of each welding operation in history. For example, if the welding path is a straight line segment, the trajectory data is also a line segment of the same length. This data is obtained by using a laser line scanner or structured light projector to acquire 3D point cloud data of the weld surface. Point cloud processing is used to extract the weld centerline or characteristic edges (such as the ridgeline of a V-groove) and fit them into an abstract trajectory (such as a V-shaped broken line).

[0035] Step S220: for each historical welding trajectory in each historical data set, determine the curvature sequence of the historical welding trajectory.

[0036] Step S230: performing threshold segmentation on the curvature sequence to determine a plurality of curvature segments, historical trajectory segments corresponding to the curvature segments, curvature representative values ​​of the curvature segments, and wire feeding speed representative values ​​of each historical trajectory segment.

[0037] It should be noted that the purpose of the aforementioned intelligent wire feeding device control method for a welding robot is to analyze historical welding data to determine a speed adjustment plan for the current welding process. This method takes into account the significant differences in deposition characteristics, thermal conductivity, and wire feeding resistance between different materials, resulting in the material-dependence of the process parameters required for the same welding trajectory, namely, the wire feeding speed. During the welding process, when the trajectory curvature changes (such as when a straight line turns into an arc), the wire feeding speed must be adjusted accordingly. For example, in areas of high curvature, the welding speed decreases, requiring a reduction in wire feeding to prevent deposited metal accumulation; while in areas of low curvature, the welding speed increases, requiring an increase in wire feeding to prevent weld concavity. Failure to make timely adjustments can lead to defects such as uneven weld formation and spatter.

[0038] Therefore, the historical welding data first needs to be divided into different data sets according to the type of welding wire. Then, for the trajectory curve in each data set, the curvature of each point on the curve is calculated to obtain a curvature sequence. The curvature sequence is segmented using the Otsu multi-threshold method to obtain multiple curvature segments, with similar curvature values ​​in the same curvature segment. For each curvature segment, the absolute value of the difference between each curvature value and all other curvature values ​​in the sequence is calculated. The minimum absolute value of the difference and the corresponding curvature value are used as the representative curvature value of the curvature segment. The wire feeding speed curve corresponding to each curvature segment is then obtained, and the speed value with the highest frequency in the curve is used as the wire feeding speed for the corresponding curvature.

[0039] It should be further explained that the selection criteria for the above curvature representative value can be: select the point in the curvature segment that has the smallest difference from all other curvature points in the segment, that is, the "most representative center point". The specific calculation basis is as follows: (1) For a certain curvature segment, it contains m curvature values: ; (2) Calculate each curvature value within the segment To all other curvature values ​​within the segment The sum of the absolute distances (i.e. the sum of the absolute differences) of : ,in, ,and ; (3) Compare all value, select The smallest curvature value , which is the representative value of the curvature of this segment.

[0040] The above curvature representative value can reflect the point with the “smallest comprehensive difference” or the “most central” point in the segment, thereby reflecting the overall curvature characteristics of the segment to the greatest extent.

[0041] Step S240: Determine the quality score of the historical weld image corresponding to each historical trajectory segment.

[0042] Preferably, in one embodiment of the present invention, the intelligent wire feeding device control method for a welding robot further includes: obtaining a training dataset for a scoring model; wherein the training dataset includes weld images and their corresponding quality scores; and training the scoring model using a mean square error loss function to obtain a trained scoring model. Step S240 includes inputting historical weld images corresponding to historical trajectory segments into the trained scoring model to obtain a quality score output by the scoring model. For example, in this embodiment, for each weld segment corresponding to a curvature segment (the region on the weld corresponding to the curvature segment), an image of that region is obtained and recorded as the weld image corresponding to the curvature segment. A scoring neural network is trained using a mean square error loss function. The training dataset includes multiple welding images, and the welding conditions are labeled 0-1 by experts. The trained neural network is trained using a mean square error loss function to obtain a trained scoring neural network. Each weld image of the curvature segment is input into the scoring neural network to obtain a quality score. Segments with a score less than 0.8 in each dataset are removed, thereby obtaining multiple (wire feeding speed, quality score) pairs corresponding to each curvature. It is understandable that multiple curvatures can be obtained on each trajectory curve, and different trajectory curves may have the same curvature, but the corresponding wire feeding speeds are not necessarily the same.

