Intelligent control method, device and equipment for welding equipment
Through multi-view image acquisition and deep neural network prediction model, the problem of low response speed caused by large data processing of welding equipment is solved, and efficient control and accuracy of welding equipment is achieved.
Patent Information
- Application Number
- CN202410717253.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-04
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-06-04
AI Technical Summary
When existing welding equipment acquires multi-angle image data, the data processing volume is too large, resulting in low response speed of the control terminal and unable to make corresponding control strategies in time, affecting the efficiency of the welding process.
Multi-view image data is obtained through the multi-view image acquisition device on the welding device, and dimensionality reduction is performed using a linear projection matrix and discriminating sparse neighborhood maintenance embedding algorithm, the outline information of the area to be welded is extracted, and welding accuracy characteristics are constructed based on the deep neural network. Data prediction model, generate regulation information for welding control.
It effectively reduces the delay of the control terminal, improves the response speed and accuracy of the welding equipment, and ensures the efficiency and accuracy of the welding process.
Smart Images

Figure CN118492763B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of welding equipment, and in particular to an intelligent control method, device and equipment for welding equipment. Background Art
[0002] Under the background of intelligent manufacturing, robot technology is developing rapidly. Many occupations performed by humans are being replaced by robots. Among them, welding robots have been applied to many industries. It can be predicted that in the near future, welding robots will be able to perform almost all welding tasks, replace welding workers, and greatly save manpower. Visual sensors and camera devices can retain the image of objects, locate the welding position according to the correspondence between spatial points and pixel points, and assist robots in finding welding positions, which plays an important role in intelligent welding. However, in the prior art, multi-angle image data is obtained with the assistance of machine vision technology, so as to complete the acquisition of multi-angle image data of welding equipment when welding parts, and track the welding process according to the image data, such as tracking through deep neural networks and twin tracking networks. However, in the prior art, the collected images are tracked to a large extent, not the designated welding area, resulting in excessive data processing when identifying the welding situation, which is not conducive to the control terminal to make corresponding control strategies in time, resulting in the control terminal The response speed is too low. Summary of the invention
[0003] The present invention overcomes the deficiencies of the prior art and provides an intelligent control method, device and equipment for welding equipment.
[0004] To achieve the above object, the technical solution adopted by the present invention is:
[0005] A first aspect of the present invention provides an intelligent control method for welding equipment, comprising the following steps:
[0006] Acquire multi-view image data information during welding by a multi-view image acquisition device arranged on the welding equipment, and acquire a linear projection matrix in each projection direction according to the multi-view image data information during welding;
[0007] Acquire the process parameter characteristic information required by the welding parts, identify the working condition through the linear projection matrix in each projection direction and the process parameter characteristic information required by the welding parts, and obtain the identification result;
[0008] Acquire historical welding accuracy change characteristic data of the welding equipment, and build a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predict the welding accuracy characteristic data within the current preset time through the welding accuracy characteristic data prediction model;
[0009] The first control information or the second control information is generated according to the recognition result and the welding accuracy characteristic data within the current preset time, and the welding equipment is controlled based on the first control information or the second control information.
[0010] Furthermore, in the present method, multi-view image data information of the welding process is obtained by a multi-view image acquisition device provided on the welding equipment, and a linear projection matrix in each projection direction is obtained according to the multi-view image data information of the welding process, specifically including:
[0011] The multi-view image data information of the welding process is obtained by using a multi-view image acquisition device provided on the welding equipment, and the processed image data information is obtained by filtering and denoising the multi-view image data information of the welding process;
[0012] A discriminant sparse neighborhood preserving embedding algorithm is introduced, and the processed image data information is subjected to dimensionality reduction processing by the discriminant sparse neighborhood preserving embedding algorithm to obtain the unit matrix after dimensionality reduction processing, and the contour information of the area to be welded is extracted according to the unit matrix after dimensionality reduction processing;
[0013] Calculate the proportion between the contour information of the area to be welded and the unit matrix after the dimensionality reduction process. When the proportion between the contour information of the area to be welded and the unit matrix after the dimensionality reduction process is greater than a preset proportion, generate a linear projection matrix according to the unit matrix after the dimensionality reduction process, and output the linear projection matrix in each projection direction.
[0014] When the proportion between the contour information of the area to be welded and the unit matrix after dimensionality reduction processing is not greater than the preset proportion value, the unit matrix after dimensionality reduction processing is decomposed to generate an intra-class reconstruction scatter matrix and an inter-class reconstruction scatter matrix;
[0015] According to the intra-class reconstruction scatter matrix and the inter-class reconstruction scatter matrix, the eigenvectors corresponding to the first n largest eigenvalues in the intra-class reconstruction scatter matrix and the inter-class reconstruction scatter matrix are calculated, and the linear projection matrix is constructed based on the eigenvectors, the linear projection matrix in each projection direction.
