Intelligent control system and method for welding equipment based on multivariate information

Abstract

The present invention relates to the technical field of intelligent control of welding equipment, and discloses an intelligent control system and method for welding equipment based on multivariate information, which collects multivariate information in a welding process in real time; extracts features of the multivariate information to generate multivariate feature information; fuses the multivariate feature information to form complete welding process information; performs intelligent analysis and diagnosis on the fused multivariate feature information, compares the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, warns of abnormal conditions in the welding process, realizes intelligent warning, and can avoid abnormal conditions in the welding process; according to the result of the intelligent analysis, if abnormal welding occurs, dynamically controls the welding parameters of the welding equipment, optimizes the welding path according to actual conditions, and executes a correction strategy at the same time until there is no abnormal welding in the welding process, so as to achieve the best welding effect and improve welding quality, welding efficiency and stability.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of welding equipment, and in particular to a welding equipment intelligent control system and method based on multi-information. Background Art

[0002] In recent years, with the development of technologies such as artificial intelligence, big data and the Internet of Things, intelligent welding equipment has gradually become a development trend in the welding industry. By integrating data collected by sensors, cameras and other equipment, all-round monitoring and control of the welding process can be achieved, and problems in the welding process can be discovered in time, improving welding quality, reducing manual intervention in the welding process, and reducing labor costs. However, most of them do not dynamically control the welding parameters of the welding equipment, optimize the welding path and implement correction strategies after judging the welding process as abnormal welding.

[0003] For example, the patent with publication number CN116629543A discloses a method, device and electronic equipment for intelligent comprehensive management and control of coal mines, which involves the field of data processing. In this method, applied to a server, the method includes: obtaining a first project engineering, the first project engineering being any one of a plurality of project engineerings of a coal mine; judging whether the first project engineering is a preset risk item according to preset conditions; if the first project engineering meets the preset conditions, determining that the first project engineering is a preset risk item; judging whether the current progress of the first project engineering has reached the preset first risk point based on preset rules; if the current progress of the first project engineering has not reached the preset first risk point, determining that there is a risk of delay in the first project engineering; sending warning information to a user device so that the user corresponding to the user device can view the warning information, and the warning information includes the first project engineering with a risk of delay. It is convenient for coal mine managers to control coal mines according to the warning information.

[0004] For example, the Chinese patent with the authorization announcement number CN117094539B discloses a method, system, device and storage medium for intelligent power supply work order management and control, which generates a task work order according to the preset inspection plan and user data, and obtains the task location, equipment complexity coefficient, task importance coefficient, execution team requirement coefficient, meteorological event coefficient and building complexity coefficient of the corresponding task work order; based on the distance between the task location and the task execution terminal; the team parameters of each task execution terminal that meets the preset task execution conditions are substituted into the preset relationship between the task matching degree and the equipment complexity coefficient, task importance coefficient, execution team requirement coefficient, meteorological event coefficient and building complexity coefficient, and the task work order is issued to the task execution terminal with the highest matching degree. It can automatically issue power supply task work orders to the power supply team, and promptly notify the power supply team to inspect the hidden dangers of on-site equipment and check the content of power consumption.

[0005] The above patents have the problems raised by this background technology: the above coal mine intelligent comprehensive management and control method determines whether the first project is a preset risk item and whether the current progress of the first project has reached the preset first risk point through preset conditions and preset rules. When there is a risk of delay, it is necessary to send an early warning message to the user device; the above power supply intelligent work order management and control method generates a task work order through a preset investigation plan and user data, and sends the task work order to a task execution terminal with the highest matching degree according to the preset nearby task execution conditions. These two patents do not have the problem of dynamically controlling the welding parameters of the welding equipment and optimizing the welding path and executing the correction strategy after judging the welding process as abnormal welding. To solve this problem, the present invention proposes a welding equipment intelligent management and control system and method based on multivariate information. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] In view of the above-mentioned problems existing in the existing intelligent control system and method for welding equipment based on multi-information, the present invention is proposed.

[0008] Therefore, the object of the present invention is to provide a welding equipment intelligent control system and method based on multi-information.