[0043] It should be further explained that the above-mentioned scoring model can adopt a neural network. The architecture and training scheme of the neural network are relatively mature and well-known technologies. For its specific implementation method, please refer to the relevant technology, and the embodiments of the present invention will not be repeated.

[0044] Step S250: Based on the curvature representative value, the wire feeding speed representative value and the quality score, a mapping relationship between the curvature representative value, the impact value and the allowable range of the wire feeding speed is determined.

[0045] Preferably, in one embodiment of the present invention, the above-mentioned step S250 includes: for each curvature representative value, determining the coordinates of the binary group consisting of the wire feeding speed representative value and the quality score in the target coordinate system, and obtaining a scatter plot; wherein, the target coordinate system is used to represent the coordinate axis with the horizontal axis being the wire feeding speed representative value and the vertical axis being the quality score; based on the scatter plot, calculating the first parameter and the second parameter; wherein, the first parameter is used to characterize the sensitivity of the welding quality to the change in the wire feeding speed; the second parameter is used to characterize the linear significance of the relationship between the wire feeding speed and the welding quality; based on the first parameter and the second parameter, determining the influence value; based on the maximum and minimum values ​​of the wire feeding speed representative value in the scatter plot, determining the allowable range of the wire feeding speed; obtaining the mapping relationship between the curvature representative value, the influence value and the allowable range of the wire feeding speed.

[0046] It should be noted that the implementation method of establishing the mapping relationship between the curvature representative value and the allowable range of wire feeding speed in the above scheme may include: (1) Data collection: Collect a large amount of historically successful welding process data. Each piece of data includes: welding trajectory, curvature of each point on the trajectory, and wire feeding speed values ​​corresponding to each point on the trajectory, which have been verified to be high-quality in practice.

[0047] (2) Curvature segmentation and representative value extraction: For each historical trajectory, the curvature is segmented according to the above-mentioned Otsu multi-threshold segmentation method, and a curvature representative value is determined for each curvature segment.

[0048] (3) Statistical speed range: For the same curvature representative value (allowing slight deviation), find all wire feeding speed values ​​corresponding to the representative value in all historical data to form a speed set.

[0049] (4) Determine the allowable range: Perform statistical analysis on the speed set (e.g. calculate the mean ±3 , or directly take the range from the minimum to the maximum value), determine the allowable range of wire feeding speed corresponding to the representative value of curvature .

[0050] (5) Forming a mapping table / function: Repeat the above process to establish the corresponding wire feed speed allowable range for all different curvature representative values, and finally form a "curvature representative value - wire feed speed allowable range" mapping query table or fit a mapping function. During actual welding, the system can quickly determine the upper and lower limits of the wire feed speed that should be used based on the curvature representative value calculated in real time by looking up the table or calculating the function.

[0051] Preferably, in one embodiment of the present invention, the above-mentioned calculation of the first parameter and the second parameter based on the scatter plot includes: performing principal component analysis on the scatter points in the scatter plot to obtain multiple two-dimensional projection vectors and their corresponding projection values; calculating the first angle value between the projection direction of the two-dimensional projection vector corresponding to the maximum projection value and the horizontal axis, and normalizing the first angle value to obtain the first parameter; calculating the second angle value between the projection direction of the two-dimensional projection vector corresponding to the minimum projection value and the horizontal axis, and normalizing the second angle value to obtain the second parameter.