[0016] Furthermore, in this method, the process parameter characteristic information required by the welding parts is obtained, and the working condition is identified through the linear projection matrix in each projection direction and the process parameter characteristic information required by the welding parts, and the identification result is obtained, which specifically includes:
[0017] Acquire process parameter characteristic information required for the welding part, and acquire the process parameter characteristic range required for the to-be-welded area of the welding part in each projection direction according to the process parameter characteristic information required for the welding part;
[0018] Acquire real-time welding process parameter characteristic information in each projection direction according to the linear projection matrix in each projection direction, and determine whether the real-time welding process parameter characteristic information is within the process parameter characteristic range;
[0019] When the real-time welding process parameter characteristic information is within the process parameter characteristic range, the working condition corresponding to the real-time welding process parameter characteristic information is regarded as a normal working condition, and the normal working condition is output as a recognition result;
[0020] When the real-time welding process parameter characteristic information is not within the process parameter characteristic range, the working condition corresponding to the real-time welding process parameter characteristic information is regarded as an abnormal working condition, and the abnormal working condition is output as a recognition result.
[0021] Furthermore, in the present method, historical welding accuracy change characteristic data of the welding equipment is obtained, and a welding accuracy characteristic data prediction model is constructed based on the historical welding accuracy change characteristic data of the welding equipment. The welding accuracy characteristic data prediction model is used to predict the welding accuracy characteristic data within the current preset time, specifically including:
[0022] Obtain historical welding accuracy change characteristic data of welding equipment, and build a welding accuracy characteristic data prediction model based on a deep neural network, and build welding accuracy change characteristic data based on a time series according to the historical welding accuracy change characteristic data of welding equipment;
[0023] Initialize a training set according to welding accuracy change feature data based on a time series, and input the training set into a welding accuracy feature data prediction model for training, so as to obtain a trained welding accuracy feature data prediction model;
[0024] The welding accuracy characteristic data of the welding equipment within the last preset time is obtained, and the welding accuracy characteristic data of the welding equipment within the last preset time is input into the trained welding accuracy characteristic data prediction model for prediction, so as to obtain the welding accuracy characteristic data within the preset time.
[0025] Furthermore, in the present method, generating the first control information or the second control information according to the recognition result and the welding accuracy characteristic data within the current preset time specifically includes:
[0026] When the recognition result is a normal condition, a continue welding instruction is generated, and the welding equipment is controlled to continue welding according to the continue welding instruction;
[0027] When the recognition result is an abnormal condition, real-time welding process parameter characteristic information and target welding process parameter characteristic information are obtained, and welding process deviation parameter information is calculated based on the real-time welding process parameter characteristic information and the target welding process parameter characteristic information;
[0028] Determine whether the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, and when the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, generate a stop processing instruction, and generate first control information according to the stop processing instruction;
[0029] When the welding process deviation parameter information is not greater than the welding accuracy characteristic data within the current preset time, a continue processing instruction is generated, and the second control information is generated according to the continue processing instruction.
[0030] Furthermore, in the present method, welding control of the welding equipment is performed based on the first control information or the second control information, specifically including:
[0031] If it is the first control information, the welding accuracy characteristic data of each welding equipment in the current welding equipment workshop within the current preset time is obtained, and the welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is obtained;
[0032] The welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is regarded as the processing equipment of the welding parts and the welding parts processing task is allocated;
[0033] If it is the second control information, the optimal welding equipment control parameter corresponding to the welding process deviation parameter information is obtained, and the welding equipment is controlled according to the optimal welding equipment control parameter corresponding to the welding process deviation parameter information.
[0034] A second aspect of the present invention provides an intelligent control device for welding equipment, the device comprising a memory and a processor, the memory comprising an intelligent control method program for welding equipment, and when the intelligent control method program for welding equipment is executed by the processor, the following steps are implemented:
[0035] Acquire multi-view image data information during welding by a multi-view image acquisition device arranged on the welding equipment, and acquire a linear projection matrix in each projection direction according to the multi-view image data information during welding;
[0036] Acquire the process parameter characteristic information required by the welding parts, identify the working condition through the linear projection matrix in each projection direction and the process parameter characteristic information required by the welding parts, and obtain the identification result;
[0037] Acquire historical welding accuracy change characteristic data of the welding equipment, and build a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predict the welding accuracy characteristic data within the current preset time through the welding accuracy characteristic data prediction model;
[0038] The first control information or the second control information is generated according to the recognition result and the welding accuracy characteristic data within the current preset time, and the welding equipment is controlled based on the first control information or the second control information.
