A Welding Control Method and Device for Identification Signs

Abstract

The present invention relates to the technical field of welding control, and particularly to a welding control method and device for identification signs. The method includes the following steps: obtaining a high-definition monitoring image of the identification sign to be welded and a welding work log; performing visual recognition of the texture of the identification sign on the high-definition monitoring image of the identification sign to be welded, and simulating the welding visual effect, so as to obtain the welding visual effect of the identification sign; accurately identifying welding points based on the welding visual effect of the identification sign, and planning a dynamic welding path, so as to construct a dynamic welding path; performing real-time welding control on the identification sign to be welded based on the dynamic welding path, and then optimizing the smoothness of the movement trajectory of the welding torch, so as to construct an optimal smooth movement trajectory of the welding torch; obtaining the real-time working parameters of the current welding torch; performing adaptive welding torch parameter adjustment on the real-time working parameters of the current welding torch based on the optimal smooth movement trajectory of the welding torch, so as to obtain an adaptive welding torch adjustment parameter. The present invention improves the welding quality of the identification sign and reduces welding errors.

Description

Technical Field

[0001] The present invention relates to the technical field of welding control, and particularly to a welding control method and device for identification signs. Background Art

[0002] As an important link in the production process of identification signs, sign welding directly affects the quality and production efficiency of the finished products. With the rapid development of industrial automation technology, traditional manual welding methods have gradually revealed problems such as low precision, low efficiency, and easy errors in manual operations, making it difficult to meet the requirements of large-scale and high-precision production. Especially in the production process of identification signs, the control of welding position, angle, and welding force plays a crucial role in the appearance and function of the final product.

[0003] Traditional identification sign welding control methods mostly rely on manual operations and mechanical control, usually using manual or simple mechanical devices for positioning and welding. However, this control method often faces problems such as unstable welding precision, position deviation, and uneven welding. Especially in the scenarios of complex shapes and small batch production, the limitations of traditional control methods are more prominent. In order to improve production efficiency and product quality, automated welding technology has gradually become an important trend in the industry development. Therefore, there is a need to develop a more intelligent identification sign welding control method. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention proposes a welding control method and device for identification signs to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a welding control method for identification signs, including the following steps:

[0006] Step S1: Obtain the high-definition monitoring image of the identification sign to be welded and the welding work log; perform visual recognition of the identification sign texture on the high-definition monitoring image of the identification sign to be welded, and perform welding visual effect simulation to obtain the welding visual effect of the identification sign;

[0007] Step S2: Accurately identify the welding points based on the welding visual effect of the identification sign, and perform dynamic welding path planning to construct a dynamic welding path;

[0008] Step S3: Perform real-time welding control on the identification sign to be welded based on the dynamic welding path, and then perform smoothing optimization on the movement trajectory of the welding torch to construct an optimal smooth movement trajectory of the welding torch;

[0009] Step S4: Obtain the real-time working parameters of the current welding torch; perform adaptive welding torch parameter adjustment on the real-time working parameters of the current welding torch based on the optimal smooth movement trajectory of the welding torch to obtain adaptive welding torch adjustment parameters;

[0010] Step S5: Analyze the changing trend of the welding molten pool temperature based on the real-time welding monitoring image, and dynamically regulate the welding parameters of the adaptive torch adjustment parameters, thereby constructing a dynamic welding parameter regulation strategy;

[0011] Step S6: Analyze local welding defects based on the real-time welding monitoring image, and optimize the welding defect control of the dynamic welding parameter regulation strategy to construct a dynamic welding control optimization model.

[0012] Through high-definition monitoring image acquisition, the system can accurately obtain the texture, shape, and surface state of the signboard, providing detailed basic data for subsequent analysis. Based on the images, a welding visual effect simulation is carried out, which can not only predict potential defects that may occur during the welding process in advance, but also evaluate the impact of different welding parameters on the final effect. This provides data support for the subsequent welding path planning and parameter adjustment, ensuring that the welding process can proceed as expected. The pre-simulation of the welding visual effect helps identify potential quality problems and avoid unexpected defects during the actual welding process, thereby improving the controllability of welding quality. Through the analysis of the visual effect, the system can accurately identify each welding point on the signboard and determine the priority and difficulty of each welding point, ensuring the optimization of the welding sequence and method. The dynamic planning of the welding path ensures the real-time and targeted selection of the path. By processing the visual information of the area to be welded, the welding path can not only be flexibly adjusted according to the complexity of different parts, but also effectively avoid potential obstacles or influences, reducing the probability of rework during the welding process. The dynamic path planning makes the welding process more efficient, and the welding torch does not need to pass through unnecessary repetitive paths, thus saving welding time and improving production efficiency. Combined with the real-time dynamic path, the welding system can perform immediate control to ensure that the welding torch is always in the best working state during the welding process. Real-time adjustment can avoid unstable factors caused by path changes, thereby improving the consistency and stability of welding. The smooth optimization of the movement trajectory not only reduces vibrations and oscillations during the welding process, but also reduces welding defects caused by uneven acceleration and deceleration, improving the quality of the welding joint. The smoothness of the trajectory also reduces mechanical wear on the equipment and extends the service life of the equipment. The optimized movement trajectory of the welding torch can ensure higher precision during the welding process, enabling each welding point to meet high standards and avoiding quality problems caused by excessive or insufficient welding. By obtaining real-time working parameters, the system can accurately monitor the state of the welding torch (such as temperature, current, air flow, etc.) to ensure that the welding torch is always under ideal working conditions during the welding process. By adaptively adjusting parameters according to the optimal movement trajectory of the welding torch, the welding system can respond in real time to different working environments and actual welding situations, automatically adjusting key parameters such as welding current and speed to ensure that the heat distribution and material fusion during the welding process reach the optimal state. Adaptive adjustment ensures the stability of the welding process, especially in complex or high-precision welding scenarios, avoiding errors caused by manual parameter setting and improving the quality consistency of each welding. By analyzing the temperature change trend of the real-time welding monitoring image, the stability of the molten pool during the welding process can be accurately judged. If it is found that the temperature is too high or too low, the system can quickly react and adjust the welding parameters to avoid overheating or insufficient heat, ensuring welding quality. Based on the temperature change trend, the system can achieve dynamic regulation of welding parameters, avoiding the limitations that may be caused by fixed parameter settings in traditional welding methods. Dynamic adjustment enables the welding process to respond more precisely to different welding conditions.By continuously adjusting welding parameters, optimizing the welding effect, avoiding defects during the welding process, the overall production efficiency and welding quality are improved. Through real-time image monitoring and welding defect analysis, the system can quickly detect defects such as pores, cracks, and lack of fusion that may occur during the welding process, provide timely feedback and make corrections. Combining the defect analysis results, the system can finely optimize the dynamic welding parameter control strategy to avoid the recurrence of the same defects and enhance the welding quality. By continuously accumulating data and optimizing the defect control strategy, an efficient dynamic welding control optimization model is established, which can be applied in different production batches to continuously improve the welding quality and stability and meet higher production standards.

[0013] Preferably, step S1 includes the following steps:

[0014] Step S11: Obtain the high-definition monitoring image of the signboard to be welded and the welding work log;

[0015] Step S12: Conduct visual recognition of the signboard texture on the high-definition monitoring image of the signboard to be welded to generate signboard texture visual features;

[0016] Step S13: Sharpen and enhance the details of the signboard texture visual features to obtain a sharpened and enhanced signboard image;

[0017] Step S14: Optimize the contrast of multiple regions of the sharpened and enhanced signboard image to obtain a contrast-optimized signboard image;

[0018] Step S15: Based on the welding work log, simulate the welding visual effect of the contrast-optimized signboard image to obtain the signboard welding visual effect.

[0019] Through the acquisition of high-definition monitoring images, the system can obtain detailed texture information, morphology, and status of the sign surface. The high-definition images provide clear and accurate raw data for subsequent image processing, texture recognition, and analysis. The welding work log contains various parameters and settings during the welding process (such as current, voltage, speed, etc.), providing necessary background data for subsequent welding simulation and optimization, making the welding control more precise and targeted. Through texture visual recognition technology, the system can effectively identify various texture features on the sign surface, such as surface roughness, texture direction, and shape. These features are crucial for heat conduction, material fusion, etc. during the welding process and can provide a basis for subsequent analysis. The extracted texture features can provide detailed basis for subsequent welding path planning, parameter adjustment, etc., ensuring a high degree of adaptation of the welding process to the material surface features and avoiding welding problems caused by uneven or irregular surfaces. The detail sharpening and enhancement technology can highlight the subtle features in the image, making each texture, defect, and tiny detail of the sign more clearly visible. This is very important for fine welding control, especially in complex sign structures, and can help identify potential small defects. The sharpened and enhanced image is clearer and more detailed, facilitating subsequent processing steps such as contrast optimization and defect analysis, making the entire image processing chain more efficient, and enabling the monitoring, analysis, and optimization during the welding process to achieve higher precision. Through multi-region contrast optimization, the system can perform personalized adjustments according to different regions of the sign image, improving the visual clarity of different regions. For the welding area, surrounding area, and other key positions, the system can specifically enhance or optimize the contrast, thus ensuring that the details and differences in each region of the image can be accurately captured. The optimized contrast image can more clearly display the details of the welding area, contributing to welding quality analysis, precise identification of welding points, early detection of defects, etc., enhancing the sensitivity and accuracy of anomaly detection during the welding process. Combining the welding work log for welding visual effect simulation enables the system to simulate the welding visual effect before welding, including the smoothness of the weld, color change, molten pool formation, etc. This provides an intuitive preview of the welding effect for operators and the system. The simulation can help predict possible problems, such as overheating, uneven welding, etc. Through the simulated welding visual effect, the system can evaluate the impact of different welding parameters on the welding effect, and thus make optimization adjustments. The simulation results provide precise basis for parameter selection in the subsequent welding process, reducing the errors that may occur in actual operation. Through visual simulation and parameter optimization, potential problems during the welding process can be discovered and corrected before actual welding, avoiding the waste of rework and repair in actual welding, and enhancing the efficiency and quality of the welding process.

[0020] Preferably, the specific steps of step S15 are as follows:

[0021] Conduct a welding process requirement analysis on the welding work log to generate the welding process requirement characteristics of the signboard to be welded;

[0022] Identify the welding process flow in the welding work log and extract the multi-stage welding process flow;

[0023] Locate the key identification nodes of the contrast-optimized signboard image based on the welding work log to obtain multiple key identification nodes;

[0024] Conduct an analysis of the spatial distribution of the identification points for multiple key identification nodes to generate spatial distribution data of the identification nodes;

[0025] Calculate the welding process parameters for each identification node based on the welding process requirement characteristics of the signboard to be welded, thereby generating the welding process parameters for each identification node;

[0026] Simulate the welding visual effect of each identification node based on the multi-stage welding process flow to obtain the welding visual effect of the signboard.