[0052] It should be noted that: since there may be a situation where one wire feeding speed corresponds to multiple quality scores, the embodiment of the present invention does not use similarity calculation, but instead uses the method of constructing a coordinate system to represent the binary group, and then analyzes and obtains the influence relationship between wire feeding speed and quality. For each curvature value in each data set, a corresponding wire feeding speed-quality score coordinate system can be constructed to obtain the corresponding coordinate points of each (wire feeding speed, quality score) binary group to form a scatter plot. All the coordinate points in the scatter plot are used as input to the principal component analysis PCA algorithm to obtain multiple two-dimensional projection vectors and the projection value corresponding to each vector. The projection vector represents the projection direction, and the corresponding projection value represents the projection length of all data points in the corresponding projection direction. The larger the projection length, the wider the distribution range of the data points in this direction. Specifically: See also Figure 4 In the wire feeding speed-quality score coordinate system, the horizontal axis represents the wire feeding speed and the vertical axis represents the quality score. Figure 4 The oblique line in represents the projection direction corresponding to the maximum projection value. If the angle between the maximum and minimum projection values ​​approaches 90°, the data distribution of the coordinate points approaches a linear shape, and the relationship between wire feed speed and welding quality under this curvature is more clearly defined. If the maximum projection value approaches 90°, a smaller change in wire feed speed leads to a larger change in wire feed quality under this curvature, meaning that welding quality is more strongly affected by wire feed speed. The inverse tangent angle of the ratio of the second element to the first element in the two-dimensional projection vector corresponding to the maximum projection value is denoted as a, where a is the angle between the projection direction corresponding to the maximum projection value and the horizontal coordinate in the coordinate system. The angle b corresponding to the minimum projection value is obtained using the same method. The acute angle between these two angles is obtained in the coordinate system, denoted as c. The ratio of c to 90° is denoted as d. A larger ratio d indicates a more clear relationship between welding quality and wire feed speed under this curvature, meaning that a larger change in wire feed speed is likely to cause a change in welding quality. The ratio of a to 90° is denoted as a0. A larger a0 indicates a greater impact of wire feed speed on welding quality at that curvature. For each curvature value in each dataset, the corresponding d and a0 are obtained, and the product of these two, k, is used as the degree of influence of wire feed speed on welding quality at that curvature, also known as the impact value. A larger K value indicates that higher wire feed speed is required to achieve higher welding quality at that curvature, or in other words, high wire feed quality can only be achieved within a smaller wire feed speed range.

[0053] It should be further explained that the principal component analysis (PCA) algorithm is a relatively mature and well-known technology. For its specific implementation, please refer to the relevant technology, and the embodiments of the present invention will not be described in detail.

[0054] Step S150: Based on the impact value, determine the allowable range of wire feeding speed for each actual trajectory segment.

[0055] What needs to be explained is that before actual welding is performed, the type of welding wire used in the actual welding is first obtained, and then a data set of the corresponding type of welding wire in the historical welding data is obtained; the welding trajectory corresponding to the actual welding is obtained, that is, an abstract representation of the weld shape, which is recorded as the actual trajectory. The actual trajectory is calculated using the same method to obtain multiple curvature segments, which are recorded as actual trajectory segments. For each actual trajectory segment, a curvature representative value is calculated, and then the influence value k corresponding to the same curvature representative value in the historical data is obtained. The larger the k value, the higher the welding quality of the actual trajectory segment, and the stricter the requirements on the speed range of the wire feeding speed. In addition, the above step S250 determines the mapping relationship between the curvature representative value, the influence value, and the allowable range of the wire feeding speed. After the influence value is determined in the above step S140, the allowable range of the wire feeding speed for each actual trajectory segment can be determined by the above mapping relationship.

[0056] Step S160: Determine the welding direction and the optimal wire feeding speed during actual welding based on the allowable range of the wire feeding speed and the impact value.

