Dynamic optimization method and system of welding process combined with tracking monitoring

Abstract

The present invention discloses a method and system for dynamic optimization of a welding process combined with tracking and monitoring, and relates to the field of welding technology. The method comprises: obtaining a historical welding record set, parsing the record set based on a target welding scenario, and determining a calibrated welding process set. Configure a forward-looking acquisition step, initialize and activate a weld acquisition component for continuous acquisition, and acquire weld contour data. Perform three-dimensional reconstruction through the weld contour data to generate a displacement weld model containing multiple weld cross sections. Extract a weld feature parameter set, and use it as an index constraint to traverse the calibrated welding process set for matching and obtain a welding process sequence. Perform continuous welding control of the target weld based on the welding process sequence, activate the tracking and monitoring component, continuously monitor the welding results, generate welding feedback information, and perform welding process response accordingly. Thereby achieving the technical effect of improving the dynamic adjustment capability of the welding process and improving the welding quality and stability.

Description

Technical Field

[0001] The present invention relates to the field of welding technology, and in particular to a welding process dynamic optimization method and system combined with tracking monitoring. Background Art

[0002] Welding is an indispensable and important process in the manufacturing industry. It is widely used in the manufacturing process of various structural parts and equipment. The quality of the weld directly affects the performance and safety of the finished product. The existing weld process optimization methods often determine relatively fixed welding process parameters by conducting welding experiments on samples. They lack the ability of real-time monitoring and dynamic adjustment, and have technical problems such as poor real-time optimization and process promotion, which affect the quality of the weld. Summary of the invention

[0003] The present invention provides a method and system for dynamic optimization of welding process combined with tracking monitoring to solve the technical problems in the prior art of poor real-time optimization and process promotion, which affect the quality of welds, and achieve the technical effect of improving the dynamic adjustment capability of welding process and improving welding quality and stability.

[0004] In a first aspect, the present invention provides a method for dynamic optimization of a welding process combined with tracking monitoring, wherein the method comprises:

[0005] A historical welding record set is obtained, and based on the welding requirements of the target welding scene, the historical welding record set is parsed to determine a calibration welding process set.

[0006] The forward-looking acquisition step is configured, and the weld acquisition component is initialized based on the forward-looking acquisition step, and the weld acquisition component is activated to perform continuous acquisition to obtain weld contour data.

[0007] A three-dimensional reconstruction is performed based on the weld profile data to generate a displacement weld model, wherein the displacement weld model includes a weld section series consisting of a plurality of weld sections.

[0008] According to the displacement weld model, a weld feature parameter set is extracted, and the weld feature parameter set is used as an index constraint to traverse the calibrated welding process set to perform welding process matching and obtain a welding process sequence.

[0009] Based on the welding process sequence, continuous welding control of the target weld is performed, and the tracking monitoring component is activated to continuously monitor and collect welding control results for welding evaluation and generate welding feedback information.

[0010] The welding process response is performed according to the welding feedback information.

[0011] In a second aspect, the present invention further provides a welding process dynamic optimization system combined with tracking monitoring, wherein the system comprises:

[0012] A process calibration module is used to obtain a historical welding record set, parse the historical welding record set based on the welding requirements of the target welding scene, and determine a calibrated welding process set.

[0013] The forward-looking acquisition module is used to configure the forward-looking acquisition step length, initialize the weld acquisition component based on the forward-looking acquisition step length, activate the weld acquisition component for continuous acquisition, and obtain weld profile data.

[0014] A cross-section reconstruction module is used to perform three-dimensional reconstruction based on the weld contour data to generate a displacement weld model, wherein the displacement weld model includes a weld cross-section series consisting of multiple weld cross sections.

[0015] A process matching module is used to extract a weld feature parameter set according to the displacement weld model, and use the weld feature parameter set as an index constraint to traverse the calibrated welding process set to perform welding process matching and obtain a welding process sequence.

[0016] A tracking and monitoring module is used to perform continuous welding control of a target weld based on the welding process sequence, activate a tracking and monitoring component, continuously monitor and collect welding control results for welding evaluation, and generate welding feedback information.

[0017] A process response module is used to perform welding process response according to the welding feedback information.

[0018] The present invention discloses a method and system for dynamic optimization of a welding process combined with tracking and monitoring, comprising: obtaining a historical welding record set, parsing the historical welding record set based on welding requirements of a target welding scene, and determining a calibrated welding process set; configuring a forward-looking acquisition step, initializing a weld acquisition component based on the forward-looking acquisition step, activating the weld acquisition component for continuous acquisition, and acquiring weld contour data; performing three-dimensional reconstruction based on the weld contour data, and generating a displacement weld model, wherein the displacement weld model includes a weld section series composed of multiple weld sections; extracting a weld feature parameter set according to the displacement weld model, and using the weld feature parameter set as an index constraint, traversing the calibrated welding process set for welding process matching, and acquiring a welding process sequence; performing continuous welding control of a target weld based on the welding process sequence, and activating the tracking and monitoring component, continuously monitoring and collecting welding control results for welding evaluation, and generating welding feedback information; and performing a welding process response according to the welding feedback information. The welding process dynamic optimization method and system combined with tracking monitoring disclosed in the present invention solve the technical problems of poor real-time and process promotion of process optimization and affecting weld quality, and achieve the technical effect of improving the dynamic adjustment capability of welding process and improving welding quality and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic flow chart of the welding process dynamic optimization method combined with tracking monitoring of the present invention;

[0020] Figure 2 It is a structural schematic diagram of the welding process dynamic optimization system combined with tracking monitoring of the present invention.