[0009] In order to solve the above technical problems, the present invention provides a welding equipment intelligent control method based on multi-information:

[0010] Real-time collection of multiple information during welding process;

[0011] Extracting features of multivariate information to generate multivariate feature information;

[0012] Fusion of multi-dimensional feature information to form complete welding process information;

[0013] Intelligently analyze and diagnose the fused multivariate feature information, compare the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, and warn of abnormal conditions in the welding process;

[0014] According to the results of intelligent analysis, if abnormal welding occurs, the welding parameters of the welding equipment will be dynamically controlled, and the welding path will be optimized according to the actual situation.

[0015] As a preferred solution of the welding equipment intelligent control method based on multivariate information described in the present invention, the multivariate feature information generation strategy is as follows:

[0016] The features of multivariate information are constructed into a feature graph. The number of edges directly connected to each node is counted. Secondly, the number of shortest paths connecting each pair of nodes is counted. Then, the number of paths passing through meta-nodes in the shortest path of each pair of nodes is counted. Finally, the importance of meta-nodes is calculated and sorted in descending order of importance to generate multivariate feature information.

[0017] As a preferred solution of the welding equipment intelligent control method based on multivariate information described in the present invention, the function expression of the importance of the meta-node is as follows:

[0018]

[0019] In the formula, s v represents the importance of meta-node v, c ab (v) represents the number of paths that pass through the meta-node v in the shortest path between each pair of nodes (a, b), c ab represents the number of shortest paths connecting node a and node b, a≠v≠b means that each pair of nodes (a, b) and meta-node v are not the same node in the feature graph;

[0020] The number of shortest paths between each pair of nodes is counted as follows:

[0021] First, an initial node is selected and all other nodes are marked as unvisited; then the adjacent nodes connected to the initial node are visited, and the distance value between the initial node and the adjacent nodes is continuously updated. If the path from the initial node to the adjacent node is shorter than the known path, the distance value of the adjacent node is updated and the updated adjacent node is marked as visited; finally, the statistics are stopped until all nodes are marked as visited.

[0022] As a preferred solution of the intelligent control method of welding equipment based on multivariate information described in the present invention, the strategy for early warning of abnormal welding is as follows:

[0023] Each group of fused multivariate feature information is recorded as a sample to form a sample set. The sample set is clustered and divided into normal welding and abnormal welding. The distribution density of each group of samples is calculated respectively, and a density threshold is set. The distribution density of each group of samples is compared with the density threshold. If the distribution density of the sample is less than the density threshold, it indicates abnormal welding and an abnormal situation in the welding process needs to be warned. If the distribution density of the sample is greater than or equal to the density threshold, it indicates normal welding.

[0024] As a preferred solution of the intelligent control method of welding equipment based on multivariate information described in the present invention, the functional expression of the distribution density of the sample is as follows:

[0025]

[0026] In the formula, d represents the distribution density of the sample, n represents the number of multivariate feature information in the sample, and K h represents the kernel function, h represents the bandwidth parameter, which is used to control the width of the kernel function, x represents the normal multivariate feature information, and x i Represents the i-th multivariate feature information in the sample.

[0027] As a preferred solution of the intelligent control method of welding equipment based on multivariate information described in the present invention, if abnormal welding occurs, the welding parameters of the welding equipment are dynamically controlled, and the dynamic control strategy is as follows:

[0028] If the temperature fluctuation during welding is large, adjust the welding current;

[0029] If the pressure fluctuation during welding is large, adjust the welding voltage;

[0030] If the speed fluctuation during welding is large, adjust the welding current and welding voltage;

[0031] If the weld length, weld width and weld shape are abnormal during the welding process, the welding speed and welding current are adjusted, the welding path is optimized, and the correction strategy is implemented;

[0032] If the workpiece deforms greatly during welding, adjust the welding speed and welding current.

[0033] As a preferred solution of the intelligent control method of welding equipment based on multivariate information described in the present invention, the strategy for optimizing the welding path is as follows:

[0034] First, the coordinates of the welding points, the shape of the workpiece, and the material information are collected; then a welding path planning strategy is generated, with the shortest path as the goal, to obtain the optimal welding path; finally, the generated path is simulated and verified;

[0035] The welding path planning strategy is as follows:

[0036] Generate workpiece contours and welding points based on workpiece shape and material information, and use corresponding welding equipment to calculate the shortest welding path of the welding points, and perform welding in sequence according to the shortest welding path;

[0037] The function expression of the shortest welding path is as follows:

[0038]

[0039] Where minf represents the shortest welding path, e represents the welding point, m represents the total number of welding points on the path, and (pe, qe) represents the coordinates of the e-th welding point on the welding path.