[0039] A third aspect of the present invention provides an intelligent control device for welding equipment, comprising:
[0040] An acquisition module, which acquires multi-view image data information during welding through a multi-view image acquisition device provided on the welding device, and acquires a linear projection matrix in each projection direction according to the multi-view image data information during welding;
[0041] An identification module, through which characteristic information of process parameters required for welding parts is obtained, and working condition identification is performed through a linear projection matrix in each projection direction and characteristic information of process parameters required for welding parts to obtain identification results;
[0042] A prediction module, through which historical welding accuracy change characteristic data of the welding equipment is obtained, and a welding accuracy characteristic data prediction model is constructed based on the historical welding accuracy change characteristic data of the welding equipment, and the welding accuracy characteristic data prediction model is used to predict the welding accuracy characteristic data within the current preset time;
[0043] A control module generates first control information or second control information according to the recognition result and welding accuracy characteristic data within a current preset time, and performs welding control on the welding equipment based on the first control information or the second control information.
[0044] The present invention solves the defects existing in the background technology and has the following beneficial effects:
[0045] The present invention obtains multi-view image data information in the welding process through a multi-view image acquisition device set on the welding equipment, and obtains a linear projection matrix in each projection direction according to the multi-view image data information in the welding process, thereby obtaining the process parameter characteristic information required by the welding parts, and performing working condition recognition through the linear projection matrix in each projection direction and the process parameter characteristic information required by the welding parts, and obtaining the recognition result, thereby obtaining the historical welding accuracy change characteristic data of the welding equipment, and constructing a welding accuracy characteristic data prediction model according to the historical welding accuracy change characteristic data of the welding equipment, and predicting the welding accuracy characteristic data within the current preset time through the welding accuracy characteristic data prediction model, and finally generating the first control information or the second control information according to the recognition result and the welding accuracy characteristic data within the current preset time, and performing welding control on the welding equipment based on the first control information or the second control information. The present invention performs judgment processing by discriminating the sparse neighborhood preservation embedding algorithm, can timely process the image data collected from multiple perspectives, eliminate redundant image data, so that the collected image data can meet the predetermined requirements of the image data, and can effectively reduce the control delay of the control terminal for the welding equipment. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.
[0047] Figure 1 The overall flow chart of the intelligent control method of welding equipment is shown;
[0048] Figure 2 A partial flow chart of an intelligent control method for welding equipment is shown;
[0049] Figure 3 A device block diagram of an intelligent control device for welding equipment is shown;
[0050] Figure 4 A module schematic diagram of an intelligent control device for welding equipment is shown. DETAILED DESCRIPTION
[0051] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.
[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.
[0053] like Figure 1 As shown, the first aspect of the present invention provides an intelligent control method for welding equipment, comprising the following steps:
[0054] S102: acquiring multi-view image data information during the welding process through a multi-view image acquisition device provided on the welding device, and acquiring a linear projection matrix in each projection direction according to the multi-view image data information during the welding process;
[0055] S104: Acquire process parameter characteristic information required for welding parts, perform working condition identification through the linear projection matrix in each projection direction and the process parameter characteristic information required for welding parts, and obtain identification results;
[0056] S106: Acquire historical welding accuracy change characteristic data of the welding equipment, and construct a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predict the welding accuracy characteristic data within the current preset time through the welding accuracy characteristic data prediction model;
[0057] S108: Generate first control information or second control information according to the recognition result and welding accuracy characteristic data within the current preset time, and perform welding control on the welding equipment based on the first control information or the second control information.
[0058] It should be noted that the present invention performs judgment processing by discriminating a sparse neighborhood preserving embedding algorithm, which can timely process image data collected from multiple perspectives and eliminate redundant image data, so that the collected image data can meet the predetermined requirements of image data, and can effectively reduce the control delay of the control terminal for the welding equipment.
[0059] Furthermore, in the present method, multi-view image data information of the welding process is obtained by a multi-view image acquisition device provided on the welding equipment, and a linear projection matrix in each projection direction is obtained according to the multi-view image data information of the welding process, specifically including:
[0060] The multi-view image data information of the welding process is obtained by using a multi-view image acquisition device provided on the welding equipment, and the processed image data information is obtained by filtering and denoising the multi-view image data information of the welding process;
[0061] It should be noted that image acquisition equipment includes laser scanning equipment, infrared imaging equipment, cameras and other equipment.