[0021] Explanation of the accompanying drawings: process calibration module 11, forward-looking acquisition module 12, cross-section reconstruction module 13, process matching module 14, tracking and monitoring module 15, process response module 16. DETAILED DESCRIPTION

[0022] The technical solution provided in the embodiments of the present invention is to solve the technical problems of poor real-time performance and poor process promotion of process optimization in the prior art, which affects the quality of the weld. The overall idea adopted is as follows:

[0023] First, a historical welding record set is obtained, and based on the welding requirements of the target welding scene, the historical welding record set is parsed to determine a calibration welding process set.

[0024] Then, the forward-looking acquisition step is configured, and the weld acquisition component is initialized based on the forward-looking acquisition step, and the weld acquisition component is activated to perform continuous acquisition to obtain weld contour data.

[0025] Then, three-dimensional reconstruction is performed based on the weld profile data to generate a displacement weld model, wherein the displacement weld model includes a weld section series consisting of a plurality of weld sections.

[0026] Next, according to the displacement weld model, a weld feature parameter set is extracted, and with the weld feature parameter set as an index constraint, the calibrated welding process set is traversed to perform welding process matching and obtain a welding process sequence.

[0027] Furthermore, continuous welding control of the target weld is performed based on the welding process sequence, and the tracking monitoring component is activated to continuously monitor and collect welding control results for welding evaluation and generate welding feedback information.

[0028] Finally, the welding process response is performed according to the welding feedback information.

[0029] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them.

[0030] Embodiment 1

[0031] Figure 1 The figure is a flow chart of a method for dynamic optimization of a welding process combined with tracking monitoring according to the present invention, wherein the method comprises:

[0032] A historical welding record set is obtained, and based on the welding requirements of the target welding scene, the historical welding record set is parsed to determine a calibration welding process set.

[0033] Specifically, collect and organize past welding records, including various welding parameters (such as current, voltage, welding speed, etc.), welding results (such as weld quality, welding efficiency, etc.), welding environment (such as material type, material thickness, ambient temperature, etc.) and other information.

[0034] Preferably, the historical welding records obtained include the welding history records of the target welding scene and the connection history records of the same source scenes of the target welding scene (such as welding scenes using the same welding equipment or used in similar welding fields), thereby ensuring that the data diversity and data volume of the historical welding records obtained meet the subsequent analysis requirements.

[0035] Specifically, the historical welding records are parsed, useful information is extracted, and the above-mentioned calibrated welding process set is formed. This includes identifying which welding parameters are related to excellent welding results, or which welding environments are more likely to cause welding problems. Then, the data corresponding to the factors strongly associated with the above-mentioned welding quality in the historical welding record set are extracted and sorted. Then, the welding requirements of the target welding scene are clarified, the relevant data under the corresponding requirements in the historical welding records are screened, and the calibrated welding process set is analyzed and determined. Among them, welding requirements include welding quality requirements, material compatibility requirements, production efficiency requirements, cost requirements, etc.

[0036] In some embodiments, obtaining a historical welding record set, parsing the historical welding record set based on welding requirements of a target welding scene, and determining a calibration welding process set include:

[0037] Taking the welding requirement as a screening constraint, the historical welding record set is traversed to obtain a calibrated welding record set.

[0038] A plurality of weld feature parameter sets of the calibrated welding record set are extracted, and cluster analysis is performed on the calibrated welding record set based on the plurality of weld feature parameter sets to obtain a plurality of calibrated substrate record clusters.

[0039] A plurality of cluster centers of a plurality of the calibrated base material record clusters are obtained, and the welding processes corresponding to the plurality of cluster centers are associated and stored with the weld feature parameter set to obtain the calibrated welding process set.

[0040] Specifically, the screening conditions are set according to the welding requirements, such as welding materials, welding methods, environmental conditions and quality requirements, and then the historical welding record set is traversed to screen out the records that meet the conditions to form a calibrated welding record set.

[0041] Next, weld characteristic parameters, such as weld width, depth, heat-affected zone size, microstructure distribution, etc., are extracted from the calibration welding record set, and the extracted characteristic parameters are standardized to ensure the comparability between different features. Then, an appropriate clustering algorithm (such as K-means, hierarchical clustering, DBSCAN, etc.) is selected to perform cluster analysis on the calibration welding record set to obtain multiple calibration substrate record clusters, which contain all welding records of a weld condition that meets the requirements of the target welding scenario. Through cluster analysis, multiple groups of records that may produce similar welding results under specific conditions can be obtained.