[0040] As a preferred solution of the intelligent control method of welding equipment based on multivariate information described in the present invention, the correction strategy is as follows:

[0041] After adjusting the welding speed and welding current and optimizing the welding path, continue to judge the abnormal welding situation in the welding process. If there is no abnormal welding in the welding process, it means that the strategy of adjusting the welding speed and welding current and optimizing the welding path is effective, and the correction strategy ends;

[0042] If there is abnormal welding during welding, correct the welding speed and welding current;

[0043] If there is no abnormal welding during the welding process after the welding speed and welding current are corrected, it means that the welding speed and welding current are adjusted incorrectly;

[0044] If abnormal welding occurs during the welding process after the welding speed and welding current are corrected, continue to correct the shortest welding path;

[0045] If there is no abnormal welding during the welding process after the shortest welding path is corrected, it means that the shortest welding path is calculated incorrectly;

[0046] If abnormal welding exists in the welding process after the shortest welding path is corrected, the welding speed, welding current and the shortest welding path are corrected simultaneously until there is no abnormal welding in the welding process, and the correction strategy is terminated;

[0047] The function expression for correcting the shortest welding path is as follows:

[0048]

[0049] Where λ represents the weight coefficient of the smoothness penalty.

[0050] As a preferred solution of the intelligent control method of welding equipment based on multi-dimensional information described in the present invention, the multi-dimensional information includes temperature, pressure, speed, welding time, weld morphology and workpiece deformation.

[0051] An intelligent control system for welding equipment based on multivariate information, comprising: a data acquisition module, a feature extraction module, a feature fusion module, an intelligent analysis module and a dynamic control module;

[0052] The data acquisition module is used to collect multivariate information during the welding process in real time;

[0053] The feature extraction module is used to extract the features of multivariate information and generate multivariate feature information;

[0054] The feature fusion module is used to fuse multivariate feature information to form complete welding process information;

[0055] The intelligent analysis module is used to perform intelligent analysis and diagnosis on the fused multivariate feature information, compare the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, and warn of abnormal conditions in the welding process;

[0056] The dynamic control module is used to dynamically control the welding parameters of the welding equipment according to the results of intelligent analysis if abnormal welding occurs, and optimize the welding path according to the actual situation.

[0057] The beneficial effects of the present invention are as follows: the present invention collects multivariate information in the welding process in real time; extracts the features of the multivariate information to generate multivariate feature information; fuses the multivariate feature information to form complete welding process information to ensure the accuracy of subsequent intelligent analysis; intelligently analyzes and diagnoses the fused multivariate feature information, compares the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, warns of abnormal conditions in the welding process, and realizes intelligent warning, which can avoid abnormal conditions in the welding process, reduce the number of welding times and material waste, and thus reduce welding costs; according to the results of the intelligent analysis, if abnormal welding occurs, the welding parameters of the welding equipment are dynamically controlled, and the welding path is optimized according to the actual situation, and the correction strategy is executed at the same time until there is no abnormal welding in the welding process, so as to achieve the best welding effect and improve the welding quality, welding efficiency and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. 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 creative labor. Among them:

[0059] Figure 1 A method flow chart of a method for intelligent control of welding equipment based on multivariate information according to the present invention;

[0060] Figure 2 It is a characteristic diagram of a welding equipment intelligent control method based on multivariate information of the present invention;

[0061] Figure 3 A PCB board outline diagram of a welding equipment intelligent control method based on multivariate information of the present invention;

[0062] Figure 4 A correction strategy flow chart of a welding equipment intelligent control method based on multivariate information according to the present invention;

[0063] Figure 5This is a system structure diagram of an intelligent management and control system for welding equipment based on multi-information according to the present invention. DETAILED DESCRIPTION

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0065] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0066] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0067] Example 1

[0068] In this embodiment, a method flow chart of a method for intelligent management and control of welding equipment based on multivariate information is provided, such as Figure 1 As shown, a method for intelligent control of welding equipment based on multivariate information includes:

[0069] This method is applied to wave soldering or reflow soldering to solder the high-temperature melted liquid tin to the component plug-ins of the PCB board.