[0062] A discriminant sparse neighborhood preserving embedding algorithm is introduced, and the processed image data information is subjected to dimensionality reduction processing by the discriminant sparse neighborhood preserving embedding algorithm to obtain the unit matrix after dimensionality reduction processing, and the contour information of the area to be welded is extracted according to the unit matrix after dimensionality reduction processing;
[0063] It should be noted that the contour information includes the dimensions of the contour in three directions of the three-dimensional space, the position information of the contour in the figure, and the like.
[0064] Calculate the proportion between the contour information of the area to be welded and the unit matrix after the dimensionality reduction process. When the proportion between the contour information of the area to be welded and the unit matrix after the dimensionality reduction process is greater than a preset proportion, generate a linear projection matrix according to the unit matrix after the dimensionality reduction process, and output the linear projection matrix in each projection direction.
[0065] When the proportion between the contour information of the area to be welded and the unit matrix after dimensionality reduction processing is not greater than the preset proportion value, the unit matrix after dimensionality reduction processing is decomposed to generate an intra-class reconstruction scatter matrix and an inter-class reconstruction scatter matrix;
[0066] According to the intra-class reconstruction scatter matrix and the inter-class reconstruction scatter matrix, the eigenvectors corresponding to the first n largest eigenvalues in the intra-class reconstruction scatter matrix and the inter-class reconstruction scatter matrix are calculated, and the linear projection matrix is constructed based on the eigenvectors, the linear projection matrix in each projection direction.
[0067] It should be noted that during the welding process, since there is a relative movement between the camera and the welding workpiece, in the process of identification, tracking can be performed through machine learning such as twin tracking networks and deep learning networks, so as to calculate the proportion between the contour information of the area to be welded and the unit matrix after dimensionality reduction processing. When the proportion between the contour information of the area to be welded and the unit matrix after dimensionality reduction processing is greater than the preset proportion, it means that the tracking area meets the predetermined requirements, and a linear projection matrix is constructed based on the eigenvector, and a linear projection matrix in each projection direction is used to delete a large amount of irrelevant image data, thereby reducing the tracking range, reducing the amount of data processing in the tracking process, and reducing the control delay of the control terminal for the welding equipment.
[0068] like Figure 2 As shown, further, in this method, the process parameter characteristic information required by the welding parts is obtained, the working condition is identified through the linear projection matrix in each projection direction and the process parameter characteristic information required by the welding parts, and the identification result is obtained, which specifically includes:
[0069] S202: Acquire process parameter characteristic information required for the welding part, and acquire the process parameter characteristic range required for the to-be-welded area of the welding part in each projection direction according to the process parameter characteristic information required for the welding part;
[0070] S204: acquiring real-time welding process parameter characteristic information in each projection direction according to the linear projection matrix in each projection direction, and determining whether the real-time welding process parameter characteristic information is within the process parameter characteristic range;
[0071] S206: When the real-time welding process parameter characteristic information is within the process parameter characteristic range, the working condition corresponding to the real-time welding process parameter characteristic information is regarded as a normal working condition, and the normal working condition is output as a recognition result;
[0072] S208: When the real-time welding process parameter characteristic information is not within the process parameter characteristic range, the working condition corresponding to the real-time welding process parameter characteristic information is regarded as an abnormal working condition, and the abnormal working condition is output as a recognition result.
[0073] It should be noted that the process parameter characteristic information required for welding parts includes welding profile deviation range, welding area position, etc. This method can identify the real-time welding process adoption characteristic information.
[0074] Furthermore, in the present method, historical welding accuracy change characteristic data of the welding equipment is obtained, and a welding accuracy characteristic data prediction model is constructed based on the historical welding accuracy change characteristic data of the welding equipment. The welding accuracy characteristic data prediction model is used to predict the welding accuracy characteristic data within the current preset time, specifically including:
[0075] Obtain historical welding accuracy change characteristic data of welding equipment, and build a welding accuracy characteristic data prediction model based on a deep neural network, and build welding accuracy change characteristic data based on a time series according to the historical welding accuracy change characteristic data of welding equipment;
[0076] Initialize a training set according to welding accuracy change feature data based on a time series, and input the training set into a welding accuracy feature data prediction model for training, so as to obtain a trained welding accuracy feature data prediction model;
[0077] The welding accuracy characteristic data of the welding equipment within the last preset time is obtained, and the welding accuracy characteristic data of the welding equipment within the last preset time is input into the trained welding accuracy characteristic data prediction model for prediction, so as to obtain the welding accuracy characteristic data within the preset time.