[0057] Preferably, in one embodiment of the present invention, the above-mentioned step S160 includes: based on the allowable range of wire feeding speed, obtaining the speed segment corresponding to each actual trajectory segment; performing speed segment merging iteration on the target speed segment until all speed segments are merged to obtain the updated speed segment; based on the allowable range of wire feeding speed and the influence value of the updated speed segment, determining the welding direction and the optimal wire feeding speed during actual welding.

[0058] Among them, the Nth speed segment merging iteration includes: determining the target speed segment; the target speed segment is the speed segment with a ranking value of N when the influence values ​​in the current speed segment are sorted from large to small; determining the intersection speed segment of the target speed segment; wherein the intersection speed segment is a speed segment in which the wire feeding speed allowable ranges in the previous speed segment and the next speed segment of the target speed segment intersect with the wire feeding speed allowable range of the target speed segment; merging the target speed segment and the intersection speed segment to obtain a merged speed segment; wherein the wire feeding speed allowable range of the merged speed segment is the intersection of the wire feeding speed allowable range of the target speed segment and the wire feeding speed allowable range of the intersection speed segment; the influence value of the merged speed segment is the larger value between the influence value of the target speed segment and the influence value of the intersection speed segment; and updating the speed segment using the merged speed segment.

[0059] Preferably, in one embodiment of the present invention, the above-mentioned step S160 includes: determining a speed conversion segment based on the updated speed segment; wherein the speed conversion segment is used to characterize the speed segment that requires wire feeding speed adjustment during the actual welding process; for each speed turning point in each welding direction, calculating the absolute value of the difference between the influence value of the speed conversion segment and the influence value of the next speed segment; based on the normalized absolute value of the difference and the influence value of the speed conversion segment, determining the conversion risk value of each speed turning point; wherein the conversion risk is used to characterize the probability of welding quality defects caused by adjusting the wire feeding speed at the speed turning point; determining the total conversion risk value of all speed turning points in each welding direction; determining the target welding direction; wherein the target welding direction is the welding direction with the smallest total conversion risk value.

[0060] What needs to be explained is: considering that the frequent acceleration and deceleration of the wire feeding speed of the welding wire can easily lead to unstable welding effects, the embodiment of the present invention can first find the actual trajectory segment with strict speed requirements, and determine the speed of each actual trajectory segment in order from large to small according to the strictness of the speed requirements, thereby obtaining a smaller number of wire feeding speed changes, and thus obtaining higher quality welding results. The weld has two endpoints: endpoint A and endpoint B. Since the wire feeding speed needs to be adjusted from one speed segment to another, the speed adjustment is different for different welding directions. For example, the speed adjustment for welding from left to right is different from the speed adjustment for welding from right to left. Then, starting welding from which endpoint can obtain the least speed change can result in a more stable and higher quality welding result. Specifically: Each trajectory segment corresponds to an allowable range of wire feed speeds, denoted as a speed segment. First, the actual trajectory segment with the largest k value is found. The intersection of the wire feed speeds of this segment with the left and right segments is calculated. If there is an intersection, the intersection speed is obtained and merged with the speed segment on that side to form a merged speed segment. Next, the actual trajectory segment with the second-largest k value is found. The intersection of the wire feed speeds of this segment with the left and right segments is calculated. If there is an intersection, the intersection speed is obtained and merged with the speed segment on that side to form a merged speed segment. This iterative process is repeated until all speed segments are merged. For each merged speed segment, the larger k value of the two merged trajectory segments is used as the k value of the merged speed segment. The larger the k value, the higher the priority for determining the speed value of this segment, thereby achieving higher welding quality. This results in an updated k value for each speed segment. Speed ​​segments with larger k values ​​are prioritized in the speed adjustment scheme, and the speeds of other segments are then adjusted based on the speed values ​​of this segment. The merged speed segment is included in the subsequent speed segments.