[0027] Through the analysis of welding work logs, the system can identify and extract the welding process requirements of identification plates, such as key factors like welding current, welding speed, heat input, etc. According to the characteristics of the identification plate, such as material, thickness, shape, etc., corresponding welding process requirement features are generated, providing an accurate basis for subsequent welding control. The analysis of welding process requirements helps the system understand the welding requirements of different identification plates, enabling each welding task to be customized according to actual needs, avoiding the welding inadaptability caused by unified parameter settings, and ensuring the efficiency and precision of each welding task. Through in-depth analysis of process requirements, potential process problems can be identified early, avoiding situations such as overheating or uneven welding during welding, and improving the stability and final quality of welding. By identifying and extracting the welding process flow, the system can divide the welding process into multiple stages (such as preheating, welding, cooling, etc.) and set different control parameters for each stage. This enables each stage to be optimized according to specific requirements, improving the overall efficiency of the welding process. The identification of multi-stage process flow can help the system evaluate the operation requirements of different stages and set the most suitable process parameters at each stage, thereby reducing unnecessary energy waste and improving the stability of welding. The precise identification of multi-stage processes provides a clear basis for controlling temperature, time, speed, etc. at each stage, contributing to the full-process supervision of the welding process and ensuring that every step in the process execution strictly complies with standards. Using the image optimized by contrast, the system can accurately locate the key welding nodes on the identification plate, such as seam points, corners, or special-shaped areas, etc. These nodes are usually the parts that require the most attention during the welding process, and accurately identifying them helps to finely control the welding quality. By accurately positioning the key nodes, the welding system can concentrate resources and efforts to conduct key welding control on these key parts, avoiding quality fluctuations caused by unclear nodes and improving the overall welding accuracy. Through the spatial distribution analysis of key identification nodes, the system can understand the spatial positions and relative relationships of these nodes on the identification plate. This analysis helps to determine the welding sequence, control the accuracy of welding positions, and optimize the welding path planning. Through the generated spatial distribution data, the system can calculate the optimal welding path, avoiding repeated movements or overheat input areas, while reducing errors during the welding process and improving efficiency. Based on the spatial distribution data and welding process requirement features of each identification node, the system can calculate the welding process parameters (such as welding current, welding speed, welding time, etc.) for each node one by one. This precise calculation ensures that each welding node can be welded with the most suitable parameters, avoiding welding quality fluctuations. Welding nodes with different positions and characteristics may have different requirements for parameters such as heat input and welding speed. Calculating and adjusting the welding process parameters one by one can ensure that each welding point can achieve the best welding effect. The precisely calculated process parameters ensure the welding quality of each welding node, avoiding defects caused by improper parameter settings, and improving the controllability of the welding process and the quality of the final product.By inputting the welding process parameters of each identification node into the multi-stage welding process flow, the system can simulate the welding process and predict the welding effect of each node. This prediction helps to identify in advance possible welding defects during the welding process, such as overheating, cracks, pores, etc. By simulating the welding effect, the system can adjust the welding process parameters in real time to further optimize the welding process. For example, when the welding effect is not ideal, the current or welding speed can be adjusted according to the simulation results to achieve the best welding effect.

[0028] Preferably, the specific steps of step S2 are as follows:

[0029] Step S21: Accurately identify welding points based on the welding visual effect of the identification plate to obtain multiple welding points;

[0030] Step S22: Optimize the identification plate image according to the contrast to perform image visual positioning and marking on multiple welding points to obtain the positioning coordinates of each welding point in the image;

[0031] Step S23: Analyze the three-dimensional structure of the identification plate according to the contrast-optimized identification plate image to generate the three-dimensional geometric information of the identification plate;

[0032] Step S24: Based on the three-dimensional geometric information of the identification plate, perform dynamic welding path planning on the positioning coordinates of each welding point in the image, thereby constructing a dynamic welding path.

[0033] Through the analysis of the welding visual effect, the system can accurately identify each welding point on the sign. This step extracts the position information of the welding points through image analysis and ensures that each welding point has a clear and definite identification in the visual effect, thus ensuring the accuracy of the welding quality. After accurately identifying the welding points, the welding path, welding parameters, etc. can be optimized and adjusted according to each welding point. For complex signs, it can ensure that even tiny welding points can be precisely welded. With the accurate identification of the welding points, the situation of manual identification errors or omissions is avoided, thereby reducing the risk of miswelding or missed welding and improving the overall consistency of the welding effect. Through image vision positioning marks, the system can obtain the accurate coordinate information of each welding point in the image. Through the image optimized by contrast, the boundary of the welding point is clearer, making the positioning more accurate. The acquisition of the positioning coordinates is not only a visual identification but also can convert the image data into real space coordinates, which is crucial for the welding control system and can accurately guide the welding operation. Through three-dimensional structure analysis, the system not only obtains the two-dimensional image information of the sign but also can accurately understand the three-dimensional geometric shape of the sign, including important information such as its thickness, curved surface shape, angle, etc. This provides the necessary spatial information support for the subsequent welding path planning. The three-dimensional geometric information of the sign helps the welding system better adapt to complex geometric shapes and avoid welding defects caused by irregular or curved surface areas. The system can optimize and adjust the welding process parameters according to these geometric features. By combining the three-dimensional geometric information of the sign with the positioning coordinates of the welding points, the system can generate a dynamic welding path. The path planning not only considers the accurate position of the welding points but also dynamically adjusts the welding path according to the overall shape of the sign, avoiding the limitations of traditional static path planning. The dynamic path planning enables the welding process to adjust the welding path in real time, avoiding unnecessary multiple operations or path overlaps and maximizing the welding efficiency and accuracy. Especially for signs with complex shapes, the dynamic path planning can effectively cope with the challenges of various geometric features.

[0034] Preferably, the specific steps of step S3 are as follows:

[0035] Step S31: Perform real-time welding control on the sign to be welded based on the dynamic welding path and collect real-time welding monitoring images;

[0036] Step S32: Identify the actual welding contact points in the real-time welding monitoring images and mark the actual welding contact points;

[0037] Step S33: Analyze the surface change of the welding points for the actual welding contact points to generate the high-temperature welding surface change characteristics of the welding points;

[0038] Step S34: Calculate the change in the welding curvature of the identification plate for the high-temperature welding surface change characteristics of the welding points, thereby generating the data on the change in the welding curvature of the identification plate;

[0039] Step S35: Conduct a welding path mutation analysis on the dynamic welding path based on the data on the change in the welding curvature of the identification plate, and mark the welding path mutation points;

[0040] Step S36: Optimize the smoothness of the torch movement trajectory based on the welding path mutation points, thereby constructing the optimal smooth trajectory of the torch movement.

[0041] Through dynamic welding paths and real-time welding control, the system can respond in a timely manner according to changes in welding points and path adjustments. This real-time control can ensure that the welding path and parameters always match the actual condition of the signboard, guaranteeing a stable and efficient welding process. Collecting real-time welding monitoring images provides accurate image data for subsequent welding point analysis, surface change analysis, etc. These image data help the system continuously monitor the welding quality and promptly detect potential problems, such as welding contact point deviations, pores, cracks and other welding defects. The system accurately identifies the actual welding contact points through image analysis algorithms. These points are usually key areas with the largest heat input and potential defects during the welding process. Precise identification of the contact points can ensure more targeted subsequent control. After marking the welding contact points, the system can track and monitor the status of these points, including welding quality, heat changes, etc. Ensure that each welding point is effectively managed, reducing the risk of omission or miswelding. During the welding process, the surface change in the high-temperature area is an important indicator of welding quality. Analyzing the surface change of the welding points can reveal the impact of heat input on the welding area, avoiding overheating or uneven temperature distribution. Through the analysis of the surface change of the welding points, the system can evaluate the heat-affected zone of the welding in real time. Through the surface change characteristics of the welding points, the system can better predict potential problems during the welding process, such as hot cracks, deformation, etc. Ensure proper heat input for each welding point, thus avoiding welding defects. Curvature change calculation can help analyze the surface changes during the welding process, especially in the area around the welding points. During the welding process, surface changes are closely related to the thermal expansion and contraction of the material. Curvature change can reveal the temperature changes and physical reactions around the welding points. Through the change in welding curvature, the system can monitor the temperature, stress distribution and thermal deformation during the welding process in real time, and promptly identify potential welding defects, such as warping, deformation, etc. By calculating the curvature change, the system can adjust the welding process to ensure that each welding point during the welding process meets the required geometric standards, avoiding quality problems caused by excessive curvature changes and improving the stability and reliability of welding quality. By analyzing the welding curvature change data, the system can identify the mutation points on the welding path. Path mutations are usually hot spots for errors during the welding process, which may lead to welding defects, such as heat stress concentration, material deformation, etc. Identifying these mutation points helps avoid the formation of incorrect paths. After marking the mutation points of the welding path, the system can adjust the welding path in real time, avoid quality problems caused by transition areas, and optimize the smoothness and efficiency of the welding path. Through the analysis and adjustment of the mutation points of the welding path, the smoothness of the path during the welding process is ensured, unnecessary process fluctuations are avoided, and the stability of welding is further improved. The optimization of the torch movement trajectory ensures a smooth transition of the torch movement near the mutation points of the welding path. By avoiding sharp movement changes, not only are welding defects reduced during the welding process, but also the stress concentration problem in the transition area is avoided.The smooth optimization of the welding torch movement trajectory can minimize welding defects caused by unsmooth trajectory changes, maintain the consistency of the welding process, and thus improve the welding quality and production efficiency.

[0042] Preferably, the specific steps of step S4 are as follows:

[0043] Step S41: Obtain the real-time working parameters of the current welding torch;

[0044] Step S42: Perform multi-stage trajectory segmentation on the optimal welding torch movement smooth trajectory to obtain multi-stage welding trajectories;

[0045] Step S43: Calculate the welding process parameters for each segment of the multi-stage welding trajectory according to the welding visual effect of the identification plate to obtain the welding process parameters for each segment of the trajectory;

[0046] Step S44: Calculate the welding current for the real-time working parameters of the current welding torch to obtain the real-time welding current of the current welding torch;

[0047] Step S45: Identify the welding speed of the welding torch for the real-time working parameters of the current welding torch and extract the welding speed of the current welding torch;

[0048] Step S46: Based on the welding process parameters of each segment of the trajectory, perform adaptive welding torch parameter adjustment on the real-time welding current of the current welding torch and the welding speed of the current welding torch to obtain the adaptive welding torch adjustment parameters.