[0078] It should be noted that, since the performance of welding equipment will degrade after a certain number of years of use, such as the control accuracy of the driving motor driving the welding source, the welding gun, the welding power supply and other equipment, the actual speed of the driving motor deviates from the predetermined speed, resulting in deviations in welding thickness, welding length area, welding width area, etc., this method can obtain welding accuracy feature data within a preset time. Among them, deep neural networks include convolutional neural networks, recurrent neural networks, long-short time memory neural networks, etc.
[0079] Furthermore, in the present method, generating the first control information or the second control information according to the recognition result and the welding accuracy characteristic data within the current preset time specifically includes:
[0080] When the recognition result is a normal condition, a continue welding instruction is generated, and the welding equipment is controlled to continue welding according to the continue welding instruction;
[0081] When the recognition result is an abnormal condition, real-time welding process parameter characteristic information and target welding process parameter characteristic information are obtained, and welding process deviation parameter information is calculated based on the real-time welding process parameter characteristic information and the target welding process parameter characteristic information;
[0082] Determine whether the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, and when the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, generate a stop processing instruction, and generate first control information according to the stop processing instruction;
[0083] When the welding process deviation parameter information is not greater than the welding accuracy characteristic data within the current preset time, a continue processing instruction is generated, and the second control information is generated according to the continue processing instruction.
[0084] It should be noted that when the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, it means that when the current welding equipment continues to weld, because the accuracy of the welding equipment does not meet the requirements, the welding equipment will continue to be used, which will also make the welded parts not meet the predetermined requirements. On the contrary, the welded parts will meet the predetermined requirements. This method can further improve the welding rationality in the welding process.
[0085] Furthermore, in the present method, welding control of the welding equipment is performed based on the first control information or the second control information, specifically including:
[0086] If it is the first control information, the welding accuracy characteristic data of each welding equipment in the current welding equipment workshop within the current preset time is obtained, and the welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is obtained;
[0087] The welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is regarded as the processing equipment of the welding parts and the welding parts processing task is allocated;
[0088] If it is the second control information, the optimal welding equipment control parameter corresponding to the welding process deviation parameter information is obtained, and the welding equipment is controlled according to the optimal welding equipment control parameter corresponding to the welding process deviation parameter information.
[0089] It should be noted that this method can further improve the welding rationality during the welding process.
[0090] In addition, obtaining multi-view image data information during welding by a multi-view image acquisition device provided on the welding device may include the following steps:
[0091] By configuring image acquisition devices on the welding equipment and obtaining the location information of each image acquisition device, obtaining the real-time environment information of the location information of each image acquisition device, and initializing the camera parameter information of each image acquisition device;
[0092] Obtain the image clarity feature data corresponding to each camera parameter information under each environmental parameter information through big data, and construct a knowledge graph with the environmental parameter information as the first node, the camera parameter information as the second node, and the image clarity feature data as the third node;
[0093] A topological structure diagram is constructed based on the first node, the second node, and the third node through directed edge description relationships, the topological structure diagram is input into the knowledge graph for storage, the camera parameter information of each image acquisition device and the real-time environmental information of the location information of each image acquisition device are input into the knowledge graph for data matching, and the real-time clarity feature data of each image acquisition device is obtained;
[0094] A genetic algorithm is introduced, and a genetic generation is set according to the genetic algorithm. When the real-time clarity feature data of the image acquisition device is lower than a preset clarity feature threshold, inheritance is performed according to the genetic generation, and the camera parameter information of the image acquisition device is readjusted until the real-time clarity feature data of the image acquisition device is not lower than the preset clarity feature threshold, and the camera device is controlled according to the camera parameter information of the image acquisition device to obtain multi-view image data information during the welding process.
[0095] It should be noted that the camera parameter information of the image acquisition device includes the infrared wavelength, frequency, brightness, contrast, exposure and other characteristics of the infrared thermal imager. Since a large amount of gas will be generated during welding or the visibility in the environmental factors is low, or the temperature and humidity change greatly, this is not conducive to the infrared thermal imager or the image collected by the camera does not meet the predetermined requirements. This method can further improve the rationality of data acquisition.