[0061] Assume that the direction of welding from point A to point B is recorded as the first direction, and the direction of welding from point B to point A is recorded as the second direction. When moving from one speed segment to another along the first direction, the wire feed speed must be adjusted within the current segment to avoid problems with wire feeding and quality issues caused by the turning point structure when passing through the turning point at a speed that is not suitable for the next segment.

[0062] See also Figure 5 and Figure 6 , Figure 5 and Figure 6 The speed segments corresponding to the first and second welding directions are shown in the figure, where the different gray values ​​in the figure represent the influence of the wire feeding speed on the quality in each speed segment, and the darker the gray value, the greater the influence. Figure 5 For example, along this welding direction, to minimize significant risks when passing through the transition point between two speed segments, the speed conversion process is relocated to the previous segment. The light-colored squares below the numbered blocks in the figure represent the speed conversion process. The speed segment in which the square is located indicates the speed segment in which the conversion from the current speed segment to the next speed segment is completed. When transitioning from one speed segment to another, there is a significant risk of poor weld quality at the transition point between the two segments due to inappropriate wire feed speed. Therefore, when calculating quality stability for different welding directions, the present invention uses the difference in k values ​​between the two speed segments to calculate the potential risk at the transition point. Taking the trajectory path in the first direction as an example, the absolute value of the difference between the k values ​​of each speed transition segment (the speed segment in which the speed conversion occurs) and the next speed segment is calculated. This difference is then normalized to obtain s. The smaller s, the lower the probability of quality risk at the transition point. Taking the trajectory path in the first direction as an example, the risk value of each turning point is expressed by the product of the k value of the previous speed conversion segment and the s value of the two speed segments corresponding to the turning point. Then, the risk values ​​of all turning points can be obtained, and the sum value is calculated as the total conversion risk value of the trajectory path in the first direction, that is, when welding according to the trajectory path in the first direction, the degree of influence of the corresponding wire feed speed conversion on the welding quality, or the probability of welding quality defects caused by adjusting the wire feed speed at the speed turning point. The same method can be used to obtain the total conversion risk value of the other reverse trajectory path. The direction corresponding to the smaller of the two conversion risk total values ​​is used as the welding direction in actual welding. When welding according to this welding direction, the corresponding wire feed speed conversion has less influence on the welding quality.

[0063] Preferably, in one embodiment of the present invention, step S160 includes determining the optimal wire feeding speed for each speed segment along the welding direction, with the goal of minimizing the speed variation between adjacent speed segments. This embodiment, for example, employs a dynamic programming algorithm or a greedy algorithm, with the goal of minimizing the speed variation between adjacent speed segments, to sequentially determine the optimal wire feeding speed value for each speed segment starting from a starting point, ensuring that the speed is always within the allowable range and that the overall variation is smooth. The resulting speed sequence can adapt to the wire feeding requirements of different curvature segments while avoiding drastic speed fluctuations, thereby ensuring the stability and quality of the welding process.

[0064] It should be noted that the sum of squares of adjacent speed differences can be used as the objective function of the above dynamic programming algorithm or greedy algorithm. The objective function can be expressed as:

[0065] in, and Respectively speed range and Optimal wire feeding speed for each speed segment; is the total amount of the speed segment; Constraints when solving for the optimal wire feed speed can include: (1) Velocity boundary constraint:

[0066] in, The minimum value allowed for wire feeding speed, in m / s, can be set to 0.1; The maximum allowable wire feeding speed is in m / s and its value can be set to 2.0.

[0067] (2) Smoothness constraint:

[0068] in, The maximum variation allowed between adjacent speed segments is in m / s and its value can be set to 0.15; (3) Iteration convergence criteria: The iterative convergence criterion can be set to the value change of the objective function for two consecutive iterations being less than the preset objective function change threshold, that is, the following conditions are met twice in a row:

[0069] in, is the number of iterations; For the The objective function value calculated at the iteration; To preset the objective function change threshold, its value can be set to 0.0001; The iterative convergence criterion can also be set to the maximum number of iterations, that is:

[0070] in, is the maximum number of iterations.