[0049] The present invention obtains the real-time working parameters of the welding torch (such as current, welding speed, temperature, etc.) to provide accurate data for subsequent adaptive adjustment. This enables the entire welding process to dynamically adjust parameters according to the actual working conditions to meet different welding requirements. By obtaining parameters in real time, possible deviations or problems in the welding process can be detected in a timely manner and adjusted to ensure the stability and accuracy of the welding process. Dividing the welding torch movement trajectory into multiple stages helps optimize the process parameters separately for each stage. By refining each segment of the trajectory, the welding accuracy can be improved, and large-scale trajectory changes can be avoided from affecting the welding quality. After segmenting the welding trajectory, the system can dynamically adjust the welding parameters according to the specific requirements of each stage, enhancing the accuracy and flexibility of welding control. Through multi-stage trajectory segmentation, the welding process can be adjusted according to the characteristics of each segment of the trajectory, further reducing the influence of welding defects and thermal stress, and ensuring that each segment of welding achieves the best effect. For each segment of the welding trajectory, the system can perform refined calculation of welding process parameters to ensure that each stage meets specific process requirements. This helps avoid parameter mismatches during the entire welding process and improves the welding accuracy. Combining the welding visual effect of each segment of the trajectory with the identification sign helps the system predict the welding effect of each stage and make corresponding adjustments to ensure that the welding quality is not affected by changes between stages. Through accurate parameter calculation for each segment of the trajectory, the process consistency of the entire welding process can be ensured, and the stability and reliability of the welding quality can be improved. By calculating the welding current of the welding torch in real time, the change of current during the welding process and its impact on the welding quality can be accurately judged. Current is a key factor affecting the heat input and the quality of the welded joint during the welding process, and real-time monitoring helps avoid welding problems caused by current fluctuations. By accurately calculating the current value at present, welding defects (such as poor welding, uneven welding pool, etc.) caused by excessive or too small current can be avoided. According to the real-time welding current value, the system can quickly feedback and dynamically adjust the welding parameters to optimize the welding process and improve the welding quality. Welding speed is an important factor affecting the welding effect. Real-time identification and extraction of the welding speed can effectively judge the heat input and cooling rate during the welding process, and avoid the negative impacts (such as insufficient welding or uneven weld seams) caused by too fast or too slow welding speed. By accurately identifying the welding speed of the welding torch, the system can make timely adjustments to ensure the optimization of the welding speed in each segment of the welding process. According to the welding process requirements of each segment of the trajectory, the system will adaptively adjust key parameters such as welding current and welding speed. This automatic adjustment ensures that the welding process always meets the optimal process requirements, thus avoiding errors caused by manual adjustment. Adaptive adjustment can accurately control the welding parameters according to the characteristics of different trajectory stages, effectively avoiding welding defects such as incomplete welding and overheating, and ensuring the quality of each segment of welding. By dynamically adjusting the parameters of the welding torch, the system can maintain an efficient and stable working state in a complex welding environment, further improving the welding quality and efficiency.

[0050] Preferably, the specific steps of step S5 are as follows:

[0051] Step S51: Analyze the heat distribution in the welding area of the real-time welding monitoring image to generate the heat distribution characteristics of the welding area;

[0052] Step S52: Evaluate the state of the welding molten pool based on the heat distribution characteristics of the welding area to obtain the state characteristics of the welding molten pool;

[0053] Step S53: Analyze the temperature change trend of the state characteristics of the welding molten pool to generate the temperature change trend of the welding molten pool;

[0054] Step S54: Perform abnormal temperature prediction evolution based on the temperature change trend of the welding molten pool, thereby generating abnormal temperature prediction data of the welding molten pool;

[0055] Step S55: Dynamically regulate the welding parameters of the adaptive torch adjustment parameters based on the abnormal temperature prediction data of the welding molten pool, thereby constructing a dynamic welding parameter regulation strategy.

[0056] Through the analysis of real-time welding monitoring images, the system can accurately understand the heat distribution in the welding area, which provides key data for subsequent welding parameter adjustment. By analyzing the heat distribution in the welding area, problems such as overheating or uneven heat distribution can be avoided, ensuring that the heat input during the welding process meets expectations, thereby improving the welding quality. Accurately grasping the heat distribution in the welding area helps predict possible thermal stresses and prevent welding defects (such as cracks, pores, etc.) caused by uneven temperatures. By analyzing the heat distribution characteristics in the welding area, the system can evaluate the state of the welding molten pool in real time, which is crucial for controlling the quality and uniformity of welding. The state of the welding molten pool directly affects the welding quality. The system can avoid unstable or incomplete welding caused by an oversized or undersized molten pool by evaluating the molten pool state. Real-time monitoring and evaluation of the molten pool state ensure that every link in the welding process is in the best state, thereby enhancing the controllability and stability of the entire welding process. By analyzing the characteristics of the welding molten pool state, the system can predict the change trend of the molten pool temperature, which helps to adjust the heat input during the welding process in real time to avoid too high or too low temperatures. Temperature is a key factor affecting welding quality. The accurate prediction of the temperature change trend can help optimize the welding process and avoid welding defects (such as overburning, insufficient melting, etc.) caused by inaccurate temperature control. The analysis of the temperature change trend helps to precisely control the heat input during the welding process, maintain the uniformity and stability of welding, and avoid quality problems caused by excessive fluctuations. By predicting the temperature change trend of the welding molten pool, the system can timely detect abnormal temperature fluctuations and give early warnings, which provides effective preventive measures for adjusting welding parameters and avoiding quality problems. Overheating may lead to overburning or welding defects, and too cold a temperature may lead to insecure welding. By predicting abnormal temperatures, intervention can be carried out before problems occur to ensure the stability of the welding process. Based on the predicted data of abnormal temperatures in the welding molten pool, the system can automatically dynamically adjust the working parameters of the welding torch (such as current, welding speed, welding voltage, etc.) to ensure that every stage in the welding process is in the best temperature control state. By dynamically adjusting welding parameters, abnormal temperatures occurring during the welding process can be dealt with in real time, optimizing the heat input, and ensuring that the welded joints are uniform and strong. The dynamic control strategy enables the parameter adjustment in the welding process to quickly respond to temperature changes, avoiding adverse effects on welding quality caused by excessive temperature changes, thereby ensuring the stability and high-quality output of the welding process.

[0057] Preferably, the specific steps of step S6 are as follows:

[0058] Step S61: Identify the welding results of multiple segments based on real-time welding monitoring images and extract the welding results of multiple process stages;

[0059] Step S62: Evaluate the welding quality of each welding point for the welding results of multiple process stages and generate a welding quality evaluation value for each welding point;

[0060] Step S63: Based on the welding quality evaluation value of each welding point, perform local welding defect analysis to generate local welding defect data;

[0061] Step S64: Based on the local welding defect data, optimize the welding defect control of the dynamic welding parameter regulation strategy to construct a dynamic welding control optimization model.

[0062] The present invention conducts multi-stage analysis on the welding process through real-time welding monitoring images, can accurately identify the welding results of each process stage, so as to comprehensively master the process progress and welding effect of each stage. The results of each welding stage are independently extracted and identified, thus ensuring the quality control of each link in the welding process. The welding effects of different stages can be evaluated separately, avoiding the problem of a certain link affecting the overall welding quality. Through the quality evaluation of each welding point, the quality of different welding points can be accurately judged, avoiding potential errors brought by overall evaluation, which can more carefully discover possible quality problems. The generation of the welding quality evaluation value provides a quantitative quality index for each welding point, helps to clarify which welding points meet the standards and which may have defects, so as to make targeted adjustments. The welding point quality evaluation value provides strong data support for subsequent defect analysis and optimization control, making the positioning of welding defects more accurate. Through the quality evaluation of each welding point, local defects (such as pores, cracks, irregular penetration, etc.) that may occur during the welding process can be detected early, which can conduct defect analysis immediately after welding is completed, avoiding large-scale quality problems. Local defect analysis can accurately identify which welding points or areas have problems, thus providing clear guidance for subsequent repair and adjustment. Based on the generation of defect data, it can effectively guide the adjustment of the welding process, improve parameters such as temperature control and current to reduce the occurrence of similar defects and improve the overall welding quality. Through the analysis of local welding defect data, the links with problems in the welding process, such as welding current, welding speed, heat input, etc., can be adjusted, so as to achieve targeted defect control and optimize the welding quality. Based on the defect control optimization model, the system can automatically adjust the welding parameters during the welding process to cope with different types of defects. This dynamic adjustment can adapt to welding environment and process changes in real time, thus ensuring the stability of welding quality.

[0063] In this specification, a welding control device for a signboard is provided, which is used to execute the welding control method for a signboard as described above, and includes:

[0064] A welding vision simulation module, configured to obtain a high-definition monitoring image of the signboard to be welded and a welding work log; perform visual texture recognition on the high-definition monitoring image of the signboard to be welded and simulate the welding vision effect, so as to obtain the welding vision effect of the signboard;

[0065] The welding path planning module is used to accurately identify welding points based on the welding visual effect of the signboard, and perform dynamic welding path planning, so as to construct a dynamic welding path;

[0066] The welding torch movement trajectory module is used to perform real-time welding control on the signboard to be welded based on the dynamic welding path, and then smooth and optimize the movement trajectory of the welding torch, so as to construct the optimal smooth trajectory of the welding torch movement;

[0067] The adaptive parameter adjustment module is used to obtain the current real-time working parameters of the welding torch; based on the optimal smooth trajectory of the welding torch movement, perform adaptive welding torch parameter adjustment on the current real-time working parameters of the welding torch, so as to obtain the adaptive welding torch adjustment parameters;

[0068] The dynamic welding control module is used to analyze the change trend of the welding pool temperature according to the real-time welding monitoring image, and perform dynamic welding parameter control on the adaptive welding torch adjustment parameters, so as to construct a dynamic welding parameter control strategy;

[0069] The welding control optimization module is used to analyze local welding defects based on the real-time welding monitoring image, and optimize the welding defect control of the dynamic welding parameter control strategy to construct a dynamic welding control optimization model.

[0070] Through high-definition monitoring images and precise texture recognition, the present invention can provide accurate visual data for subsequent welding. The welding work log combined with welding records ensures that all welding parameters required in the simulation process are fully considered. Through the simulation of welding visual effects, potential visual defects such as uneven penetration depth and excessive heat-affected zone can be detected in advance before welding, reducing the possibility of rework after welding. Through precise visual simulation and welding effect analysis, the position of the welding point can be accurately identified, avoiding welding inaccuracies caused by human errors in traditional methods. Based on the precise positioning of the welding point, the dynamic path planning module can flexibly adjust the welding path according to different welding requirements (such as sign size, material properties, etc.) to ensure that each welding point can be precisely welded. The dynamic planned path avoids unnecessary repetitive operations, reduces welding time, and improves welding quality at the same time. The smooth optimization of the welding torch movement trajectory can avoid vibrations or oscillations during welding, reducing local defects (such as cracks, pores, etc.) that occur during welding. Through precise trajectory control, the movement of the welding torch is optimized, thereby improving the stability of the welding process and reducing human operation errors. During the welding process, the real-time working parameters of the welding torch (such as current, voltage, welding speed, etc.) will be dynamically adjusted according to real-time monitoring data and movement trajectory. As the welding environment and materials change, the system can adaptively adjust the parameters to ensure that the quality of each welding reaches the best level. Through the analysis of the temperature change trend, the system can timely capture abnormal changes in the temperature of the welding molten pool, issue warnings and make adjustments. Precise control of temperature is crucial for welding quality. The dynamic regulation module can adjust welding parameters through the temperature change trend to optimize the shape and depth of the welding molten pool and ensure welding quality. Through real-time image monitoring, local detection of welding defects (such as pores, cracks, incomplete penetration, etc.) can be carried out and timely feedback can be provided. Based on the defect detection, the system automatically adjusts welding parameters, optimizes the welding process, eliminates or reduces defects, and improves the final welding quality. Through optimizing welding control, welding defects are effectively reduced, the rework rate is lowered, and production efficiency and the stability of the welding process are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a schematic flow chart of the steps of a welding control method for a sign according to the present invention;

[0072] Figure 2 It is a schematic detailed implementation step flow chart of step S1;

[0073] Figure 3 It is a schematic detailed implementation step flow chart of step S2. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0074] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0075] Embodiments of the present application provide a method and a device. The execution subjects of the welding control method and device for identification signs include, but are not limited to, mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc. that carry this system, which can be regarded as general computing nodes of the present application. The data processing platform includes, but is not limited to, at least one of an audio and image management system, an information management system, and a cloud data management system.