[0070] S1. Real-time collection of multi-dimensional information during welding.

[0071] The multivariate information includes temperature, pressure, speed, welding time, weld morphology and workpiece deformation.

[0072] In specific applications, sensors are used to collect the temperature, pressure, speed and welding time of the welding process in real time, and cameras are used to collect the weld morphology and workpiece deformation. At the same time, the multi-dimensional information is stored in the form of blocks on the cloud platform to achieve real-time traceability of the multi-dimensional information.

[0073] S2. Extract the features of multivariate information and generate multivariate feature information for subsequent intelligent analysis and diagnosis.

[0074] The strategy for generating multivariate feature information is as follows:

[0075] The features of multivariate information are constructed into a feature graph. The number of edges directly connected to each node is counted. Secondly, the number of shortest paths connecting each pair of nodes is counted. Then, the number of paths passing through meta-nodes in the shortest path of each pair of nodes is counted. Finally, the importance of meta-nodes is calculated and sorted in descending order of importance to generate multivariate feature information.

[0076] The function expression of the importance of a meta node is as follows:

[0077]

[0078] In the formula, s v represents the importance of meta-node v, c ab (v) represents the number of paths that pass through the meta-node v in the shortest path between each pair of nodes (a, b), c ab represents the number of shortest paths connecting node a and node b, and a≠v≠b means that each pair of nodes (a, b) and meta-node v are not the same node in the feature graph.

[0079] The number of shortest paths between each pair of nodes is counted as follows:

[0080] First, an initial node is selected and all other nodes are marked as unvisited; then the adjacent nodes connected to the initial node are visited, and the distance value between the initial node and the adjacent nodes is continuously updated. If the path from the initial node to the adjacent node is shorter than the known path, the distance value of the adjacent node is updated and the updated adjacent node is marked as visited; finally, the statistics are stopped until all nodes are marked as visited.

[0081] It needs to be explained that each node represents a feature of multivariate information, and the connection between nodes represents the correlation or interaction between the features of multivariate information.

[0082] In specific applications, the collected multivariate information is preprocessed, including cleaning, missing value processing and outlier detection of the multivariate information, to ensure the quality and integrity of the multivariate information. After preprocessing, the features of the multivariate information are extracted, and temperature features, pressure features, speed features, welding time features, weld morphology features and workpiece deformation features are extracted respectively. The temperature features include average temperature, maximum temperature, temperature change rate and temperature fluctuation degree. The pressure features include average pressure, maximum pressure, pressure change rate and pressure fluctuation degree. The speed features include average speed, maximum speed, speed change rate and speed fluctuation degree. The welding time features include total welding time, welding duration and welding time distribution. The weld morphology features include weld length, weld width and weld shape. The workpiece deformation features include workpiece deformation degree and workpiece deformation position.

[0083] The features of multivariate information are constructed into feature maps, such as Figure 2 As shown in the figure, one node represents a feature of multivariate information, and the connection between nodes indicates the correlation or interaction between the features of multivariate information. There are 20 features of multivariate information here, indicating that there are 20 nodes in the feature graph. Because the features of multivariate information are correlated or interact with each other, there will be connections between nodes. The number of edges directly connected to each node indicates the importance of the node in the entire feature graph. The more directly connected edges there are, the more important the node is in the entire feature graph, that is, the importance of the feature of the multivariate information.

[0084] Secondly, count the number of shortest paths connecting each pair of nodes;

[0085] Then, the number of paths passing through the meta-node in the shortest path of each pair of nodes is counted, and finally the importance of the meta-node is calculated by the formula. The two nodes in each pair of nodes are not the same node as the meta-node. The calculation results are sorted in descending order of importance to generate multivariate feature information. Here we can know that the multivariate feature information includes the degree of temperature fluctuation, the degree of pressure fluctuation, the degree of speed fluctuation, welding duration, weld length, weld width, weld shape and degree of workpiece deformation. The generated multivariate feature information is prepared for subsequent intelligent analysis and diagnosis.