[0071] It should be noted that the above-mentioned dynamic programming algorithm and greedy algorithm are both relatively mature and well-known technologies. For their specific implementation methods, please refer to the relevant technologies, and the embodiments of the present invention will not be described in detail.

[0072] Step S170: Based on the welding direction and the optimal wire feeding speed, the intelligent wire feeding device of the welding robot is controlled to complete the welding.

[0073] Preferably, in one embodiment of the present invention, step S170 includes: obtaining in real time the current welding gun position of the intelligent wire feeding device in the welding robot and the actual trajectory segment the welding robot is about to enter; determining the optimal wire feeding speed corresponding to the actual trajectory segment; and dynamically fine-tuning the speed of the wire feeding motor in the intelligent wire feeding device through a closed-loop feedback mechanism to ensure that the actual wire feeding speed matches the optimal wire feeding speed. For example, in this embodiment, during the actual welding control process, the system implements precise control based on pre-determined trajectory segments (speed segments) and their corresponding optimal wire feeding speed parameters through the following steps: First, the welding control system reads the current welding gun position and the trajectory segment about to be entered in real time, and calls the preset wire feeding speed parameters for that segment; then, through a closed-loop feedback mechanism (such as an encoder monitoring the actual wire feeding speed and a current sensor detecting the arc state), the wire feeding motor speed is dynamically fine-tuned to ensure that the actual wire feeding speed matches the theoretical value; Simultaneously, the system predicts the trajectory curvature change trend in advance and initiates a smooth transition of the wire feeding speed at the end of the previous trajectory segment to avoid speed step changes.

[0074] It should be noted that during the actual welding process, welding quality data (such as molten pool shape and spatter) will be continuously recorded throughout the entire welding process for subsequent optimization of speed segment division and wire feeding parameters, forming a control closed loop for continuous improvement.

[0075] It should be further clarified that the aforementioned closed-loop feedback mechanism is a control strategy that achieves precise control by monitoring system output in real time and comparing it with a desired value to adjust the input. It is understood that the closed-loop feedback mechanism is a relatively mature and well-known technology. Its specific implementation is described in related art and will not be further described in detail in the present embodiments.

[0076] So far, the present invention is completed.

[0077] To summarize, in an embodiment of the present invention, an actual welding trajectory is obtained based on point cloud data of the welding area; the curvature of each point in the actual welding trajectory is calculated to obtain a curvature sequence of the actual welding trajectory; the curvature sequence is threshold segmented to determine multiple curvature segments, actual trajectory segments corresponding to the curvature segments, and curvature representative values ​​of each actual trajectory segment; in a target historical data set, an influence value corresponding to the curvature representative value is determined; based on the influence value, an allowable range of wire feeding speed for each actual trajectory segment is determined; based on the allowable range of wire feeding speed and the influence value, the welding direction and the optimal wire feeding speed during actual welding are determined; based on the welding direction and the optimal wire feeding speed, the intelligent wire feeding device of the welding robot is controlled to complete welding. The present invention uses a prediction-compensation approach driven by historical data to provide wire feeding speed adjustment amounts in advance at the point where the trajectory curvature changes, which is beneficial to reducing the adjustment lag time, keeping the wire feeding action synchronized with the welding movement, making the weld formation more uniform, and significantly reducing spatter; on the other hand, the Otsu multi-threshold segmentation and PCA quantified wire feeding speed-quality scoring relationship are used to automatically identify trajectory segments that are sensitive to speed, and prioritize locking the speed of key segments based on k-value sorting, reducing unnecessary frequent acceleration and deceleration, thereby reducing mechanical impact and improving the welding quality of the entire weld; on the other hand, risk value assessment and two-way path comparison are introduced when determining the final speed sequence, and the welding direction with the smoothest change is selected through dynamic programming or greedy algorithm, so as to achieve a gentle wire feeding transition at the speed turning point, and further suppress defects such as biting and depression caused by sudden speed changes.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for controlling an intelligent wire feeding device for a welding robot, characterized in that: The method comprises: Based on the point cloud data of the welding area, the actual welding trajectory is obtained; Calculating the curvature of each point in the actual welding trajectory to obtain a curvature sequence of the actual welding trajectory; performing threshold segmentation on the curvature sequence to determine a plurality of curvature segments, actual trajectory segments corresponding to the curvature segments, and curvature representative values ​​of the actual trajectory segments; Determining an influence value corresponding to the curvature representative value in a target historical data set; wherein the target historical data set is a historical data set corresponding to the type of welding wire used in actual welding; and the influence value is used to characterize the degree of influence of the wire feeding speed of the actual trajectory segment on the welding quality; Determining an allowable range of wire feeding speed for each of the actual trajectory segments based on the impact value; Determining a welding direction and an optimal wire feeding speed during actual welding based on the allowable range of wire feeding speed and the impact value; Based on the welding direction and the optimal wire feeding speed, the intelligent wire feeding device of the welding robot is controlled to complete the welding.