[0076] Please refer to Figures 1 to 3 , the present invention provides a welding control method for identification signs, and the method includes the following steps:

[0077] Step S1: Obtain the high-definition monitoring image of the identification sign to be welded and the welding work log; perform visual recognition of the identification sign texture on the high-definition monitoring image of the identification sign to be welded, and perform welding visual effect simulation to obtain the welding visual effect of the identification sign.

[0078] Step S2: Based on the welding visual effect of the identification sign, accurately identify the welding points and perform dynamic welding path planning to construct a dynamic welding path.

[0079] Step S3: Based on the dynamic welding path, perform real-time welding control on the identification sign to be welded, and then perform smoothing optimization of the welding torch movement trajectory to construct an optimal smooth trajectory of the welding torch movement.

[0080] Step S4: Obtain the real-time working parameters of the current welding torch; perform adaptive welding torch parameter adjustment on the real-time working parameters of the current welding torch based on the optimal smooth trajectory of the welding torch movement to obtain adaptive welding torch adjustment parameters.

[0081] Step S5: Analyze the change trend of the welding pool temperature according to the real-time welding monitoring image, and perform dynamic regulation of the welding parameters on the adaptive welding torch adjustment parameters to construct a dynamic welding parameter regulation strategy.

[0082] Step S6: Based on the real-time welding monitoring image, perform local welding defect analysis, and perform welding defect control optimization on the dynamic welding parameter regulation strategy to construct a dynamic welding control optimization model.

[0083] In the embodiments of the present invention, refer to Figure 1 , which is a schematic flow chart of the steps of a welding control method for identification signs of the present invention. In this embodiment, the steps of the welding control method for identification signs include:

[0084] Step S1: Obtain the high-definition monitoring image of the identification sign to be welded and the welding work log; perform visual recognition of the identification sign texture on the high-definition monitoring image of the identification sign to be welded, and perform welding visual effect simulation to obtain the welding visual effect of the identification sign.

[0085] In this embodiment, a high-resolution camera (such as a single-lens reflex camera or an industrial camera) is used to photograph the sign to be welded. Ensure that the photographing is carried out under good lighting conditions to avoid the influence of shadows and reflections on the image quality. Set the camera parameters (such as shutter speed, ISO, aperture, etc.) to ensure the clarity of the image and the retention of details. At the same time, a tripod can be used to fix the camera position to ensure the consistency of the shooting angle. Before welding starts, record the welding work log, including welding materials, welding process parameters (such as current, voltage, welding speed, etc.), environmental conditions (temperature, humidity), and the expected welding effect. This information will provide a reference for subsequent analysis. The work log can be recorded using a spreadsheet or specialized welding record software to ensure the systematicness and traceability of the information. Preprocess the collected high-definition monitoring images to improve the accuracy of subsequent texture recognition, including steps such as denoising, color correction, and contrast enhancement. Image processing software (such as OpenCV or MATLAB) can be used for image preprocessing. Apply filters to remove background noise, adjust the brightness and contrast of the image to ensure that texture details are clearly visible. Apply computer vision technology to identify the texture of the sign. Use feature extraction algorithms (such as SIFT or ORB) to extract texture feature points from the image. These feature points will be used for subsequent visual effect simulation. After feature extraction, use a matching algorithm (such as FLANN or BFMatcher) to verify the effectiveness of the texture features to ensure that the extracted texture can accurately reflect the actual situation of the sign. Based on the extracted texture features and the process parameters in the welding work log, simulate the welding visual effect. This can be achieved through computer graphics technology. Use 3D modeling software (such as Blender or Maya) to build a virtual model of the welding process. During the simulation, the temperature distribution in the welding heat-affected zone, the flow of molten metal, and the influence of the cooling process after welding should be considered. These factors will directly affect the final welding effect. Compare the simulated welding visual effect with the actual sign image, record the differences and analyze the reasons. This can help identify potential welding quality problems and provide a basis for subsequent process adjustment. Record the simulation results and analysis conclusions for reference and adjustment during subsequent welding processes. All data should be stored in a database to ensure the traceability of information.

[0086] Step S2: Based on the welding visual effect of the sign, accurately identify the welding points and perform dynamic welding path planning, thereby constructing a dynamic welding path;

[0087] In this embodiment, an image processing algorithm is used to analyze the image and identify the characteristics of the welding points. Edge detection algorithms (such as Canny edge detection) and contour extraction techniques can be adopted to clarify the specific positions of the welding points. Through an image processing software (such as OpenCV), the image of the welding visual effect is analyzed to extract the coordinate information and relevant characteristics of the welding points. These characteristics include the size, shape, and position of the welding points, providing basic data for subsequent path planning. Feature matching algorithms (such as template matching or SURF) are applied to further confirm the positions of the welding points. The extracted features are compared with the preset welding point templates to ensure the accuracy of identification. After confirming the welding points, the coordinates (X, Y, Z) of each welding point are recorded and stored in a data structure for easy calling during subsequent path planning. According to the identified welding point information, a dynamic welding path is designed. Path planning algorithms (such as the A* algorithm or Dijkstra algorithm) are used to calculate the optimal path from the starting point to each welding point. During the path planning process, obstacles that may exist during the welding process (such as other equipment or workpieces) are considered to ensure that the planned path can guarantee the smooth movement of the welding torch and avoid collisions. After the preliminary path planning is completed, the path is optimized to reduce the welding time and improve the welding quality. For example, path smoothing algorithms (such as Bezier curves or spline curves) can be used to optimize the welding path to make it more smooth. During this process, adjustments to the welding speed and welding sequence should also be considered to ensure the best welding effect and heat input control. During the welding process, the status of the welding points and changes in the surrounding environment are monitored in real time. Based on sensor feedback and real-time image analysis, the welding path is dynamically adjusted, and a closed-loop control system can be implemented. When a welding point position deviation or environmental change is detected, the system automatically calculates a new welding path to ensure that the welding quality is not affected. After completing the dynamic welding path planning, path verification and testing are carried out. The welding process is simulated through simulation software (such as ROS or Gazebo) to ensure that the planned path is feasible in actual operation and meets the welding process requirements. The test results and welding quality data are recorded to evaluate the effectiveness of the path planning and provide a basis for the optimization of subsequent welding processes.

[0088] Step S3: Based on the dynamic welding path, real-time welding control is performed on the sign to be welded, and then the movement trajectory of the welding torch is smoothed and optimized to construct an optimal smooth trajectory of the welding torch movement;

[0089] In this embodiment, a welding control system is established according to the dynamic welding path generated in the previous step. This system needs to integrate real-time monitoring data, including the status of the welding point, the position of the welding torch, and environmental changes. Control algorithms (such as PID controllers) are used to adjust welding parameters (such as current, voltage, welding speed, etc.) in real time to ensure the steady-state and dynamic response during the welding process. The temperature and molten pool status of the welding point are monitored in real time to adjust the parameters in a timely manner during the welding process to ensure welding quality. High-precision sensors and cameras are deployed to monitor the welding process in real time. The sensors can include temperature sensors, pressure sensors, and displacement sensors to ensure that all key parameters during the welding process can be obtained. Image data during the welding process is analyzed through algorithms (such as image processing technology) to identify welding defects (such as pores, incomplete penetration, etc.) in real time and feed them back to the welding control system for adjustment. Based on the dynamic welding path, a preliminary movement trajectory of the welding torch is generated according to the welding process parameters. In this stage, the movement path of the welding torch from the starting point to the welding point is determined mainly through path planning algorithms. Combining the welding speed and welding sequence, the initially generated movement trajectory should avoid sharp turns and unnecessary path retracing to improve welding efficiency and quality. A smoothing algorithm (such as Bezier curve, spline curve, or B-spline) is used to optimize the initially generated movement trajectory of the welding torch. This process aims to eliminate sharp changes and discontinuities in the trajectory and make the movement of the welding torch smoother. When performing smoothing optimization, ensure that the movement speed and acceleration of the welding torch are within a reasonable range to avoid affecting welding quality and welding effect. After optimization, the smoothed movement trajectory of the welding torch is verified through simulation software (such as MATLAB Simulink or ROS), simulating the welding process, observing whether the movement of the welding torch is smooth, and evaluating welding quality. Adjust the smoothing parameters according to the simulation results to ensure that the final movement trajectory of the welding torch can be effectively executed in actual welding operations and maintain welding quality. During the actual welding process, a real-time control and feedback mechanism is implemented to ensure dynamic adjustment of the movement trajectory of the welding torch. The welding status is monitored in real time through sensor data, and the movement trajectory of the welding torch is adjusted in a timely manner. If abnormalities occur during the welding process (such as excessive temperature or position deviation), the system will immediately adjust the movement trajectory of the welding torch to ensure that the welding quality is not affected and maintain the continuity and stability of welding.

[0090] Step S4: Obtain the real-time working parameters of the current welding torch; perform adaptive welding torch parameter adjustment on the real-time working parameters of the current welding torch based on the optimal smoothed movement trajectory of the welding torch, so as to obtain the adaptive welding torch adjustment parameters;

[0091] In this embodiment, high-precision sensors (such as current sensors, temperature sensors, and voltage sensors) are deployed around the welding torch to monitor the working state of the welding torch in real time. The sensors should be able to provide accurate and real-time data feedback, including parameters such as welding current, welding voltage, welding speed, and molten pool temperature. The sensor data is transmitted to the control unit in real time through a data acquisition system. The control unit can be based on a PLC (programmable logic controller) or an embedded system and is responsible for data collection, processing, and analysis. The collected real-time working parameters are preprocessed to remove noise and outliers. Filters (such as Kalman filters) can be used to smooth the data to ensure the accuracy and reliability of the data. The real-time parameters are analyzed through data analysis algorithms (such as statistical analysis or machine learning models) to evaluate whether the current working state of the welding torch is within the expected range. This process will help identify possible problems during the welding process, such as excessive current or abnormal temperature. Taking the optimal smooth trajectory of the welding torch movement generated in the previous step as a reference, the relationship between this trajectory and the current working parameters of the welding torch is analyzed. Ensure that the movement trajectory of the welding torch can match the changes in welding parameters. Evaluate the influence of the welding torch movement trajectory under different welding speeds and current conditions to determine the ideal combination of welding parameters. According to the real-time working parameters and the optimal movement trajectory, an adaptive adjustment algorithm is designed. Fuzzy control or adaptive control strategies can be adopted to dynamically adjust the welding current, voltage, and welding speed according to real-time data. For example, if the welding current is too high, the control system can automatically reduce the current and adjust the welding speed at the same time to maintain the stability of the welding molten pool. Such adjustment can be achieved through a PID controller or other control algorithms. During the welding process, a real-time feedback mechanism is implemented to ensure the effectiveness of the adaptive welding torch parameter adjustment. The control system should continuously monitor the working parameters of the welding torch and make adjustments according to real-time feedback to cope with the changes during the welding process. Through a closed-loop control system, ensure that during the welding process, if an abnormality (such as too high molten pool temperature) is detected, the system can quickly adjust the welding parameters to ensure that the welding quality is not affected. After each welding is completed, record the parameters of the adaptive adjustment and the welding quality data for subsequent analysis and improvement. A database can be established to store the parameter records and quality assessment results of each welding. Regularly evaluate and optimize the adaptive adjustment algorithm, and adjust the control strategy according to historical data to improve the stability and quality of the welding process.