[0096] In addition, the method may further comprise the following steps:
[0097] Obtaining the order information of the current welding parts and the welding accuracy characteristic data of each current welding equipment, and obtaining the welding accuracy characteristic data required for each welding part and the welding quantity information of each welding part according to the order information of the current welding parts;
[0098] The welding accuracy characteristic data of the welding equipment is greater than the welding accuracy characteristic data required by the welding parts, and the welding quantity of each welding equipment is obtained, and the average welding time information of the unit number of welding parts is obtained through big data;
[0099] Calculate the estimated welding time information corresponding to the welding amount of each welding device according to the welding amount of each welding device and the average welding time information per unit number of welded parts, and predict the predicted time node when each welding device reaches the critical point of the preset welding accuracy characteristic threshold through the welding accuracy characteristic data prediction model;
[0100] When the estimated welding time information corresponding to the welding amount of the welding device is lower than the predicted time node when the welding device reaches the critical point of the preset welding accuracy characteristic threshold, outputting the welding amount of the welding device;
[0101] When the estimated welding time information corresponding to the welding amount of the welding equipment is not lower than the predicted time node at which the welding equipment reaches the critical point of the preset welding accuracy characteristic threshold, the welding amount of the welding equipment is readjusted until it is lower than the predicted time node at which the welding equipment reaches the critical point of the preset welding accuracy characteristic threshold.
[0102] It should be noted that, since the welding accuracy is changing, this method can further improve the rationality of the processing allocation of the welding equipment during welding and reduce the defective rate of welded parts.
[0103] like Figure 3As shown, the second aspect of the present invention provides an intelligent control device 4 for welding equipment, the device 4 includes a memory 41 and a processor 42, the memory 41 includes an intelligent control method program for welding equipment, and when the intelligent control method program for welding equipment is executed by the processor 42, the following steps are implemented:
[0104] Acquire multi-view image data information during welding by a multi-view image acquisition device arranged on the welding equipment, and acquire a linear projection matrix in each projection direction according to the multi-view image data information during welding;
[0105] Acquire the process parameter characteristic information required by the welding parts, identify the working condition through the linear projection matrix in each projection direction and the process parameter characteristic information required by the welding parts, and obtain the identification result;
[0106] Acquire historical welding accuracy change characteristic data of the welding equipment, and build a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predict the welding accuracy characteristic data within the current preset time through the welding accuracy characteristic data prediction model;
[0107] The first control information or the second control information is generated according to the recognition result and the welding accuracy characteristic data within the current preset time, and the welding equipment is controlled based on the first control information or the second control information.
[0108] like Figure 4 As shown, the third aspect of the present invention provides an intelligent control device for welding equipment, comprising:
[0109] An acquisition module 10, which acquires multi-view image data information during welding through a multi-view image acquisition device provided on the welding device, and acquires a linear projection matrix in each projection direction according to the multi-view image data information during welding;
[0110] An identification module 20, which obtains process parameter characteristic information required by the welding part through the identification module, performs working condition identification through the linear projection matrix in each projection direction and the process parameter characteristic information required by the welding part, and obtains an identification result;
[0111] A prediction module 30, which obtains historical welding accuracy change characteristic data of the welding equipment through the prediction module, and constructs a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predicts the welding accuracy characteristic data within the current preset time through the welding accuracy characteristic data prediction model;
[0112] The control module 40 generates the first control information or the second control information according to the recognition result and the welding accuracy characteristic data within the current preset time, and performs welding control on the welding equipment based on the first control information or the second control information.
[0113] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.
[0114] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0115] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.
[0116] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.
[0117] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, disks or optical disks.
[0118] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. An intelligent control method for welding equipment, characterized in that: The following steps are involved: Acquire multi-view image data information during welding by a multi-view image acquisition device provided on the welding equipment, and acquire a linear projection matrix in each projection direction according to the multi-view image data information during welding; Acquire process parameter characteristic information required for welding parts, perform working condition identification through the linear projection matrix in each projection direction and the process parameter characteristic information required for welding parts, and obtain identification results; Acquire historical welding accuracy change characteristic data of the welding equipment, and construct a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predict the welding accuracy characteristic data within a current preset time through the welding accuracy characteristic data prediction model; Generate first control information or second control information according to the recognition result and welding accuracy characteristic data within a current preset time, and perform welding control on the welding equipment based on the first control information or the second control information; Generating the first control information or the second control information according to the recognition result and the welding accuracy characteristic data within the current preset time specifically includes: When the recognition result is a normal condition, a continue welding instruction is generated, and the welding equipment is controlled to continue welding according to the continue welding instruction; When the recognition result is an abnormal condition, real-time welding process parameter characteristic information and target welding process parameter characteristic information are obtained, and welding process deviation parameter information is calculated according to the real-time welding process parameter characteristic information and the target welding process parameter characteristic information; Determine whether the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, and when the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, generate a stop processing instruction, and generate first control information according to the stop processing instruction; When the welding process deviation parameter information is not greater than the welding accuracy characteristic data within the current preset time, a continue processing instruction is generated, and second control information is generated according to the continue processing instruction; The method of controlling welding equipment based on the first control information or the second control information specifically includes: If it is the first control information, the welding accuracy characteristic data of each welding equipment in the current welding equipment workshop within the current preset time is obtained, and the welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is obtained; The welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is used as the processing equipment for the welding parts and the welding parts processing task is assigned; If it is the second control information, the optimal welding equipment control parameter corresponding to the welding process deviation parameter information is obtained, and the welding equipment is controlled according to the optimal welding equipment control parameter corresponding to the welding process deviation parameter information.