[0042] Furthermore, the cluster center of each calibration base material record cluster is calculated to obtain the cluster center characteristic parameters. The welding process parameters of each cluster center are associated and stored with the corresponding weld characteristic parameter set to form a calibration welding process set containing welding processes and weld characteristic parameters corresponding to multiple cluster centers.

[0043] Specifically, the cluster center is the center point of each calibration base material record cluster, representing the typical welding process parameters under the corresponding weld conditions. In other words, it is the average or most typical value (welding process corresponding to the cluster center) of the welding records that meet specific welding requirements under specific weld conditions (weld characteristic parameter set). Among them, welding process parameters include current, voltage, welding speed, welding method, etc.

[0044] Through the above-mentioned method steps, screening and cluster analysis of historical welding records can help find multiple groups of welding processes corresponding to various weld states that meet the welding requirements of the target welding scenario from a large number of historical welding records, thereby improving the optimization efficiency of the welding process.

[0045] The forward-looking acquisition step is configured, and the weld acquisition component is initialized based on the forward-looking acquisition step, and the weld acquisition component is activated to perform continuous acquisition to obtain weld contour data.

[0046] Specifically, the forward-looking acquisition step includes the clearance step and the window step. The clearance step refers to the time sequence length or progress ratio of data acquisition in advance during the welding process. The clearance step specifies the advance amount of forward-looking real-time process optimization of the weld and provides execution time for the process optimization process. For example, the clearance step is determined by the computing power and welding requirements used for process optimization in the target scenario. The window step refers to the size of the acquisition window for each acquisition, which determines the quantity and range of the weld profile data obtained each time.

[0047] Optionally, based on the process optimization requirements and process optimization computing power of the target scenario, adjust the clearance step size and window step size to ensure the real-time and effectiveness of data collection.

[0048] Specifically, the weld acquisition component is a hardware and software system used to collect weld profile data, including laser rangefinders, 3D scanners, cameras and other equipment. Calibrate the weld acquisition component and configure the acquisition parameters of the weld acquisition equipment according to the forward acquisition step length to ensure that the equipment can collect data according to the set clearance step length and window step length, and ensure the accuracy and reliability of the collected data.

[0049] Specifically, the initialized weld acquisition component is activated to continuously acquire weld contour data, which may be single data including laser ranging data, point cloud data, image data, etc., or may be multi-dimensional data acquired synchronously. The weld contour data reflects the shape and state of the weld to be welded, thereby better controlling the welding process.

[0050] Optionally, according to the weld detection requirements, select the appropriate data type for collection. For example, laser ranging data is suitable for accurately measuring the depth and width of welds, point cloud data is suitable for 3D reconstruction, and image data is suitable for surface defect detection. If the collected data is multi-dimensional data, it is fused and processed to generate comprehensive weld profile data to improve the accuracy and completeness of the data.

[0051] The above method steps realize continuous monitoring of the weld by configuring the forward-looking acquisition step, and then obtain the weld contour data in advance during the welding process, which is helpful for high-real-time welding process optimization and decision-making, and is also conducive to timely discovery and handling of potential problems.

[0052] A three-dimensional reconstruction is performed based on the weld profile data to generate a displacement weld model, wherein the displacement weld model includes a weld section series consisting of a plurality of weld sections.

[0053] In some embodiments, performing three-dimensional reconstruction based on the weld profile data to generate a displacement weld model includes:

[0054] Based on the preset process control granularity, the weld profile data is lightly processed to extract a weld profile slice set.

[0055] A three-dimensional reconstruction is performed based on the weld contour slice set to obtain a weld cross-section set.

[0056] The weld section set is serialized according to the acquisition timing mark of the weld contour slice set to obtain the weld section series.

[0057] The weld cross-section series is stored in a displacement register, and along with the real-time update of the weld profile data, the displacement update of the weld cross-section series is performed, and the displacement weld model is generated.

[0058] Specifically, first, based on the preset process control granularity, lightweight processing of weld profile data is a data compression process, wherein the preset process control granularity defines the data sampling interval during the lightweight processing, that is, the interval for extracting data. The above steps help to reduce the amount of data while maintaining sufficient understanding of the weld profile, thereby achieving effective data management and processing, and facilitating subsequent 3D reconstruction and displacement weld model generation.

[0059] Specifically, the seam contour slice set includes multiple weld slice data, among which the seam contour data can be considered as continuous weld 3D feature data, and the multiple weld slice data extracted after light processing can be considered as the cross-sectional data at multiple process control granularity points of the aforementioned continuous weld 3D feature data. This method of data extraction and processing allows the data volume to be controlled within a reasonable range, while ensuring sufficient weld characteristic information, providing the necessary basis for subsequent 3D reconstruction and displacement weld model generation.