[0086] S3. Fusion of multi-dimensional feature information to form complete welding process information to ensure the accuracy of subsequent intelligent analysis.

[0087] In specific applications, eight kinds of multivariate feature information, including temperature fluctuation degree, pressure fluctuation degree, speed fluctuation degree, welding duration, weld length, weld width, weld shape and workpiece deformation degree, are weightedly fused. The importance of each multivariate feature information is set according to the actual situation to fully understand the status of the welding process.

[0088] S4. Perform intelligent analysis and diagnosis on the fused multivariate feature information, compare the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, and warn of abnormal conditions in the welding process.

[0089] The strategy for early warning of abnormal welding is as follows:

[0090] Each group of fused multivariate feature information is recorded as a sample to form a sample set. The sample set is clustered and divided into normal welding and abnormal welding. The distribution density of each group of samples is calculated respectively, and a density threshold is set. The distribution density of each group of samples is compared with the density threshold. If the distribution density of the sample is less than the density threshold, it indicates abnormal welding and an abnormal situation in the welding process needs to be warned. If the distribution density of the sample is greater than or equal to the density threshold, it indicates normal welding.

[0091] The function expression of the sample distribution density is as follows:

[0092]

[0093] In the formula, d represents the distribution density of the sample, n represents the number of multivariate feature information in the sample, and K h represents the kernel function, h represents the bandwidth parameter, which is used to control the width of the kernel function, x represents the normal multivariate feature information, and x i Represents the i-th multivariate feature information in the sample.

[0094] What needs to be explained is: kernel function K h It is calculated by calculating the normal multivariate feature information x and the i-th multivariate feature information x in the sample i The difference between , returns a non-negative weight value, and the size of the weight value depends on the i-th multivariate feature information x in the sample i The distance from the normal multivariate feature information x, if the i-th multivariate feature information x in the sample i If the i-th multivariate feature information x in the sample is close to the normal multivariate feature information x, the weight value is larger. i The farther away from the normal multivariate feature information x, the smaller the weight value; at the same time, the bandwidth parameter h controls the width of the kernel function. A smaller bandwidth parameter will result in a larger weight value returned by the kernel function only when it is very close to the normal multivariate feature information x. A larger bandwidth parameter will result in the i-th multivariate feature information x in the sample. i When the kernel function is not very close to the normal multivariate feature information x, the weight value returned by the kernel function will also be larger.

[0095] In specific applications, based on the fused multivariate feature information, each group of fused multivariate feature information is recorded as a group of samples to form a sample set, and the multivariate feature information is clustered and divided into normal welding and abnormal welding. The distribution density of the multivariate feature information is calculated respectively, and the distribution density of the multivariate feature information is compared with the density threshold. When the distribution density of a group of multivariate feature information is less than the density threshold, the group of multivariate feature information is marked as an abnormal group. At this time, the corresponding node in the feature graph will be highlighted, and the abnormal degree of the corresponding multivariate feature information will be displayed. For example, if the temperature fluctuation degree in the welding process is large, then the corresponding node representing the temperature fluctuation degree in the feature graph will be highlighted and the abnormal degree will be displayed as 2. If the temperature fluctuation degree in the welding process is small, then the corresponding node representing the temperature fluctuation degree in the feature graph will be highlighted and the abnormal degree will be displayed as 1, indicating that under the welding operation with the welding parameters in this group of multivariate feature information, it is abnormal welding, and there is a problem with the welding quality. It is necessary to warn of abnormal conditions in the welding process. When the distribution density of a group of multivariate feature information is greater than or equal to the density threshold, it means that it is normal welding without abnormal conditions, and intelligent warning of the welding process is realized.

[0096] S5. According to the results of intelligent analysis, if abnormal welding occurs, the welding parameters of the welding equipment are dynamically controlled, and the welding path is optimized according to the actual situation to achieve the best welding effect and improve the welding quality and stability.