2. The intelligent control method for welding equipment according to claim 1, characterized in that: The multi-view image data information of the welding process is obtained by a multi-view image acquisition device provided on the welding device, and the linear projection matrix in each projection direction is obtained according to the multi-view image data information of the welding process, specifically including: Acquire multi-view image data information during the welding process by using a multi-view image acquisition device provided on the welding device, and acquire processed image data information by filtering and denoising the multi-view image data information during the welding process; Introducing a discriminant sparse neighborhood preserving embedding algorithm, and performing dimensionality reduction processing on the processed image data information by using the discriminant sparse neighborhood preserving embedding algorithm to obtain a unit matrix after the dimensionality reduction processing, and extracting contour information of the area to be welded according to the unit matrix after the dimensionality reduction processing; Calculating the proportion between the contour information of the area to be welded and the unit matrix after the dimensionality reduction process, when the proportion between the contour information of the area to be welded and the unit matrix after the dimensionality reduction process is greater than a preset proportion, generating a linear projection matrix according to the unit matrix after the dimensionality reduction process, and outputting the linear projection matrix in each projection direction; When the weight ratio between the contour information of the area to be welded and the unit matrix after the dimensionality reduction process is not greater than the preset weight ratio, the unit matrix after the dimensionality reduction process is decomposed to generate an intra-class reconstruction divergence reconstruction matrix and an inter-class reconstruction divergence matrix; According to the intra-class reconstruction divergence reconstruction matrix and the inter-class reconstruction divergence matrix, calculations are performed to obtain the eigenvectors corresponding to the first n largest eigenvalues in the intra-class reconstruction divergence reconstruction matrix and the inter-class reconstruction divergence matrix, and a linear projection matrix is constructed based on the eigenvectors, a linear projection matrix in each projection direction.
3. The intelligent control method for welding equipment according to claim 1, characterized in that: Acquiring process parameter characteristic information required by welding parts, performing working condition identification through the linear projection matrix in each projection direction and the process parameter characteristic information required by welding parts, and obtaining identification results, specifically including: Acquire process parameter characteristic information required by the welding part, and acquire the process parameter characteristic range required by the to-be-welded area of the welding part in each projection direction according to the process parameter characteristic information required by the welding part; Acquire real-time welding process parameter characteristic information in each projection direction according to the linear projection matrix in each projection direction, and determine whether the real-time welding process parameter characteristic information is within the process parameter characteristic range; When the real-time welding process parameter characteristic information is within the process parameter characteristic range, the working condition corresponding to the real-time welding process parameter characteristic information is regarded as a normal working condition, and the normal working condition is output as a recognition result; When the real-time welding process parameter characteristic information is not within the process parameter characteristic range, the working condition corresponding to the real-time welding process parameter characteristic information is regarded as an abnormal working condition, and the abnormal working condition is output as a recognition result.
4. The intelligent control method for welding equipment according to claim 1, characterized in that: Acquiring historical welding accuracy change characteristic data of the welding equipment, and building a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predicting the welding accuracy characteristic data within the current preset time by using the welding accuracy characteristic data prediction model, specifically including: Acquire historical welding accuracy change characteristic data of welding equipment, and construct a welding accuracy characteristic data prediction model based on a deep neural network, and construct welding accuracy change characteristic data based on a time series according to the historical welding accuracy change characteristic data of the welding equipment; Initializing a training set according to the welding accuracy change characteristic data based on the time series, and inputting the training set into the welding accuracy characteristic data prediction model for training, so as to obtain a trained welding accuracy characteristic data prediction model; The welding accuracy characteristic data of the welding equipment within the last preset time is obtained, and the welding accuracy characteristic data of the welding equipment within the last preset time is input into the trained welding accuracy characteristic data prediction model for prediction, so as to obtain the welding accuracy characteristic data within the preset time.