[0092] Step S5: Analyze the change trend of the welding molten pool temperature based on the real-time welding monitoring image, and perform dynamic regulation of the welding parameters on the adaptive welding torch adjustment parameters, so as to construct a dynamic welding parameter regulation strategy;

[0093] In this embodiment, a high-resolution thermal imaging camera or an infrared sensor is used to monitor the temperature of the welding molten pool in real time, ensuring that the device can continuously track and record the temperature changes during the welding process, providing high-frequency temperature data. The monitored images and the corresponding temperature data are transmitted to a data processing system for subsequent analysis. This process requires ensuring the real-time and accuracy of the data to avoid welding quality problems caused by delays. The collected real-time temperature data is preprocessed, including noise elimination and data smoothing. Filters (such as moving average filters or Kalman filters) can be used to process the temperature data to improve the accuracy of subsequent analysis. Key features, such as temperature peaks, temperature fluctuation ranges, and change rates, are extracted from the processed data. These features will be used to analyze the thermal dynamic behavior of the welding molten pool. Statistical analysis and data mining algorithms (such as time series analysis or regression analysis) are applied to deeply analyze the temperature change trend of the welding molten pool. Graphical tools (such as Matplotlib or Tableau) can be used to generate temperature change graphs to help visualize the trend and identify the patterns and anomalies of temperature changes. For example, a rapid increase in temperature may indicate improper welding parameter settings. By analyzing these trends, appropriate adjustment strategies are determined. Based on the temperature change trend and its impact on welding quality, a dynamic welding parameter regulation strategy is designed. This strategy should include adjustment schemes for welding current, voltage, and welding speed to respond to temperature changes. Fuzzy control or adaptive control algorithms are used to ensure that the welding parameters can be automatically adjusted according to the real-time temperature feedback. For example, when the temperature is too high, the system can automatically reduce the welding current or adjust the welding speed to prevent overheating. During the welding process, a dynamic regulation mechanism is implemented to ensure that the welding parameters can respond to the changes in the molten pool temperature in real time. The control system should continuously monitor the temperature data and adjust the welding parameters according to the set strategy. A closed-loop control system is designed to ensure that if an abnormal temperature (such as the temperature exceeding the set range) is detected, the system can immediately execute the adjustment operation to ensure the stability of the welding process and the welding quality.

[0094] Step S6: Perform local welding defect analysis based on the real-time welding monitoring images, and optimize the welding defect control of the dynamic welding parameter regulation strategy to construct a dynamic welding control optimization model.

[0095] In this embodiment, a high-resolution imaging device is used to collect images during the welding process in real time. These images should be able to clearly display the welding area and capture details of welding defects, such as pores, cracks, and incomplete penetration. Ensure that the frequency of image collection is high enough to capture the rapidly changing situations during the welding process, guaranteeing the comprehensiveness and real-time nature of the data. Image processing and computer vision techniques are used to analyze the collected welding images. Edge detection algorithms (such as Canny edge detection) and morphological processing methods (such as opening and closing operations) are used to extract welding defect features. Machine learning algorithms (such as convolutional neural network CNN) are applied for defect recognition and classification. By training the model, it can automatically identify different types of welding defects and output the location and type of the defects. The identified welding defect data is recorded in a database, including information such as defect type, location, and severity. This data will be used for subsequent analysis and model optimization. Statistical analysis is performed on the defect data to identify common defect types and their occurrence frequencies and conditions, helping to determine the key factors that need to be concerned during the welding process. Based on the analysis results of local welding defects, the existing dynamic welding parameter regulation strategies are evaluated to identify which parameters have a significant impact on the defects, such as welding current, welding speed, and welding gas flow rate. Optimization algorithms (such as genetic algorithms or particle swarm optimization) are used to adjust the welding parameters to reduce the occurrence of defects. For example, if it is found that high current leads to an increase in pores, the current setting is adjusted, and a new regulation strategy is formed by combining other welding parameters. Based on the optimized welding parameters and defect analysis results, a dynamic welding control optimization model is constructed. This model should be able to respond in real time to changes during the welding process and intelligently adjust the welding parameters to control defects. Select a suitable modeling method, such as fuzzy logic control, neural network, or rule-based system, to ensure that the model can effectively handle the complex welding process and achieve dynamic regulation. The constructed dynamic welding control optimization model is verified in an experimental environment to evaluate its effectiveness in the actual welding process. By comparing the welding quality and defect occurrence rate, the effectiveness of the model is tested. After confirming the effectiveness of the model, it is applied to actual welding production, and the welding quality is continuously tracked through a real-time monitoring system. According to the actual welding feedback, the model is continuously adjusted and optimized to ensure its adaptability and stability.

[0096] In this embodiment, refer to Figure 2 , which is a schematic diagram of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0097] Step S11: Obtain high-definition monitoring images of the sign to be welded and the welding work log.

[0098] Step S12: Perform visual recognition of the sign texture on the high-definition monitoring image of the sign to be welded to generate sign texture visual features.

[0099] Step S13: Sharpen and enhance the visual features of the sign texture to obtain a sharpened and enhanced sign image;

[0100] Step S14: Optimize the contrast of multiple regions of the sharpened and enhanced sign image to obtain a contrast-optimized sign image;

[0101] Step S15: Simulate the welding visual effect of the contrast-optimized sign image based on the welding work log to obtain the welding visual effect of the sign.

[0102] In this embodiment, a high-resolution camera (such as a DSLR or an industrial camera) is used to capture the signboard to be welded under good lighting conditions, ensuring that the image is clear and free of blur, suitable for subsequent visual recognition processing. During the shooting process, pay attention to selecting the appropriate angle and distance to capture the details and textures of the signboard to the greatest extent. Collect the work logs related to the welding process, including welding parameters (such as current, voltage, welding speed, etc.), welding material information, welding time, and remarks of the operator, etc., ensuring that the format of the work logs is clear and structured for subsequent data processing and analysis. Store the high-definition monitoring images and welding work logs in a dedicated database or file system to ensure data security and traceability. At the same time, record the acquisition time of the data and related metadata. Perform preprocessing on the obtained high-definition monitoring images, including denoising, grayscale conversion, and normalization. A Gaussian filter can be used to remove image noise to ensure the accuracy of subsequent texture recognition. Adopt texture recognition algorithms (such as LBP (Local Binary Pattern), GLCM (Gray-Level Co-Occurrence Matrix), or SIFT (Scale-Invariant Feature Transform)) to extract the texture features of the image. These algorithms can effectively capture the subtle texture changes on the surface of the signboard. The generated texture features can be used to describe the surface state of the signboard and any potential defects. Store the extracted texture features in the database for subsequent processing and analysis, ensuring that the feature data is associated with the original image and work log. Select a suitable sharpening algorithm, such as the Laplacian operator or UnsharpMask. These algorithms can enhance the edges and details of the image. Apply the selected sharpening algorithm to the signboard image to be processed. By adjusting the sharpening intensity, ensure that the enhancement effect is moderate to avoid oversharpening of the image. After processing, carefully check the sharpening effect to ensure that the details of the signboard are clearer and the texture features are obvious. Save the sharpened image as a new file, ensuring that its format is suitable for subsequent processing. PNG or JPEG format can be used depending on specific requirements. Select a suitable contrast optimization algorithm, such as histogram equalization or adaptive histogram equalization (CLAHE). The latter can enhance the local contrast of the image while preserving details. Divide the sharpened and enhanced signboard image into multiple regions. Each region can be divided according to the complexity and features of the image content for local contrast optimization. Apply the selected contrast optimization algorithm to each region to ensure that the natural features of the image are preserved during the processing and avoid producing unnatural contrast effects. Finally, integrate the processed regions to generate a complete contrast-optimized signboard image. Save the optimized image in a new file format to ensure its good visual performance and suitability for subsequent welding effect simulation. Extract welding parameters from the welding work log, such as welding temperature, welding time, welding materials, etc. These parameters will be used to simulate the visual effects during the welding process. Select a suitable welding effect simulation algorithm, and physical-based simulation methods or image processing algorithms can be considered.For the image-based welding effect simulation technology, the extracted welding parameters are combined with the contrast-optimized sign image to conduct visual simulation of the welding effect. This can be achieved by adjusting the brightness, color, texture, etc. of the image, simulating the possible color changes and surface effects after welding, paying attention to maintaining the authenticity of the simulation results to ensure they are consistent with the actual welding effect. Save the visual effect of the sign welding obtained from the simulation as a new image file, and a high-quality image format can be used for subsequent display and analysis. Evaluate the simulation results to ensure that the visual performance of the welding effect can truly reflect the expected effect in the actual welding process. If necessary, parameter adjustment and re-simulation can be carried out according to the evaluation results.

[0103] In this embodiment, the specific steps of step S15 are as follows:

[0104] Conduct a welding process requirement analysis on the welding work log to generate the welding process requirement characteristics of the sign to be welded;

[0105] Identify the welding process flow of the welding work log and extract the multi-stage welding process flow;

[0106] Based on the welding work log, locate the key identification nodes of the contrast-optimized sign image to obtain multiple key identification nodes;

[0107] Conduct an analysis of the spatial distribution of the identification points for multiple key identification nodes to generate the spatial distribution data of the identification nodes;

[0108] Based on the welding process requirement characteristics of the sign to be welded, calculate the welding process parameters for each identification node for the spatial distribution data of the identification nodes, thereby generating the welding process parameters for each identification node;

[0109] Based on the multi-stage welding process flow, conduct a visual simulation of the welding effect for the welding process parameters of each identification node, thereby obtaining the visual effect of the sign welding.

[0110] In this embodiment, collect and organize welding work logs, including welding parameters, material information, environmental conditions, etc., to ensure the integrity and accuracy of the data for in-depth analysis. Import the log information into data analysis software (such as Excel, Pandas library in Python) for structured processing. By analyzing the parameters in the welding work logs, identify the key welding process requirement characteristics, which may include welding material type, welding temperature range, welding speed, welding current, and voltage, etc. Use statistical analysis methods to calculate the mean, variance, and other statistical characteristics of each parameter, and identify the factors that have the greatest impact on welding quality. Organize the extracted welding process requirement characteristics into a document, recording the definition of each characteristic and its corresponding reference value, which will provide a detailed basis for subsequent welding process design and optimization. By analyzing the welding work logs, identify the various stages of the welding process, which may include the preparation stage, welding stage, cooling stage, etc. Combine the timestamps and parameter changes recorded in the welding work logs to draw the timeline of the welding process. Display the identified welding process stages in the form of a flowchart. Use flowchart software (such as Visio or Lucidchart) to create a visual process flowchart, indicating the main operations, parameter settings, and time requirements for each stage in the flowchart to ensure the clarity and understandability of the process. Organize the welding process flowchart and the corresponding detailed description document into a report for subsequent process implementation and review. When necessary, communicate with on-site welding technicians to verify whether the extracted process conforms to actual operations. Preprocess the contrast-optimized sign image, including denoising, enhancing contrast, and grayscale conversion, to improve the identifiability of key nodes. Use image processing software (such as OpenCV or MATLAB) to import the processed image. Adopt edge detection algorithms (such as Canny edge detection) or feature detection algorithms (such as SIFT or SURF) to identify the key nodes in the image. By setting appropriate thresholds and parameters, ensure that the detected nodes have good repeatability and accuracy. Record the coordinate information of the detected key nodes in the database to ensure that this information can be combined with the welding process requirement characteristics for subsequent analysis. Statistically analyze the coordinate information of each key node, calculate the spatial distribution characteristics of the nodes, including aggregation degree, distribution range, etc. Use statistical analysis software (such as R or Python) for data analysis, draw scatter plots or heat maps to display the spatial distribution of key nodes, and visualize the spatial distribution data to intuitively understand the distribution characteristics of the nodes. This can help identify which areas have dense nodes and which areas are relatively sparse. Combine the welding process requirement characteristics to analyze the potential impact of node distribution on the welding process. Organize the generated spatial distribution data of identified nodes into a document or database record for subsequent use in calculating welding process parameters. Based on the requirement characteristics, select appropriate welding parameter calculation methods, which may include empirical formulas, machine learning models, or physics-based simulation methods. For each key node,Combined with its spatial distribution characteristics and welding process requirements, calculate the welding process parameters, which may include welding current, voltage, welding speed, etc. Record and organize the calculation results into a table to ensure that the parameters of each node are clearly retrievable. When necessary, verify and adjust the parameters. The rationality and effectiveness of the calculation can be verified by simulating the welding process or conducting small-scale trial welding. Select a suitable welding effect simulation software or tool (such as ANSYS, SolidWorks, or a custom image processing script) to ensure that it can support the simulation of multi-stage welding processes. Input the welding process parameters of each identified node into the simulation tool, ensuring that the parameter settings conform to the actual welding conditions to avoid deviations in the simulation results. Start the simulation tool and run the visual effect simulation of the welding process. During this process, observe the images generated by the simulation and analyze the details and quality of the welding effect. If the generated welding effect does not match the expectations, adjust the input parameters as needed and conduct the simulation again. Integrate the welding visual effects of all key nodes into a report, record the welding effect and parameter settings of each node, evaluate the simulation results to ensure that they can reflect the expected effects in the actual welding process, and provide a basis for subsequent welding process optimization.