[0097] If abnormal welding occurs, the welding parameters of the welding equipment are dynamically controlled. The dynamic control strategy is as follows:

[0098] If the temperature fluctuation during welding is large, adjust the welding current;

[0099] If the pressure fluctuation during welding is large, adjust the welding voltage;

[0100] If the speed fluctuation during welding is large, adjust the welding current and welding voltage;

[0101] If the weld length, weld width and weld shape are abnormal during the welding process, the welding speed and welding current are adjusted, the welding path is optimized, and the correction strategy is implemented;

[0102] If the workpiece deforms greatly during welding, adjust the welding speed and welding current.

[0103] What needs to be explained is: increasing the welding current can increase the stability of the molten pool and reduce temperature fluctuations; adjusting the welding voltage can improve arc stability and reduce pressure fluctuations; appropriately increasing the welding current and welding voltage can increase the welding energy, thereby adapting to changes in welding speed; slowing down the welding speed can increase the welding time and make the weld shape more uniform, while appropriately increasing the welding current can increase the size of the molten pool and fill the weld; slowing down the welding speed can reduce heat input and reduce the risk of workpiece deformation, while increasing the welding current can increase the welding speed and reduce the welding time, thereby reducing the possibility of workpiece deformation.

[0104] The strategy for optimizing the welding path is as follows:

[0105] First, the coordinates of the welding points, the shape of the workpiece and the material information are collected; then a welding path planning strategy is generated, with the shortest path as the goal, to obtain the optimal welding path; finally, the generated path is simulated and verified.

[0106] The welding path planning strategy is as follows:

[0107] The workpiece contour and welding points are generated according to the workpiece shape and material information, and the corresponding welding equipment is used to calculate the shortest welding path of the welding points, and welding is performed in sequence according to the shortest welding path.

[0108] The function expression of the shortest welding path is as follows:

[0109]

[0110] Where minf represents the shortest welding path, e represents the welding point, m represents the total number of welding points on the path, (p e ,q e ) represents the coordinates of the e-th welding point on the welding path.

[0111] The correction strategy is as follows:

[0112] After adjusting the welding speed and welding current and optimizing the welding path, continue to judge the abnormal welding situation in the welding process. If there is no abnormal welding in the welding process, it means that the strategy of adjusting the welding speed and welding current and optimizing the welding path is effective, and the correction strategy ends;

[0113] If there is abnormal welding during welding, correct the welding speed and welding current;

[0114] If there is no abnormal welding during the welding process after the welding speed and welding current are corrected, it means that the welding speed and welding current are adjusted incorrectly;

[0115] If abnormal welding occurs during the welding process after the welding speed and welding current are corrected, continue to correct the shortest welding path;

[0116] If there is no abnormal welding during the welding process after the shortest welding path is corrected, it means that the shortest welding path is calculated incorrectly;

[0117] If abnormal welding exists in the welding process after the shortest welding path is corrected, the welding speed, welding current and the shortest welding path are corrected at the same time until there is no abnormal welding in the welding process, and the correction strategy is terminated.

[0118] The function expression for correcting the shortest welding path is as follows:

[0119]

[0120] Where λ represents the weight coefficient of the smoothness penalty.

[0121] In specific applications, optimizing the welding path first requires collecting welding point coordinates, workpiece shape and material information, and then generating a welding path planning strategy. Using CAD software to generate a PCB board outline based on the PCB board, the PCB board outline is as follows: Figure 3As shown in the figure, the components of the PCB board are welded with liquid tin. The component positions of the PCB board are the welding points. The relevant parameters of the welding equipment are determined according to the PCB board outline and the welding points, including the working mode and welding speed of the welding equipment. The shortest path is used as the objective function to obtain the optimal welding path. Finally, the generated path is simulated and verified to ensure that abnormal welding will not occur in the path, resulting in deformation of the workpiece. It is determined whether there is abnormal welding in the welding process within 5 seconds after the optimization of the welding path. If there is abnormal welding, the correction strategy is executed. The correction strategy process is as follows Figure 4 As shown, when the welding speed, welding current and the shortest welding path are corrected at the same time and abnormal welding still exists, it means that the weld length, weld width and weld shape are not the main reasons for abnormal welding in the welding process. At this time, it is necessary to consider whether the abnormal welding is caused by the large deformation of the workpiece, and the welding speed and welding current need to be corrected; the correction of the welding speed and welding current includes reducing the welding speed and increasing the welding current to reduce the weld width, wherein the correction of the shortest welding path is to add a curvature penalty term on the basis of the original formula to consider the smoothness of the welding path, and the weight coefficient λ of the smoothness penalty needs to be adjusted according to actual needs, such as the working mode of the welding equipment, the shape of the workpiece and the welding requirements. For example, when high-precision welding is required, a larger λ value is required to ensure the smoothness of the welding path;