5. An intelligent control device for welding equipment, characterized in that: The device includes a memory and a processor, wherein the memory includes an intelligent control method program for welding equipment, and when the intelligent control method program for welding equipment is executed by the processor, the following steps are implemented: Acquire multi-view image data information during welding by a multi-view image acquisition device provided on the welding equipment, and acquire a linear projection matrix in each projection direction according to the multi-view image data information during welding; Acquire process parameter characteristic information required for welding parts, perform working condition identification through the linear projection matrix in each projection direction and the process parameter characteristic information required for welding parts, and obtain identification results; Acquire historical welding accuracy change characteristic data of the welding equipment, and construct a welding accuracy characteristic data prediction model based on the historical welding accuracy change characteristic data of the welding equipment, and predict the welding accuracy characteristic data within a current preset time through the welding accuracy characteristic data prediction model; Generate first control information or second control information according to the recognition result and welding accuracy characteristic data within a current preset time, and perform welding control on the welding equipment based on the first control information or the second control information; Generating the first control information or the second control information according to the recognition result and the welding accuracy characteristic data within the current preset time specifically includes: When the recognition result is a normal condition, a continue welding instruction is generated, and the welding equipment is controlled to continue welding according to the continue welding instruction; When the recognition result is an abnormal condition, real-time welding process parameter characteristic information and target welding process parameter characteristic information are obtained, and welding process deviation parameter information is calculated according to the real-time welding process parameter characteristic information and the target welding process parameter characteristic information; Determine whether the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, and when the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, generate a stop processing instruction, and generate first control information according to the stop processing instruction; When the welding process deviation parameter information is not greater than the welding accuracy characteristic data within the current preset time, a continue processing instruction is generated, and second control information is generated according to the continue processing instruction; The method of controlling welding equipment based on the first control information or the second control information specifically includes: If it is the first control information, the welding accuracy characteristic data of each welding equipment in the current welding equipment workshop within the current preset time is obtained, and the welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is obtained; The welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is used as the processing equipment for the welding parts and the welding parts processing task is assigned; If it is the second control information, the optimal welding equipment control parameter corresponding to the welding process deviation parameter information is obtained, and the welding equipment is controlled according to the optimal welding equipment control parameter corresponding to the welding process deviation parameter information.
6. An intelligent control device for welding equipment, characterized in that: include: An acquisition module, which acquires multi-view image data information during welding through a multi-view image acquisition device provided on the welding device, and acquires a linear projection matrix in each projection direction according to the multi-view image data information during welding; An identification module, through which process parameter characteristic information required for welding parts is obtained, working condition identification is performed through the linear projection matrix in each projection direction and the process parameter characteristic information required for welding parts, and an identification result is obtained; A prediction module, through which historical welding accuracy change characteristic data of a welding device are obtained, and a welding accuracy characteristic data prediction model is constructed according to the historical welding accuracy change characteristic data of the welding device, and the welding accuracy characteristic data prediction model is used to predict the welding accuracy characteristic data within a current preset time; A control module, through which the first control information or the second control information is generated according to the recognition result and welding accuracy characteristic data within a current preset time, and welding control is performed on the welding equipment based on the first control information or the second control information; Generating the first control information or the second control information according to the recognition result and the welding accuracy characteristic data within the current preset time specifically includes: When the recognition result is a normal condition, a continue welding instruction is generated, and the welding equipment is controlled to continue welding according to the continue welding instruction; When the recognition result is an abnormal condition, real-time welding process parameter characteristic information and target welding process parameter characteristic information are obtained, and welding process deviation parameter information is calculated according to the real-time welding process parameter characteristic information and the target welding process parameter characteristic information; Determine whether the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, and when the welding process deviation parameter information is greater than the welding accuracy characteristic data within the current preset time, generate a stop processing instruction, and generate first control information according to the stop processing instruction; When the welding process deviation parameter information is not greater than the welding accuracy characteristic data within the current preset time, a continue processing instruction is generated, and second control information is generated according to the continue processing instruction; The method of controlling welding equipment based on the first control information or the second control information specifically includes: If it is the first control information, the welding accuracy characteristic data of each welding equipment in the current welding equipment workshop within the current preset time is obtained, and the welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is obtained; The welding equipment whose welding accuracy characteristic data within the current preset time is greater than the welding process deviation parameter information is used as the processing equipment for the welding parts and the welding parts processing task is assigned; If it is the second control information, the optimal welding equipment control parameter corresponding to the welding process deviation parameter information is obtained, and the welding equipment is controlled according to the optimal welding equipment control parameter corresponding to the welding process deviation parameter information.