[0111] In this embodiment, refer to Figure 3 , which is a schematic diagram of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0112] Step S21: Accurately identify welding points based on the welding visual effect of the signboard to obtain multiple welding points;

[0113] Step S22: Optimize the signboard image according to the contrast to visually locate and mark the multiple welding points to obtain the positioning coordinates of each welding point in the image;

[0114] Step S23: Analyze the three-dimensional structure of the signboard based on the contrast-optimized signboard image to generate the three-dimensional geometric information of the signboard;

[0115] Step S24: Dynamically plan the welding path for the positioning coordinates of each welding point in the image based on the three-dimensional geometric information of the signboard, thereby constructing a dynamic welding path.

[0116] In this embodiment, a welding effect image is acquired, and the image quality is ensured to be good so as to clearly display the characteristics of the welding points. If necessary, image preprocessing is performed, including denoising, enhancing contrast and sharpness, to improve the recognizability of the welding points. A suitable welding point recognition algorithm is selected, such as a threshold segmentation algorithm based on image processing, edge detection (such as Canny edge detection) or a machine learning method (such as a convolutional neural network CNN), for the automatic recognition of welding points. For welding points in a complex background, using a deep learning method can improve the recognition accuracy. The selected recognition algorithm is applied to process the welding effect image to extract the coordinate information of the welding points. For each welding point, connected component analysis can be used to confirm its position. The recognized welding points are recorded in the database, ensuring that the coordinate information of each point is detailed and accurate. The recognized welding points are manually verified to ensure the accuracy of the recognition results. If the recognition effect is not ideal, it can be optimized by adjusting the algorithm parameters or changing the algorithm. The sign image with optimized contrast is imported, ensuring that the image is clear and has good contrast for accurate visual positioning. An image processing tool (such as OpenCV) is used to draw the positioning marks of each welding point on the image with optimized contrast. Different colored dots or cross-shaped marks can be used to represent different welding points. For each welding point, its specific coordinates in the image are recorded, and it is ensured that these coordinates can correspond to the recognition results of the welding points. The positioning coordinates of all welding points and their corresponding image marking information are stored in the database for subsequent processing and query. The image coordinates of each welding point and its corresponding relationship in the three-dimensional space are recorded to prepare for subsequent three-dimensional analysis and path planning. The marking results are checked to ensure that the marks of each welding point are accurate. If errors are found, the marking positions need to be adjusted in time and the coordinates are re-recorded. Stereo vision technology is adopted, and multiple contrast-optimized sign images at different angles are used for three-dimensional reconstruction. If there is only one image, a depth estimation algorithm (such as monocular depth estimation) can be used to obtain depth information. Camera calibration technology is used to ensure clear geometric relationships in the captured images, thereby improving the accuracy of three-dimensional reconstruction. By processing multi-view images, a three-dimensional reconstruction algorithm (such as structured light, laser scanning or bundle adjustment) is used to generate the three-dimensional geometric model of the sign. Software such as MeshLab or Blender can be used for the generation and processing of the three-dimensional model. The generated three-dimensional model should include the surface details, shape of the sign and the spatial positions of the welding points. Geometric information is extracted from the generated three-dimensional model, including the three-dimensional coordinates, surface normals, curvatures, etc. of the welding points. This information will be helpful for subsequent welding path planning. The extracted geometric information is stored in a structured database for subsequent query and calculation. A suitable path planning algorithm is selected, such as the A* algorithm, Dijkstra algorithm or RRT (rapidly-exploring random tree) algorithm. These algorithms can find the optimal path in a complex three-dimensional space. According to the distribution of the welding points and the geometric shape of the sign,Determine important parameters for path planning, such as the starting point, ending point, and traversable area. Based on the three-dimensional coordinate information of each welding point and the positioning coordinates in the image, dynamically generate a welding path to ensure that the path follows the principle of the shortest distance as much as possible, reducing the movement time during the welding process. Consider the possible obstacles and environmental factors during the welding process to ensure that the generated welding path is safe and efficient. Optimize the generated welding path, considering the welding sequence and welding process requirements, to ensure that the path planning meets the requirements of actual welding operations. When necessary, conduct path simulation to verify the effectiveness and feasibility of the welding path. Record and save the finally generated welding path for subsequent welding control and execution. The path information can be exported in a specific format (such as G-code or other control instructions) for direct use by the welding equipment.

[0117] In this embodiment, step S3 includes the following steps:

[0118] Step S31: Perform real-time welding control on the sign to be welded based on the dynamic welding path, and collect real-time welding monitoring images;

[0119] Step S32: Identify the actual welding contact points on the real-time welding monitoring images and mark the actual welding contact points;

[0120] Step S33: Analyze the surface change of the welding points for the actual welding contact points to generate the high-temperature welding surface change characteristics of the welding points;

[0121] Step S34: Calculate the sign welding curvature change for the high-temperature welding surface change characteristics of the welding points to generate the sign welding curvature change data;

[0122] Step S35: Perform welding path mutation analysis on the dynamic welding path based on the sign welding curvature change data and mark the welding path mutation points;

[0123] Step S36: Smoothly optimize the torch movement trajectory based on the welding path mutation points to construct the optimal smooth torch movement trajectory.

[0124] In this embodiment, the pre-generated dynamic welding path data is used to set the motion trajectory of the welding equipment, which includes the starting position, traveling path, and ending position of the welding torch, ensuring that the welding equipment can automatically adjust its motion according to the path data. Integrate the welding control system and the motion control system to ensure that the welding torch can move accurately along the preset path. A PLC (Programmable Logic Controller) or a motion control card can be used to achieve real-time control of the welding equipment. Configure a high-definition camera or a thermal imaging camera to capture real-time welding monitoring images during the welding process. These images can provide visual information of the welding process to help monitor the welding quality. Ensure that the image acquisition system can capture images at a high frequency during the welding process to obtain the status of the welding contact point in a timely manner. Store the captured real-time welding monitoring images in a database or cloud storage for subsequent analysis and processing. Use a suitable image format (such as PNG or JPEG) to ensure image quality. Preprocess the real-time welding monitoring images, including denoising, enhancing contrast, and grayscale conversion, to ensure the visibility of the welding contact point. These operations can be performed using image processing libraries such as OpenCV. Apply edge detection algorithms (such as Canny edge detection) or machine learning-based image segmentation methods to automatically identify the welding contact point. If conditions permit, use deep learning models (such as YOLO or Mask R-CNN) for more accurate contact point detection. Mark the identified welding contact points and distinguish them using color markings or symbols. Store the marking information and the coordinates of the contact points for subsequent data analysis. Verify the identified and marked welding contact points to ensure the accuracy of the markings. If errors are found, adjust the parameters of the recognition algorithm or make manual corrections. Based on the contact point position in the welding monitoring image, extract the image data around the welding point to construct an initial model of the high-temperature welding surface. Use three-dimensional reconstruction technology and surface fitting algorithms (such as B-spline surface fitting) to analyze the welding contact points, extract the surface change characteristics of the welding points at high temperatures, calculate the curvature change, radius change, and other geometric characteristics of the welding points. These characteristics can reflect the changes in the heat-affected zone during the welding process and help analyze the welding quality. Record and store the surface change characteristics of the welding points for subsequent analysis and comparison. At the same time, visualize the data to generate a surface change diagram to facilitate understanding of the changes occurring during the welding process. Select a suitable curvature calculation method. Common methods include Gaussian curvature and mean curvature calculations. Numerical methods (such as the finite difference method) can be used to calculate the curvature change of the surface. Based on the surface change characteristics of the welding contact points, perform curvature calculations, analyze the curvature change of the welding points at high temperatures, and record the change data. Calculate the local curvature of the welding points to evaluate the impact of the heat-affected zone during the welding process on the welding quality. Organize the welding curvature change data into structured information for subsequent analysis. A curvature change diagram can be generated to visualize the change trend of the curvature during the welding process. Verify the curvature change data to ensure its consistency with the actual welding situation.If necessary, adjust the calculation method or parameters, and select a suitable analysis method, such as a threshold-based mutation detection algorithm or a wavelet transform method in signal processing, to identify mutation points in the welding path. Apply the selected analysis method to perform mutation analysis on the dynamic welding path. By comparing adjacent curvature change values, identify the mutation points and record their positions in the welding path. Mark the identified mutation points on the welding path diagram, distinguish them using different colors or symbols, and record the coordinate information of the mutation points. Record the information of the welding path mutation points in the database and perform statistical analysis to understand potential problems existing in the welding process. Select a suitable trajectory smoothing algorithm, such as Bezier curve fitting, spline curve interpolation, or Kalman filter. These algorithms can effectively smooth the torch movement trajectory and reduce the impact brought by mutation points. Based on the identified welding path mutation points, use the selected smoothing algorithm to process the torch movement trajectory to ensure that the trajectory is as smooth as possible during welding and avoid the impact caused by mutations. Visualize the smoothed torch movement trajectory to generate a movement trajectory diagram for easy observation of the trajectory changes. Compare the trajectories before and after smoothing to verify the effectiveness of the smoothing effect. Save the optimized torch movement trajectory to the database and generate corresponding control instructions for direct use by the welding equipment.