[0122] When the workpiece deforms greatly during welding, slowing down the welding speed reduces the risk of workpiece deformation, and increasing the welding current reduces the possibility of workpiece deformation.

[0123] Example 2

[0124] In this embodiment, a system structure diagram of a welding equipment intelligent management and control system based on multi-information is provided, such as Figure 5 As shown, an intelligent management and control system for welding equipment based on multivariate information includes a data acquisition module, a feature extraction module, a feature fusion module, an intelligent analysis module and a dynamic control module.

[0125] The data acquisition module is used to collect multivariate information during the welding process in real time;

[0126] The feature extraction module is used to extract the features of multivariate information and generate multivariate feature information;

[0127] The feature fusion module is used to fuse multivariate feature information to form complete welding process information;

[0128] The intelligent analysis module is used to perform intelligent analysis and diagnosis on the fused multivariate feature information, compare the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, and warn of abnormal conditions in the welding process;

[0129] The dynamic control module is used to dynamically control the welding parameters of the welding equipment according to the results of intelligent analysis if abnormal welding occurs, and optimize the welding path according to the actual situation;

[0130] The above-mentioned intelligent control system for welding equipment based on multi-information refers to the content of an intelligent control method for welding equipment based on multi-information, which will not be repeated here.

[0131] Example 3

[0132] In this embodiment, a computer device is provided, including a memory and a processor, the memory is used to store instructions, and the processor is used to execute the instructions, so that the computer device executes the steps of implementing the above-mentioned intelligent management and control method of welding equipment based on multi-dimensional information.

[0133] Example 4

[0134] In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed, the steps of the above-mentioned intelligent management and control method of welding equipment based on multi-dimensional information are implemented.

[0135] The computer-readable storage medium includes: a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and other media for storing program codes.

[0136] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A welding equipment intelligent control method based on multivariate information, characterized in that: include: Real-time collection of multiple information during welding process; Extracting features of multivariate information to generate multivariate feature information; The strategy for generating multivariate feature information is as follows: The features of multivariate information are constructed into a feature graph, and the number of edges directly connected to each node is counted. Then, the number of shortest paths connecting each pair of nodes is counted. Then, the number of paths passing through meta-nodes in the shortest path of each pair of nodes is counted. Finally, the importance of meta-nodes is calculated and sorted in descending order according to importance to generate multivariate feature information. Fusion of multi-dimensional feature information to form complete welding process information; Intelligently analyze and diagnose the fused multivariate feature information, compare the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, and warn of abnormal conditions in the welding process; According to the results of intelligent analysis, if abnormal welding occurs, the welding parameters of the welding equipment are dynamically controlled, and the welding path is optimized according to the actual situation; The strategy for optimizing the welding path is as follows: First, the coordinates of the welding points, the shape of the workpiece, and the material information are collected; then a welding path planning strategy is generated, with the shortest path as the goal, to obtain the optimal welding path; finally, the generated path is simulated and verified; The welding path planning strategy is as follows: Generate workpiece contours and welding points based on workpiece shape and material information, and use corresponding welding equipment to calculate the shortest welding path of the welding points, and perform welding in sequence according to the shortest welding path; The function expression of the shortest welding path is as follows: ； In the formula, represents the shortest path for welding, Indicates welding point, Represents the total number of welding points on the path, Indicates the first The coordinates of the welding points; The correction strategy is as follows: After adjusting the welding speed and welding current and optimizing the welding path, continue to judge the abnormal welding situation in the welding process. If there is no abnormal welding in the welding process, it means that the strategy of adjusting the welding speed and welding current and optimizing the welding path is effective, and the correction strategy ends; If there is abnormal welding during welding, correct the welding speed and welding current; If there is no abnormal welding during the welding process after the welding speed and welding current are corrected, it means that the welding speed and welding current are adjusted incorrectly; If abnormal welding occurs during the welding process after the welding speed and welding current are corrected, continue to correct the shortest welding path; If there is no abnormal welding during the welding process after the shortest welding path is corrected, it means that the shortest welding path is calculated incorrectly; If abnormal welding exists in the welding process after the shortest welding path is corrected, the welding speed, welding current and the shortest welding path are corrected simultaneously until there is no abnormal welding in the welding process, and the correction strategy is terminated; The function expression for correcting the shortest welding path is as follows: ； In the formula, Weight coefficient representing the smoothness penalty.