[0060] Further, the multiple weld slice data included in the seam contour slice set are respectively restored in three dimensions to generate multiple weld cross sections and obtain a weld cross section set. Then, according to the acquisition time sequence mark of the weld contour slice set, the weld cross section set is serialized to obtain a weld cross section series. In other words, through serialization processing, the weld cross section set is sorted according to the time sequence of acquisition, thereby forming an ordered weld cross section series. Among them, the sequence relationship direction of the weld cross section series is consistent with the welding direction of the target weld.

[0061] Specifically, the serialized weld section series is stored in the displacement register to ensure the order and integrity of the data. Then, as the welding process progresses, new weld profile data is collected in real time, and the new weld section slices are inserted into the displacement register to update the weld section series. In other words, if the weld profile data changes (for example, due to the progress of the welding process), the data of the weld section series will also be updated accordingly to reflect these changes. Finally, based on the updated weld section series, a displacement weld model is generated. The displacement weld model is a reduction model that is updated in real time with the displacement of the welding point, and is used to quickly and accurately reflect the changes in the weld profile data, thereby helping the subsequent welding process optimization.

[0062] According to the displacement weld model, a weld feature parameter set is extracted, and the weld feature parameter set is used as an index constraint to traverse the calibrated welding process set to perform welding process matching and obtain a welding process sequence.

[0063] The weld feature parameter set is a set of key parameters extracted from the displacement weld model to characterize the weld morphology and state. Exemplarily, it includes geometric feature parameters of the weld, such as weld width, depth, cross-sectional shape, weld curvature, etc.; surface features such as roughness and waviness of the weld surface. Optionally, the weld feature parameter set also includes structural feature parameters of the weld in the current welding task obtained by the interactive target welding scene, such as density, hardness, elastic modulus, etc. of the weld material.

[0064] Specifically, the seam feature parameter set is used as an index constraint, and the calibrated welding process set is traversed to perform welding process matching, that is, the real-time extracted weld feature parameter set is compared with the pre-calibrated welding process set to find the welding process that best matches the current weld feature. Optionally, a multidimensional parameter space corresponding to a variety of weld feature parameters is constructed, and then the weld feature parameter set is mapped to a point in the multidimensional parameter space, and each calibrated welding process is also mapped to a point in the multidimensional parameter space. Furthermore, based on the distance between the points (including Euclidean distance, cosine similarity, etc.), the similarity between the weld feature parameter set and each process in the calibrated welding process set is calculated, and sorted according to the similarity, the welding process with the highest similarity is selected as the matching result, and the calibrated welding process closest to the current weld feature parameter set is found.

[0065] Specifically, the displacement weld model is traversed, and multiple weld feature parameter sets corresponding to multiple weld sections are extracted. Based on the above method steps, multiple groups of optimal calibrated welding processes are obtained to generate a welding process sequence. The sequence direction of the welding process sequence is consistent with the unique direction of the displacement weld model, which characterizes the changes in the weld welding process along the weld welding direction and is used to guide the subsequent welding process.

[0066] Based on the welding process sequence, continuous welding control of the target weld is performed, and the tracking monitoring component is activated to continuously monitor and collect welding control results for welding evaluation and generate welding feedback information.

[0067] Specifically, continuous welding control of the target weld is performed based on the welding process sequence. First, the welding process sequence is synchronized with the time of the welding equipment in the target scene. The welding parameters such as welding current, voltage, speed, etc. are set according to the order in the welding process sequence, and the welding operations are performed in sequence.

[0068] Further, the tracking and monitoring component is activated to continuously monitor and collect welding control results. The tracking and monitoring component is used to monitor the welding process in real time and collect relevant data, including various sensors and monitoring equipment, such as thermal imaging cameras, photoelectric sensors, etc., to monitor the welding process in real time and collect welding data. Exemplarily, by activating the tracking and monitoring component to continuously collect the molten pool image during the welding process and the weld metal image after cooling, the output is the welding control result.

[0069] Specifically, welding evaluation is performed based on the collected welding control results. This includes comparing the difference between the actual welding result (the shape of the weld metal image after cooling) and the expected welding result (the ideal weld metal shape), and analyzing the grayscale features (information entropy, contrast, etc.) of the molten pool image during the welding process. Exemplarily, the welding control results are analyzed through machine learning models such as K-nearest neighbor (KNN), support vector machine (SVM), decision tree, random forest, neural network, etc. to perform welding evaluation and generate welding feedback information.

[0070] Optionally, welding evaluation is performed to generate welding feedback information. The welding feedback information includes qualitative quality information and quantitative quality information of the welding results. Among them, the qualitative quality information is a qualitative description of the weld quality after welding is completed (such as qualified, unqualified, and the presence of certain defects, etc.); the quantitative quality information is the corresponding quantitative indicators of the weld quality after welding is completed (such as weld strength, defect rate, penetration, excess height, etc.).

[0071] In some implementations, before performing continuous welding control of a target weld based on the welding process sequence, the method further includes:

[0072] The welding process sequence is analyzed, and equivalent processes are merged to obtain a plurality of difference process intervals and a plurality of equivalent process intervals, wherein the difference process intervals and the equivalent process intervals are set at intervals.