[0125] In this embodiment, step S4 includes the following steps:

[0126] Step S41: Obtain the real-time working parameters of the current torch;

[0127] Step S42: Perform multi-stage trajectory segmentation on the optimal torch movement smoothing trajectory to obtain multi-stage welding trajectories;

[0128] Step S43: Calculate the welding process parameters for each segment of the multi-stage welding trajectory according to the welding visual effect of the identification plate to obtain the welding process parameters for each segment of the trajectory;

[0129] Step S44: Calculate the welding current of the current torch based on the real-time working parameters of the current torch to obtain the real-time welding current of the current torch;

[0130] Step S45: Identify the torch welding speed of the current torch based on the real-time working parameters of the current torch and extract the torch welding speed of the current torch;

[0131] Step S46: Perform adaptive torch parameter adjustment on the real-time welding current of the current torch and the torch welding speed of the current torch based on the welding process parameters of each segment of the trajectory to obtain adaptive torch adjustment parameters.

[0132] In this embodiment, appropriate sensors are installed on the welding torch and the welding equipment to obtain key parameters during the welding process in real time. These sensors may include current sensors, voltage sensors, and speed sensors to ensure that they can accurately reflect the working state of the welding torch. A data acquisition system is configured to transmit the sensor data to the central control system in real time. An industrial controller (such as a PLC) or an embedded system (such as Arduino or Raspberry Pi) can be used for data acquisition and processing. A user interface is developed to display the working parameters of the welding torch in a graphical manner in real time. The user interface can display current, voltage, welding speed, and other important information for the operator to monitor. The working parameters of the welding torch obtained in real time are recorded in a database for subsequent analysis and evaluation, ensuring that the data storage format is suitable for subsequent processing and analysis. The smooth motion trajectory data of the welding torch is obtained and analyzed, and data processing tools (such as MATLAB or Python) are used for trajectory visualization and analysis to ensure that the trajectory data is available and easy to process. A suitable trajectory segmentation algorithm is selected, such as a segmentation method based on speed change or a segmentation method based on distance. By setting thresholds, natural segmentation points in the trajectory are identified, and the smooth trajectory is segmented according to the selected algorithm and divided into multiple stages. Each stage should include the welding process parameters required by the welding torch during this section of movement. The starting and ending points of each stage are recorded in the database and combined with subsequent welding process parameter calculations to ensure that the parameters of each stage can be clearly defined. With the help of the previously generated identification sign welding visual effect, the influence of each trajectory stage on the welding process parameters is analyzed, and image processing tools are used to ensure that the connection between the welding effect and the process parameters is clear. Based on the requirements of the welding process, a parameter calculation model is established. The model should include calculation formulas for parameters such as welding current, voltage, welding speed, and welding time to ensure that it can adapt to different welding stages. For each welding trajectory stage, the established model is used to calculate various welding process parameters, and the parameters such as current, speed, and welding time of each stage are recorded and stored in the database. Verify whether the calculation results meet the actual welding requirements. If necessary, adjust the calculation model or parameters to ensure that the welding process parameters of each stage can achieve the best welding effect. Apply the welding current calculation formula, which usually includes factors such as material type, welding method, and welding rate, to ensure that the formula is suitable for the current welding situation. The working parameters of the welding torch obtained in real time (such as voltage and welding speed) are input into the current calculation model, and real-time calculations are performed using the set formula to output the calculation results of the real-time welding current, ensuring that it can reflect the working state of the welding torch in a timely manner. Record the results and store them in the database for subsequent analysis and reference. Monitor the change of the real-time welding current to ensure that it is within a reasonable range. If abnormal current is found, adjust the welding parameters in time to avoid welding quality problems. Configure a welding speed sensor to monitor the moving speed of the welding torch in real time. These sensors can be photoelectric sensors or encoders.Ensure that it can accurately reflect the movement state of the welding torch. Calculate the welding speed of the welding torch through the real-time obtained welding torch movement data. Usually, the speed can be calculated by the ratio of the distance the welding torch moves to the time. Record and store the real-time welding speed in the database, ensure that the speed data is combined with the welding process parameters for subsequent analysis, monitor the real-time change of the welding speed, ensure that the speed remains within the set range during the welding process. If the speed is abnormal, adjust it in time to optimize the welding process. Design an adaptive control algorithm that can automatically adjust the working parameters of the welding torch according to the changes of the real-time welding current and welding speed. Common methods include the PID control algorithm or the fuzzy control algorithm. According to the comparison of the process parameters of each welding section with the real-time current and welding speed, adjust the current and welding speed of the welding torch in real time to ensure that the welding torch can adapt to different process requirements during the welding process. Monitor the adjusted welding current and welding speed to ensure that they meet the set process parameters. The feedback mechanism can detect and adjust the non-conforming parameters in time to ensure the stability of the welding process. Record and store the parameters of the welding torch after adaptive adjustment for subsequent analysis and optimization. Based on historical data, optimize the adjustment algorithm to improve the automation and accuracy of future welding processes.

[0133] In this embodiment, the specific steps of step S5 are as follows:

[0134] Step S51: Analyze the heat distribution in the welding area of the real-time welding monitoring image to generate the heat distribution characteristics of the welding area;

[0135] Step S52: Evaluate the state of the welding molten pool based on the heat distribution characteristics of the welding area to obtain the state characteristics of the welding molten pool;

[0136] Step S53: Analyze the temperature change trend of the state characteristics of the welding molten pool to generate the temperature change trend of the welding molten pool;

[0137] Step S54: Predict and evolve the abnormal temperature according to the temperature change trend of the welding molten pool to generate the predicted data of the abnormal temperature of the welding molten pool;

[0138] Step S55: Dynamically regulate the welding parameters of the adaptive welding torch adjustment parameters based on the predicted data of the abnormal temperature of the welding molten pool to construct a dynamic welding parameter regulation strategy.

[0139] In this embodiment, a thermal imaging camera or a high-temperature sensor is used to collect monitoring images of the welding area in real time, ensuring that the images are clear and can effectively capture the temperature changes generated during the welding process. The collected images are preprocessed, including noise removal and contrast enhancement, to improve the visualization effect of the heat distribution. These operations can be performed using image processing software (such as OpenCV). Apply the heat distribution analysis algorithm to calculate the heat distribution of the preprocessed image. This can be achieved by converting pixel values to map the colors in the image to corresponding temperature values, generating a heat distribution map. An interpolation algorithm (such as bilinear interpolation) is used to improve the accuracy of the heat distribution, ensuring that temperature changes in different regions can be accurately captured. Extract the heat distribution characteristics of the welding area from the heat distribution map, including information such as heat concentration areas and temperature gradients. These characteristics will be used for subsequent molten pool state evaluation. Organize and store the extracted feature data for subsequent analysis and comparison. Display the heat distribution characteristics in a visual manner to generate a heat distribution map, showing different temperature regions through color coding for easy identification of heat concentration areas and affected regions. Define the state characteristics of the welding molten pool, including the area, depth, temperature distribution, etc. of the molten pool. These characteristics have an important impact on welding quality. Based on the heat distribution characteristics, establish a molten pool state evaluation model. Machine learning algorithms such as support vector machines (SVM) or decision trees can be used to evaluate the state of the welding molten pool. Input the heat distribution characteristics of the welding area into the molten pool state evaluation model for real-time evaluation. The model will output the state characteristics of the molten pool, including whether it is within the normal range. Record the evaluation results in the database to ensure that the state characteristics of the molten pool during each welding process are detailedly recorded, and analyze the evaluation results for subsequent optimization of welding parameters. Collect data on the state characteristics of the molten pool from real-time monitoring, including information such as temperature and area at different time points, ensuring the integrity and accuracy of the data. Establish a temperature change trend analysis model. Time series analysis methods (such as ARIMA models) or machine learning methods (such as linear regression) can be used to predict the temperature change trend of the welding molten pool. Input the collected data on the state characteristics of the molten pool into the model for temperature change trend analysis. The model will output a trend graph of the molten pool temperature over time, showing the increase, decrease, or steady state of the temperature. Visualize the temperature change trend in the form of a chart for easy understanding of the temperature change pattern by the operator. Record the analysis results for subsequent reference and optimization of welding parameters. Determine the normal temperature range of the welding molten pool and define the criteria for abnormal temperatures. This can be based on historical data, industry standards, or laboratory test results. Use machine learning methods (such as random forests or neural networks) or statistical methods (such as Z-score analysis) to establish an abnormal temperature prediction model. Train using the collected temperature change data. Input the real-time temperature change trend data of the welding molten pool into the abnormal prediction model for real-time prediction. The model will output the abnormal temperature prediction data of the welding molten pool. Record the prediction results in the database.Combined with a real-time monitoring system to ensure that operators can promptly learn about potential abnormal situations, design a regulation strategy based on the predicted data of the abnormal temperature of the welding molten pool, which may include real-time adjustment of parameters such as welding current, welding speed, and welding interval in response to temperature changes, develop an adaptive control system that can automatically adjust the working parameters of the welding torch according to the real-time state and predicted data of the welding molten pool, and the PID control algorithm or fuzzy control algorithm can be used to achieve this. Input the predicted data of the abnormal temperature into the adaptive control system to adjust the parameters of the welding torch in real time, ensuring that when the temperature is abnormal, the welding torch can quickly adjust the current and speed to reduce the temperature and maintain the welding quality. Continuously monitor the adjusted parameters during the welding process to ensure that they meet the requirements of the welding process. If problems are found, the system should be able to automatically make adjustments and record all adjustment data for subsequent analysis.

[0140] In this embodiment, the specific steps of step S6 are as follows:

[0141] Step S61: Identify the welding results of multiple segments based on the real-time welding monitoring image and extract the welding results of multiple process stages;

[0142] Step S62: Evaluate the welding quality of each welding point for the welding results of multiple process stages to generate the welding quality evaluation value of each welding point;

[0143] Step S63: Conduct local welding defect analysis based on the welding quality evaluation value of each welding point to generate local welding defect data;

[0144] Step S64: Optimize the welding defect control of the dynamic welding parameter regulation strategy based on the local welding defect data to construct a dynamic welding control optimization model.

[0145] In this embodiment, a high-resolution welding monitoring device (such as a thermal imaging camera or a high-frame-rate video camera) is used for real-time image acquisition to ensure that the quality of the monitoring images is sufficient to capture the details during the welding process. The acquired images are preprocessed, including denoising, enhancing contrast, and image normalization, to improve the accuracy of subsequent analysis. Image processing and computer vision techniques are adopted, and edge detection and contour extraction algorithms (such as Canny edge detection and Hough transform) are used to identify the welding area in the image. By setting specific thresholds and conditions, different process stages (such as preheating, welding, and cooling) are segmented to ensure that the results of each stage can be effectively extracted. The welding results of each identified process stage are classified and labeled to form a structured data set, which includes the specific characteristics of each stage (such as weld width, depth, etc.). The extracted results are stored in a database for subsequent analysis and evaluation. The welding results of multiple process stages are visually displayed to generate dynamic charts or image sequences of the welding process to help technicians quickly understand the changes during the welding process. Welding quality assessment criteria are established, including the geometric features (such as depth, width, shape) and physical properties (such as strength, toughness) of the welds, and the qualified range of each feature is determined so that subsequent evaluation work can be based on evidence. Key features of each welding point are obtained from the extracted welding results, which can be achieved by using image processing techniques to extract information such as the geometric parameters and temperature distribution of the weld. Feature extraction algorithms (such as morphological processing) are applied to ensure that the extracted features have high accuracy and consistency. Machine learning models (such as support vector machines or decision trees) are used to evaluate the quality of the extracted welding point features. The welding point features are input into the model, and a welding quality assessment value is output. Combining historical data and expert experience, the evaluation model is continuously optimized to improve the prediction accuracy. The evaluation value of each welding point is recorded in the database and associated with other data of the welding process for comprehensive analysis. Statistical analysis is carried out to identify welding points with unqualified quality and further defect analysis is performed to determine the possible defect types during the welding process, such as porosity, inclusions, lack of penetration, welding cracks, etc. A defect classification standard is established for subsequent analysis. Feature analysis and pattern recognition techniques are used to perform defect analysis on the welding quality assessment value. Statistical methods (such as control charts) or machine learning methods (such as clustering analysis) can be used to identify defects. Local defects are identified based on the evaluation value and defect analysis results, and the welding points with problems are marked. Combining image processing techniques, a defect distribution map is generated to clarify the defect location and type. The detailed data of local welding defects (including defect type, location, severity, etc.) are recorded and stored in the database to ensure that defects can be traced and analyzed subsequently. Based on the local welding defect data, targeted welding parameter optimization strategies are developed, including the adjustment direction and amplitude of parameters such as welding current, welding speed, and welding gas flow rate. A dynamic welding control optimization model is constructed, and a feedback control mechanism is adopted to adjust the welding parameters in real time.In response to defect information during the welding process, adaptive control or fuzzy control algorithms can be used to apply the optimization model in the actual welding process, monitor the welding quality and defect conditions in real time, continuously adjust the model parameters according to the feedback results to ensure its adaptability and accuracy, continuously monitor the parameter changes during the welding process, record the welding quality data after the implementation of the optimization strategy to ensure the effectiveness of the optimization measures, and regularly evaluate and update the model to improve the welding quality control level.