2. The intelligent wire feeding device control method for a welding robot according to claim 1, characterized in that: The method further comprises: Based on the type of welding wire, the historical welding data is divided into a plurality of historical data sets; For each historical welding trajectory in each of the historical data sets, determining a curvature sequence of the historical welding trajectory; Performing threshold segmentation on the curvature sequence to determine a plurality of curvature segments, historical trajectory segments corresponding to the curvature segments, curvature representative values ​​of the curvature segments, and wire feeding speed representative values ​​of each of the historical trajectory segments; Determining a quality score of the historical weld image corresponding to each of the historical trajectory segments; Based on the curvature representative value, the wire feeding speed representative value and the quality score, a mapping relationship among the curvature representative value, the impact value and the allowable range of the wire feeding speed is determined.

3. The intelligent wire feeding device control method for a welding robot according to claim 2, characterized in that: The method further comprises: Obtaining a training data set for a scoring model; wherein the training data set includes weld images and their corresponding quality scores; The scoring model is trained using a mean square error loss function to obtain a trained scoring model; Determining the quality score of the historical weld image corresponding to each historical trajectory segment includes: The historical weld image corresponding to the historical trajectory segment is input into the trained scoring model to obtain the quality score output by the scoring model.

4. The intelligent wire feeding device control method for a welding robot according to claim 2, characterized in that: The determining, based on the curvature representative value, the wire feeding speed representative value, and the quality score, a mapping relationship between the curvature representative value, the impact value, and the allowable range of the wire feeding speed includes: For each curvature representative value, determining the coordinates of a two-tuple consisting of the wire feeding speed representative value and the quality score in a target coordinate system, and obtaining a scatter plot; wherein the target coordinate system is used to represent the horizontal axis as the wire feeding speed representative value and the vertical axis as the coordinate axis of the quality score; Based on the scatter plot, a first parameter and a second parameter are calculated; wherein the first parameter is used to characterize the sensitivity of the welding quality to the change of the wire feeding speed; and the second parameter is used to characterize the linear significance of the relationship between the wire feeding speed and the welding quality; determining the impact value based on the first parameter and the second parameter; Determining the allowable range of the wire feeding speed based on the maximum and minimum values ​​of the wire feeding speed representative values ​​in the scatter plot; A mapping relationship among the curvature representative value, the influence value, and the allowable range of the wire feeding speed is obtained.

5. The intelligent wire feeding device control method for a welding robot according to claim 4, characterized in that: The calculating the first parameter and the second parameter based on the scatter plot includes: Performing principal component analysis on the scattered points in the scatter plot to obtain a plurality of two-dimensional projection vectors and their corresponding projection values; Calculating a first angle between the projection direction of the two-dimensional projection vector corresponding to the maximum projection value and the abscissa axis, and normalizing the first angle to obtain a first parameter; A second angle value between the projection direction of the two-dimensional projection vector corresponding to the minimum projection value and the horizontal coordinate axis is calculated, and the second angle value is normalized to obtain a second parameter.