[0073] A first difference process interval is extracted, and a first preceding difference metric and a first succeeding difference metric are calculated and acquired according to a first preceding equivalent process interval and a succeeding equivalent process interval of the first difference process interval, respectively.

[0074] Based on a preset smoothing threshold, smoothing judgment is performed in combination with the first preceding difference metric and the first succeeding difference metric, and the welding process sequence is smoothed, wherein the smoothing includes linear smoothing and nonlinear smoothing.

[0075] Specifically, before the continuous welding control of the target weld is performed, first, the welding process sequence is parsed and equivalent process merging is performed. This means that continuous welding processes with the same or similar welding parameters are merged into an equivalent process interval. Then, multiple difference process intervals and multiple equivalent process intervals are obtained. The difference process interval refers to the interval between multiple equivalent process intervals.

[0076] For example, if the first-order welding process is consistent with the second-order welding process, the third-order welding process is consistent with the fourth-order welding process, and the second-order welding process is different from the third-order welding process in the welding process sequence, then the control nodes corresponding to the first-order welding process and the second-order welding process are an equivalent process interval, the control nodes corresponding to the second-order welding process and the third-order welding process are a difference process interval, and the control nodes corresponding to the third-order welding process and the fourth-order welding process are another equivalent process interval.

[0077] Further, the process difference between each difference process interval and its adjacent equivalent process interval is calculated to obtain a first preceding difference metric and a first succeeding difference metric. The difference metric refers to the degree of difference between two welding processes, which is calculated by comparing welding parameters (such as current, voltage, welding speed, etc.). Exemplarily, the distance between process indicators in the above multidimensional parameter space is calculated as the difference metric.

[0078] Specifically, based on a preset smoothing threshold, a smoothing judgment is performed in combination with the first preceding difference metric and the first succeeding difference metric. If the difference metric exceeds the smoothing threshold, the welding process sequence is smoothed to reduce the sudden change of welding parameters, achieve more coherent and optimized welding control, and thus improve welding quality and efficiency. Among them, the smoothing process includes linear smoothing and nonlinear smoothing. Linear smoothing refers to smoothing welding parameters by methods such as linear interpolation, while nonlinear smoothing refers to smoothing welding parameters by methods such as nonlinear interpolation.

[0079] Optionally, the preset smoothing threshold is set based on a dual threshold value, including a first smoothing threshold and a second smoothing threshold, which correspond to linear smoothing and nonlinear smoothing in the smoothing process, respectively. Exemplarily, the first smoothing threshold is set to be less than the second smoothing threshold. If the difference metric exceeds the first smoothing threshold but does not exceed the second smoothing threshold, linear smoothing is performed. Linear smoothing assumes that the change between the two welding processes is linear, and can quickly and efficiently smooth the welding parameters with before and after differences, which is suitable for situations with small differences. If the difference metric exceeds the second smoothing threshold, nonlinear smoothing is performed. Nonlinear smoothing assumes that the change between the two welding processes is nonlinear, which can better reduce the sudden change of welding parameters and achieve more coherent and optimized welding control, which is suitable for situations with large differences.

[0080] The welding process response is performed according to the welding feedback information.

[0081] In some embodiments, performing a welding process response according to the welding feedback information includes:

[0082] Configure difference update constraints, where the difference update constraints include period accumulation constraints, frequency accumulation constraints, and mutation accumulation constraints.

[0083] Difference weight accumulation is performed based on the welding feedback information to obtain a difference accumulation coefficient.

[0084] When the difference accumulation coefficient satisfies the difference update constraint, the historical welding record set is updated according to the welding feedback information to obtain a corrected welding record.

[0085] Based on the revised welding record, a calibration welding process set is redefined.

[0086] Specifically, the difference update constraint is a set of rules used to determine when the welding process needs to be adjusted based on the welding feedback information. It includes period accumulation constraint, frequency accumulation constraint, and mutation accumulation constraint. Among them, the period accumulation constraint is used to stipulate that the update is triggered within a certain time period; the frequency accumulation constraint stipulates that the update is triggered when the frequency of the difference reaches a certain threshold; the mutation accumulation constraint defines the degree of welding quality mutation when the update is triggered.

[0087] Furthermore, the cumulative weight is determined based on the quality quantitative information in the welding feedback information, and the quality qualitative information in the welding feedback information is weighted and accumulated to calculate the difference cumulative coefficient, wherein the difference cumulative coefficient is an indicator that comprehensively reflects the degree of welding difference accumulation. The larger the difference cumulative coefficient, the worse the current welding seam quality. Through the above method steps, the welding quality can be evaluated more accurately.

[0088] Optionally, when the difference accumulation coefficient satisfies the difference update constraint, an update operation is triggered, including updating the welding quality index value in the historical welding record set with welding feedback information, and generating a corrected welding record, which reflects the welding process application effect that meets the target welding scenario, and is used to guide subsequent welding process optimization. Through the above method steps, real-time monitoring and continuous optimization of the welding process can be achieved, thereby continuously improving the quality and efficiency of the weld.