[0146] In this embodiment, a welding control device for a sign is provided, which is used to execute the welding control method for a sign as described above, including:

[0147] A welding vision simulation module, which is used to obtain a high-definition monitoring image of the sign to be welded and a welding work log; perform sign texture visual recognition on the high-definition monitoring image of the sign to be welded and simulate the welding visual effect, so as to obtain the welding visual effect of the sign;

[0148] A welding path planning module, which is used to accurately identify welding points based on the welding visual effect of the sign and perform dynamic welding path planning, so as to construct a dynamic welding path;

[0149] A welding torch motion trajectory module, which is used to perform real-time welding control on the sign to be welded based on the dynamic welding path, and then perform smoothing optimization on the welding torch motion trajectory, so as to construct an optimal welding torch motion smooth trajectory;

[0150] An adaptive parameter adjustment module, which is used to obtain the current real-time working parameters of the welding torch; perform adaptive welding torch parameter adjustment on the current real-time working parameters of the welding torch based on the optimal welding torch motion smooth trajectory, so as to obtain an adaptive welding torch adjustment parameter;

[0151] A dynamic welding regulation module, which is used to analyze the change trend of the welding pool temperature according to the real-time welding monitoring image and perform dynamic welding parameter regulation on the adaptive welding torch adjustment parameter, so as to construct a dynamic welding parameter regulation strategy;

[0152] A welding control optimization module, which is used to perform local welding defect analysis based on the real-time welding monitoring image and perform welding defect control optimization on the dynamic welding parameter regulation strategy to construct a dynamic welding control optimization model.

[0153] Therefore, no matter from which point of view, the embodiment should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to include all changes falling within the meaning and scope of the equivalent elements of the application documents within the present invention.

[0154] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A welding control method for identification signs, characterized in that, Including the following steps: Step S1: Obtain the high-definition monitoring image of the sign to be welded and the welding work log; perform visual recognition of the sign texture on the high-definition monitoring image of the sign to be welded, and simulate the welding visual effect, so as to obtain the welding visual effect of the sign; Step S2: Based on the welding visual effect of the sign, accurately identify the welding points and plan the dynamic welding path, so as to construct the dynamic welding path; Step S3: Based on the dynamic welding path, perform real-time welding control on the sign to be welded, and then optimize the smoothness of the welding torch movement trajectory, so as to construct the optimal smooth trajectory of the welding torch movement; Step S4: Obtain the real-time working parameters of the current welding torch; based on the optimal smooth trajectory of the welding torch movement, perform adaptive welding torch parameter adjustment on the real-time working parameters of the current welding torch, so as to obtain the adaptive welding torch adjustment parameters; Step S5: Analyze the change trend of the welding pool temperature according to the real-time welding monitoring image, and perform dynamic regulation of the welding parameters on the adaptive welding torch adjustment parameters, so as to construct the dynamic welding parameter regulation strategy; Step S6: Analyze the local welding defects based on the real-time welding monitoring image, and optimize the welding defect control on the dynamic welding parameter regulation strategy to construct the dynamic welding control optimization model; Among them, the specific steps of step S2 are: Step S21: Based on the welding visual effect of the sign, accurately identify the welding points to obtain multiple welding points; Step S22: According to the contrast-optimized sign image, perform image visual positioning and marking on multiple welding points to obtain the positioning coordinates of each welding point in the image; Step S23: Analyze the three-dimensional structure of the sign according to the contrast-optimized sign image to generate the three-dimensional geometric information of the sign; Step S24: Based on the three-dimensional geometric information of the sign, perform dynamic welding path planning on the positioning coordinates of each welding point in the image, so as to construct the dynamic welding path.

2. The welding control method for a signboard according to claim 1, wherein The specific steps of step S1 are: Step S11: Obtain the high-definition monitoring image of the sign to be welded and the welding work log; Step S12: Perform visual recognition of the sign texture on the high-definition monitoring image of the sign to be welded to generate the sign texture visual features; Step S13: Sharpen and enhance the details of the sign texture visual features to obtain the sharpened and enhanced sign image; Step S14: Optimize the contrast of the sharpened and enhanced sign image in multiple regions to obtain the contrast-optimized sign image; Step S15: Based on the welding work log, simulate the welding visual effect of the contrast-optimized sign image to obtain the welding visual effect of the sign.

3. The welding control method for identification signs according to claim 2, characterized in that, The specific steps of step S15 are: Analyze the welding process requirements of the welding work log to generate the welding process requirement features of the sign to be welded; Identify the welding process flow of the welding work log and extract the multi-stage welding process flow; Based on the welding work log, perform key sign node positioning on the contrast-optimized sign image to obtain multiple key sign nodes; Analyze the spatial distribution of the identification points of multiple key sign nodes to generate the spatial distribution data of the identification nodes; Based on the welding process requirement characteristics of the sign to be welded, calculate the welding process parameters for each identification node of the spatial distribution data of the identification nodes, so as to generate the welding process parameters of each identification node; Based on the multi-stage welding process flow, simulate the welding visual effect of each identification node's welding process parameters, so as to obtain the welding visual effect of the sign.

4. The welding control method for identification signs according to claim 1, wherein The specific steps of step S3 are as follows: Step S31: Based on the dynamic welding path, perform real-time welding control on the sign to be welded, and collect real-time welding monitoring images; Step S32: Identify the actual welding contact points in the real-time welding monitoring images, and mark the actual welding contact points; Step S33: Analyze the surface change of the welding points for the actual welding contact points, so as to generate the high-temperature welding surface change characteristics of the welding points; Step S34: Calculate the welding curvature change of the sign for the high-temperature welding surface change characteristics of the welding points, so as to generate the welding curvature change data of the sign; Step S35: Based on the welding curvature change data of the sign, perform welding path mutation analysis on the dynamic welding path, and mark the welding path mutation points; Step S36: Based on the welding path mutation points, perform smooth optimization on the movement trajectory of the welding torch, so as to construct the optimal smooth movement trajectory of the welding torch.

5. The welding control method for identification signs according to claim 1, characterized in that The specific steps of step S4 are as follows: Step S41: Obtain the current real-time working parameters of the welding torch; Step S42: Perform multi-stage trajectory segmentation on the optimal smooth movement trajectory of the welding torch, so as to obtain multi-stage welding trajectories; Step S43: According to the welding visual effect of the sign, calculate the welding process parameters for each segment of the multi-stage welding trajectory, so as to obtain the welding process parameters of each segment of the trajectory; Step S44: Calculate the welding current for the current real-time working parameters of the welding torch to obtain the current real-time welding current of the welding torch; Step S45: Identify the welding speed of the welding torch for the current real-time working parameters of the welding torch, and extract the welding speed of the current welding torch; Step S46: Based on the welding process parameters of each segment of the trajectory, perform adaptive welding torch parameter adjustment on the current real-time welding current and the welding speed of the current welding torch, so as to obtain the adaptive welding torch adjustment parameters.

6. The welding control method for identification signs according to claim 1, characterized in that The specific steps of step S5 are as follows: Step S51: Analyze the heat distribution in the welding area of the real-time welding monitoring images to generate the heat distribution characteristics of the welding area; Step S52: Based on the heat distribution characteristics of the welding area, evaluate the state of the welding molten pool to obtain the state characteristics of the welding molten pool; Step S53: Analyze the temperature change trend of the state characteristics of the welding molten pool to generate the temperature change trend of the welding molten pool; Step S54: According to the temperature change trend of the welding molten pool, perform abnormal temperature prediction evolution to generate the abnormal temperature prediction data of the welding molten pool; Step S55: Based on the abnormal temperature prediction data of the welding molten pool, perform dynamic regulation of the welding parameters on the adaptive welding torch adjustment parameters, so as to construct a dynamic welding parameter regulation strategy.

7. The welding control method for a sign according to claim 1, characterized in that, The specific steps of step S6 are as follows: Step S61: Based on the real-time welding monitoring images, identify the welding results of multiple segments, and extract the welding results of multiple process stages; Step S62: Evaluate the welding quality of each welding point for the welding results of multiple process stages, and generate the welding quality evaluation value of each welding point; Step S63: Perform local welding defect analysis based on the welding quality evaluation value of each welding point, so as to generate local welding defect data; Step S64: Optimize welding defect control for the dynamic welding parameter regulation strategy based on the local welding defect data, so as to construct a dynamic welding control optimization model.

8. A welding control device for a signboard, characterized in that, For implementing the welding control method for identification signs as described in claim 1, including: A welding vision simulation module, configured to obtain a high-definition monitoring image of the identification sign to be welded and a welding work log; perform texture vision recognition on the high-definition monitoring image of the identification sign to be welded, and perform welding vision effect simulation, so as to obtain the welding vision effect of the identification sign; A welding path planning module, configured to accurately identify welding points based on the welding vision effect of the identification sign, and perform dynamic welding path planning, so as to construct a dynamic welding path; A welding torch movement trajectory module, configured to perform real-time welding control on the identification sign to be welded based on the dynamic welding path, and then perform smoothing optimization on the welding torch movement trajectory, so as to construct an optimal welding torch movement smooth trajectory; An adaptive parameter adjustment module, configured to obtain the real-time working parameters of the current welding torch; perform adaptive welding torch parameter adjustment on the real-time working parameters of the current welding torch based on the optimal welding torch movement smooth trajectory, so as to obtain an adaptive welding torch adjustment parameter; A dynamic welding regulation module, configured to analyze the change trend of the welding molten pool temperature according to the real-time welding monitoring image, and perform dynamic regulation of welding parameters on the adaptive welding torch adjustment parameter, so as to construct a dynamic welding parameter regulation strategy; A welding control optimization module, configured to perform local welding defect analysis based on the real-time welding monitoring image, and optimize welding defect control for the dynamic welding parameter regulation strategy, so as to construct a dynamic welding control optimization model.

Images