2. The intelligent control method for welding equipment based on multivariate information according to claim 1, characterized in that: The function expression of the importance of a meta node is as follows: ； In the formula, Represents a meta node The importance of Indicates that for each pair of nodes The shortest path passes through the meta node The number of paths, Represents a connected node and nodes The number of shortest paths between Represents each pair of nodes and meta nodes Not the same node in the feature graph; The number of shortest paths between each pair of nodes is counted as follows: First, an initial node is selected and all other nodes are marked as unvisited; then the adjacent nodes connected to the initial node are visited, and the distance value between the initial node and the adjacent nodes is continuously updated. If the path from the initial node to the adjacent node is shorter than the known path, the distance value of the adjacent node is updated and the updated adjacent node is marked as visited; finally, the statistics are stopped until all nodes are marked as visited.

3. The intelligent control method for welding equipment based on multivariate information as claimed in claim 2, characterized in that: The strategy for early warning of abnormal welding is as follows: Each group of fused multivariate feature information is recorded as a sample to form a sample set. The sample set is clustered and divided into normal welding and abnormal welding. The distribution density of each group of samples is calculated respectively, and a density threshold is set. The distribution density of each group of samples is compared with the density threshold. If the distribution density of the sample is less than the density threshold, it indicates abnormal welding and an abnormal situation in the welding process needs to be warned. If the distribution density of the sample is greater than or equal to the density threshold, it indicates normal welding.

4. The intelligent control method for welding equipment based on multivariate information as claimed in claim 3, characterized in that: The function expression of the sample distribution density is as follows: ； In the formula, represents the distribution density of the sample, Represents the amount of multivariate feature information in the sample, represents the kernel function, Represents the bandwidth parameter, which is used to control the width of the kernel function. represents normal multivariate feature information, Indicates the sample Multi-dimensional feature information.

5. The intelligent control method for welding equipment based on multivariate information according to claim 4, characterized in that: If abnormal welding occurs, the welding parameters of the welding equipment are dynamically controlled. The dynamic control strategy is as follows: If the temperature fluctuation during welding is large, adjust the welding current; If the pressure fluctuation during welding is large, adjust the welding voltage; If the speed fluctuation during welding is large, adjust the welding current and welding voltage; If the weld length, weld width and weld shape are abnormal during the welding process, the welding speed and welding current are adjusted, the welding path is optimized, and the correction strategy is implemented; If the workpiece deforms greatly during welding, adjust the welding speed and welding current.

6. The intelligent control method for welding equipment based on multivariate information according to claim 5, characterized in that: The multivariate information includes temperature, pressure, speed, welding time, weld morphology and workpiece deformation.

7. An intelligent control system for welding equipment based on multi-information, used to implement an intelligent control method for welding equipment based on multi-information according to any one of claims 1 to 6, characterized in that: include: Data acquisition module, feature extraction module, feature fusion module, intelligent analysis module and dynamic control module; The data acquisition module is used to collect multivariate information during the welding process in real time; The feature extraction module is used to extract the features of multivariate information and generate multivariate feature information; The feature fusion module is used to fuse multivariate feature information to form complete welding process information; The intelligent analysis module is used to perform intelligent analysis and diagnosis on the fused multivariate feature information, compare the distribution density of the multivariate feature information to determine whether the fused multivariate feature information is normal welding or abnormal welding, and warn of abnormal conditions in the welding process; The dynamic control module is used to dynamically control the welding parameters of the welding equipment according to the results of intelligent analysis if abnormal welding occurs, and optimize the welding path according to the actual situation.

Images