[0089] In some embodiments, performing a welding process response according to the welding feedback information further includes:

[0090] Interactive target welding scenarios and construct welding process response matrix.

[0091] The welding feedback information is parsed, a key feedback indicator set is extracted, and the key feedback indicator set is used as a response constraint to traverse the welding process response matrix for response matching to obtain a welding process response template.

[0092] A multi-path welding process response is performed based on the welding process response template.

[0093] Specifically, a welding response plan set for the target welding scenario is obtained through the data acquisition system, and a welding process response matrix is ​​constructed based on historical response records. The welding process response matrix is ​​a multi-dimensional matrix that records the correspondence and response strategies between different welding processes and possible welding feedback.

[0094] Specifically, the welding feedback information is parsed and a key feedback indicator set is extracted. The key feedback indicator set includes information such as weld defect status, weld shape, color, and surface texture. Then, with the key feedback indicator set as the response constraint, the welding process response matrix is ​​traversed for response matching to obtain a welding process response template. The welding process response template defines a specific processing method and a specific processing path for a weld defect. For example, for a weld with holes, it is necessary to reduce the welding speed, increase the current, or change the welding path.

[0095] Furthermore, the welding process response template performs multi-path welding process response, and processes weld defects that have occurred or may occur in real time, thereby ensuring the quality of weld welding and reducing the losses caused by welding defects.

[0096] Furthermore, the method further comprises:

[0097] Get real-time welding process and extract real-time welding speed.

[0098] An adaptive adjustment strategy is configured, and the forward-looking acquisition step is updated in combination with the real-time welding speed to obtain an adaptive acquisition step, wherein the adaptive acquisition step includes a clearance step and a window step.

[0099] Update acquisition is performed based on the adaptive acquisition step size to obtain dynamic weld profile data.

[0100] Specifically, according to the real-time welding speed, the acquisition step size is adaptively and dynamically adjusted to control the process optimization accuracy and efficiency of the welding process. Exemplarily, the adaptive adjustment strategy includes: if the real-time welding speed is low, the clearance step size is shortened accordingly, the window step size is kept unchanged, or the clearance step size is kept unchanged and the window step size is increased. Among them, shortening the clearance step size can obtain weld profile data closer to the current welding point, so as to monitor the welding process more accurately. Increasing the window step size can increase the amount of data analyzed each time, so as to understand the welding process more comprehensively.

[0101] In summary, the welding process dynamic optimization method combined with tracking monitoring provided by the present invention has the following technical effects:

[0102] By acquiring the historical welding record set, the historical welding record set is parsed based on the welding requirements of the target welding scene to determine the calibrated welding process set; the forward-looking acquisition step is configured, and the weld acquisition component is initialized based on the forward-looking acquisition step, and the weld acquisition component is activated for continuous acquisition to obtain weld contour data; three-dimensional reconstruction is performed based on the weld contour data to generate a displacement weld model, wherein the displacement weld model includes a series of weld sections composed of multiple weld sections; according to the displacement weld model, the weld feature parameter set is extracted, and the weld feature parameter set is used as an index constraint to traverse the calibrated welding process set for welding process matching to obtain the welding process sequence; based on the welding process sequence, continuous welding control of the target weld is performed, and the tracking and monitoring component is activated to continuously monitor and collect welding control results for welding evaluation and generate welding feedback information; welding process response is performed according to the welding feedback information. Thereby, the technical effect of improving the dynamic adjustment capability of the welding process and improving the welding quality and stability is achieved.

[0103] Embodiment 2

[0104] Figure 2 Schematic diagram of the structure of the welding process dynamic optimization system combined with tracking monitoring of the present invention. Figure 1 The flow chart of the welding process dynamic optimization method combined with tracking monitoring in the present invention can be shown as follows: Figure 2 The structure shown is implemented.

[0105] Based on the same concept as the welding process dynamic optimization method combined with tracking and monitoring in the above embodiment, the present invention also provides a welding process dynamic optimization system combined with tracking and monitoring, including:

[0106] The process calibration module 11 is used to obtain a historical welding record set, analyze the historical welding record set based on the welding requirements of the target welding scene, and determine a calibration welding process set.

[0107] The forward-looking acquisition module 12 is used to configure the forward-looking acquisition step length, initialize the weld acquisition component based on the forward-looking acquisition step length, activate the weld acquisition component to perform continuous acquisition, and obtain weld profile data.

[0108] The cross-section reconstruction module 13 is used to perform three-dimensional reconstruction based on the weld contour data to generate a displacement weld model, wherein the displacement weld model includes a weld cross-section series consisting of multiple weld cross sections.

[0109] The process matching module 14 is used to extract a weld feature parameter set according to the displacement weld model, and use the weld feature parameter set as an index constraint to traverse the calibrated welding process set to perform welding process matching and obtain a welding process sequence.