6. The intelligent wire feeding device control method for a welding robot according to claim 1, characterized in that: The determining of the welding direction and the optimal wire feeding speed during actual welding based on the allowable range of the wire feeding speed and the influence value includes: Based on the allowable range of wire feeding speed, obtaining the speed segment corresponding to each actual trajectory segment; Performing speed segment merging iterations on the target speed segment until all the speed segments are merged, and obtaining the updated speed segment; Determining a welding direction and an optimal wire feeding speed during actual welding based on the updated allowable wire feeding speed range of the speed section and the influence value; The Nth speed segment merging iteration includes: Determine a target speed segment; the target speed segment is the speed segment with a ranking value of N when the influence values ​​in the current speed segment are sorted from large to small; Determine an intersection speed segment of the target speed segment; wherein the intersection speed segment is the speed segment in which the allowable wire feeding speed ranges in the speed segment preceding and succeeding the target speed segment intersect with the allowable wire feeding speed range of the target speed segment; Merging the target speed segment and the intersection speed segment to obtain a merged speed segment; wherein the allowable wire feeding speed range of the merged speed segment is the intersection of the allowable wire feeding speed range of the target speed segment and the allowable wire feeding speed range of the intersection speed segment; and the influence value of the merged speed segment is the larger value of the influence value of the target speed segment and the influence value of the intersection speed segment; The speed segment is updated using the merged speed segment.

7. The intelligent wire feeding device control method for a welding robot according to claim 6, characterized in that: The determining of the welding direction during actual welding based on the updated allowable wire feeding speed range of the speed segment and the influence value includes: Based on the updated speed segment, a speed conversion segment is determined; wherein the speed conversion segment is used to represent a speed segment that requires wire feeding speed adjustment during an actual welding process; For each speed turning point in each welding direction, calculating the absolute value of the difference between the influence value of the speed conversion segment and the influence value of the next speed segment; Determining a conversion risk value for each speed transition point based on the normalized absolute value of the difference and the impact value of the speed transition segment; wherein the conversion risk is used to characterize the probability of welding quality defects resulting from adjusting the wire feeding speed at the speed transition point; Determining a total conversion risk value of all the speed turning points in each of the welding directions; Determine a target welding direction; wherein the target welding direction is the welding direction with the smallest total conversion risk value.

8. The intelligent wire feeding device control method for a welding robot according to claim 6, characterized in that: Determining the optimal wire feeding speed during actual welding based on the updated allowable wire feeding speed range of the speed section and the impact value includes: Taking minimizing the speed variation between adjacent speed segments as an optimization goal, the optimal wire feeding speed of each speed segment is determined in sequence along the welding direction.

9. The intelligent wire feeding device control method for a welding robot according to any one of claims 1 to 8, characterized in that: The method of controlling the intelligent wire feeding device of the welding robot to complete welding based on the welding direction and the optimal wire feeding speed includes: Real-time acquisition of the current welding gun position of the intelligent wire feeding device in the welding robot and the actual trajectory segment that the welding robot is about to enter; Determining the optimal wire feeding speed corresponding to the actual trajectory segment; Through a closed-loop feedback mechanism, the rotational speed of the wire feeding motor in the intelligent wire feeding device is dynamically fine-tuned so that the actual wire feeding speed matches the optimal wire feeding speed.

10. The intelligent wire feeding device control method for a welding robot according to any one of claims 1 to 8, characterized in that: The method of obtaining the actual welding trajectory based on the welding area point cloud data includes: Extracting the center line or characteristic edge of the weld from the point cloud data of the welding area; The center line or the feature edge is fitted as an actual welding trajectory.

Images