[0110] The tracking and monitoring module 15 is used to perform continuous welding control of the target weld based on the welding process sequence, activate the tracking and monitoring component, continuously monitor and collect welding control results for welding evaluation, and generate welding feedback information.

[0111] The process response module 16 is used to perform welding process response according to the welding feedback information.

[0112] Among them, the process calibration module 11 includes:

[0113] The constraint screening unit is used to traverse the historical welding record set and obtain a calibrated welding record set by taking the welding requirement as a screening constraint.

[0114] The clustering calibration unit is used to extract multiple weld feature parameter sets from the calibration welding record set, and perform cluster analysis on the calibration welding record set based on the multiple weld feature parameter sets to obtain multiple calibration substrate record clusters.

[0115] The center extraction unit is used to obtain multiple cluster centers of the multiple calibration base material record clusters, associate and store the welding processes corresponding to the multiple cluster centers with the weld feature parameter set, and obtain the calibration welding process set.

[0116] In some embodiments, the cross-sectional reconstruction module 13 includes:

[0117] The light processing unit is used to lightly process the weld profile data based on a preset process control granularity and extract a weld profile slice set.

[0118] The three-dimensional reconstruction unit is used to perform three-dimensional reconstruction based on the weld contour slice set to obtain a weld cross-section set.

[0119] A serialization unit is used to serialize the weld section set according to the acquisition timing mark of the weld contour slice set to obtain the weld section series.

[0120] The displacement updating unit is used to store the weld cross-section series in a displacement register, update the weld cross-section series with the real-time update of the weld profile data, and generate the displacement weld model.

[0121] In some embodiments, the tracking and monitoring module 15 includes:

[0122] The process interval division unit is used to analyze the welding process sequence, merge equivalent processes, and obtain multiple difference process intervals and multiple equivalent process intervals, wherein the difference process intervals and the equivalent process intervals are set at intervals.

[0123] The difference metric unit is used to extract the first difference process interval, and calculate and obtain the first preceding difference metric and the first succeeding difference metric according to the first preceding equivalent process interval and the succeeding equivalent process interval of the first difference process interval respectively.

[0124] A smoothing judgment and processing unit is used to perform smoothing judgment based on a preset smoothing threshold in combination with the first preceding difference metric and the first succeeding difference metric, and perform smoothing processing on the welding process sequence, wherein the smoothing processing includes linear smoothing and nonlinear smoothing.

[0125] In some embodiments, the process response module 16 includes:

[0126] The update constraint unit is used to configure the difference update constraint, wherein the difference update constraint includes a period accumulation constraint, a frequency accumulation constraint, and a mutation accumulation constraint.

[0127] The difference weight accumulation unit is used to accumulate the difference weight based on the welding feedback information to obtain the difference accumulation coefficient.

[0128] The welding record correction unit is used to update the historical welding record set according to the welding feedback information to obtain a corrected welding record when the difference accumulation coefficient meets the difference update constraint.

[0129] The recalibration unit is used to redefine the calibration welding process set based on the revised welding record.

[0130] In some embodiments, the process response module 16 further includes:

[0131] The response matrix unit is used to interact with the target welding scenario and construct the welding process response matrix.

[0132] The response matching unit is used to parse the welding feedback information, extract the key feedback indicator set, and use the key feedback indicator set as the response constraint to traverse the welding process response matrix for response matching to obtain the welding process response template.

[0133] The multi-path welding process response unit is used to perform multi-path welding process response based on the welding process response template.

[0134] In some implementations, the system further includes:

[0135] Welding speed acquisition unit, used to obtain real-time welding process,extraction.

[0136] The adaptive adjustment unit is used to configure an adaptive adjustment strategy, update the forward-looking acquisition step length in combination with the real-time welding speed, and obtain an adaptive acquisition step length, wherein the adaptive acquisition step length includes a clearance step length and a window step length.

[0137] The updating acquisition unit is used to perform updating acquisition based on the adaptive acquisition step size to obtain dynamic weld profile data.

[0138] It should be understood that the embodiments mentioned in this specification focus on their differences from other embodiments. The specific embodiments in the aforementioned embodiment one are also applicable to the dynamic optimization system of welding process combined with tracking monitoring described in embodiment two. For the sake of brevity of the specification, they will not be further elaborated here.

[0139] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A dynamic optimization method for welding process combined with tracking monitoring, characterized in that: The method comprises: Obtain a historical welding record set, parse the historical welding record set based on the welding requirements of the target welding scene, and determine a calibration welding process set; Configure the forward-looking acquisition step length, initialize the weld acquisition component based on the forward-looking acquisition step length, activate the weld acquisition component to perform continuous acquisition, and obtain weld profile data; Performing three-dimensional reconstruction based on the weld profile data to generate a displacement weld model, wherein the displacement weld model includes a weld section series consisting of a plurality of weld sections; Extracting a weld feature parameter set according to the displacement weld model, and using the weld feature parameter set as an index constraint, traversing the calibrated welding process set to perform welding process matching and obtain a welding process sequence; Based on the welding process sequence, continuous welding control of the target weld is performed, and the tracking monitoring component is activated to continuously monitor and collect welding control results for welding evaluation and generate welding feedback information; Perform welding process response according to the welding feedback information; Wherein, performing a welding process response according to the welding feedback information includes: Configure difference update constraints, wherein the difference update constraints include period accumulation constraints, frequency accumulation constraints, and mutation accumulation constraints; Performing difference weight accumulation based on the welding feedback information to obtain a difference accumulation coefficient; When the difference accumulation coefficient satisfies the difference update constraint, updating the historical welding record set according to the welding feedback information to obtain a corrected welding record; redefine the calibration welding process set based on the revised welding record; Wherein, performing a welding process response according to the welding feedback information further includes: Interactive target welding scenarios to build welding process response matrix; Parsing the welding feedback information, extracting a key feedback indicator set, and taking the key feedback indicator set as a response constraint, traversing the welding process response matrix for response matching, and obtaining a welding process response template; A multi-path welding process response is performed based on the welding process response template.

2. The method for dynamic optimization of welding process combined with tracking monitoring according to claim 1, characterized in that: Obtain a historical welding record set, parse the historical welding record set based on the welding requirements of the target welding scene, and determine a calibration welding process set, including: Taking the welding requirement as a screening constraint, traversing the historical welding record set to obtain a calibrated welding record set; Extracting a plurality of weld feature parameter sets from the calibration welding record set, and performing cluster analysis on the calibration welding record set based on the plurality of weld feature parameter sets to obtain a plurality of calibration substrate record clusters; A plurality of cluster centers of a plurality of the calibrated substrate record clusters are obtained, and the welding processes corresponding to the plurality of cluster centers are associated and stored with the weld feature parameter set to obtain the calibrated welding process set.

3. The method for dynamic optimization of welding process combined with tracking monitoring as claimed in claim 2, characterized in that: Performing three-dimensional reconstruction based on the weld profile data to generate a displacement weld model includes: Based on a preset process control granularity, the weld profile data is lightly processed to extract a weld profile slice set; Perform three-dimensional reconstruction based on the weld contour slice set to obtain a weld cross-section set; According to the acquisition timing mark of the weld contour slice set, the weld section set is serialized to obtain the weld section series; The weld cross-section series is stored in a displacement register, and along with the real-time update of the weld profile data, the displacement update of the weld cross-section series is performed, and the displacement weld model is generated.

4. The method for dynamic optimization of welding process combined with tracking monitoring according to claim 1, characterized in that: Before performing continuous welding control of a target weld based on the welding process sequence, the method further includes: Analyze the welding process sequence, merge equivalent processes, and obtain a plurality of difference process intervals and a plurality of equivalent process intervals, wherein the difference process intervals and the equivalent process intervals are set at intervals; Extracting a first difference process interval, and calculating and obtaining a first preceding difference metric and a first succeeding difference metric according to a first preceding equivalent process interval and a succeeding equivalent process interval of the first difference process interval respectively; Based on a preset smoothing threshold, smoothing judgment is performed in combination with the first preceding difference metric and the first succeeding difference metric, and the welding process sequence is smoothed, wherein the smoothing includes linear smoothing and nonlinear smoothing.

5. The method for dynamic optimization of welding process combined with tracking monitoring according to claim 1, characterized in that: The method further comprises: Obtain real-time welding process and extract real-time welding speed; Configure an adaptive adjustment strategy, update the forward-looking acquisition step length in combination with the real-time welding speed, and obtain an adaptive acquisition step length, wherein the adaptive acquisition step length includes a clearance step length and a window step length; Update acquisition is performed based on the adaptive acquisition step size to obtain dynamic weld profile data.

6. A dynamic optimization system for welding process combined with tracking monitoring, characterized in that: The system is used to implement the welding process dynamic optimization method combined with tracking monitoring according to any one of claims 1 to 5, and the system comprises: A process calibration module, wherein the process calibration module is used to obtain a historical welding record set, parse the historical welding record set based on the welding requirements of the target welding scene, and determine a calibration welding process set; A forward-looking acquisition module, which is used to configure the forward-looking acquisition step length, initialize the weld acquisition component based on the forward-looking acquisition step length, activate the weld acquisition component to perform continuous acquisition, and obtain weld profile data; A cross-section reconstruction module, the cross-section reconstruction module is used to perform three-dimensional reconstruction based on the weld profile data to generate a displacement weld model, wherein the displacement weld model includes a weld cross-section series consisting of multiple weld cross sections; A process matching module, wherein the process matching module is used to extract a weld feature parameter set according to the displacement weld model, and use the weld feature parameter set as an index constraint to traverse the calibrated welding process set to perform welding process matching and obtain a welding process sequence; A tracking and monitoring module, wherein the tracking and monitoring module is used to continuously control the welding of the target weld based on the welding process sequence, activate the tracking and monitoring component, continuously monitor and collect welding control results for welding evaluation, and generate welding feedback information; A process response module is used to perform welding process response according to the welding feedback information.

Images