Autonomous controllable high-power laser two-dimensional cutting control method and system

Abstract

The invention relates to the field of laser cutting control, in particular to an autonomous controllable high-power laser two-dimensional cutting control method and system. The method comprises the following steps: acquiring a real-time monitoring image of a to-be-cut steel structure workpiece and a preset cutting control log; performing three-dimensional form point cloud modeling on the real-time monitoring image of the to-be-cut steel structure workpiece to construct a three-dimensional workpiece form model; full-stage cutting demand mining is carried out based on a preset cutting control log, and self-adaptive laser power adjustment is carried out, so that self-adaptive laser power parameters are obtained; carrying out multi-time-point cutting form demand analysis and dynamic cutting path planning on the three-dimensional workpiece form model so as to generate a multi-time-point dynamic cutting path; and real-time cutting control is conducted according to the self-adaptive laser power parameters and the multi-time-point dynamic cutting path, cutting track dynamic optical flow tracking is conducted, and a time sequence laser cutting track sequence is constructed. According to the laser two-dimensional cutting control system, efficient and accurate laser two-dimensional cutting control is achieved.

Description

Technical Field

[0001] The present invention relates to the field of laser cutting control, and in particular to an autonomous and controllable high-power laser two-dimensional cutting control method and system. Background Art

[0002] With the continuous development of modern industrial manufacturing technology, steel structure cutting, as a key processing step, is widely used in industries such as construction, shipbuilding, aerospace, and transportation. Especially in large-scale production, the quality and efficiency of steel structure cutting directly impact the overall production schedule and cost. Traditional steel structure cutting methods, such as mechanical cutting and flame cutting, are increasingly unable to meet increasingly stringent cutting requirements due to their limited precision and efficiency. With the continuous advancement of laser technology, laser cutting technology, with its high precision, high efficiency, and contactless cutting method, has gained widespread application in the steel structure cutting field.

[0003] However, while high-power laser cutting technology can provide relatively ideal accuracy and speed in the steel structure cutting process, it still faces some challenges. In actual applications, factors such as the complex shape of steel structures, the heterogeneity of different materials, and the thermal effects during the cutting process often lead to problems such as unstable cutting paths, insufficient cutting accuracy, and even abnormal cutting. In addition, since the adjustment of laser power usually relies on manual settings or fixed parameters, the laser cutting effect may not be consistent when facing different workpieces or changing cutting conditions.

[0004] In order to improve cutting quality, reduce cutting errors, and increase cutting efficiency, traditional laser cutting methods have gradually exposed their shortcomings, especially in terms of cutting path planning, laser power adjustment, and real-time error compensation. Traditional cutting path planning often relies on experience and rules, lacking the ability to provide real-time feedback and flexible adjustment. As a result, in the cutting process of complex or variable steel workpieces, problems such as cutting deviations or excessive heat-affected zones often occur. In addition, the current adjustment of laser power is mostly based on manual settings, which makes it difficult to dynamically adjust according to the real-time needs of the workpiece, thus affecting the overall quality and consistency of the cutting. Therefore, against this background, it is particularly urgent to develop a high-power laser two-dimensional cutting control method based on intelligent control. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention proposes an autonomous and controllable high-power laser two-dimensional cutting control method and system to solve at least one of the above technical problems.

[0006] To achieve the above objectives, the present invention provides an autonomous and controllable high-power laser two-dimensional cutting control method, comprising the following steps: Step S1: obtaining a real-time monitoring image of a steel structure workpiece to be cut and a preset cutting control log; performing three-dimensional morphological point cloud modeling on the real-time monitoring image of the steel structure workpiece to be cut to construct a three-dimensional workpiece morphological model; Step S2: mining the full-stage cutting requirements based on the preset cutting control log, and performing adaptive laser power adjustment to obtain adaptive laser power parameters; Step S3: performing multi-time point cutting shape requirement analysis and dynamic cutting path planning on the three-dimensional workpiece shape model, thereby generating a multi-time point dynamic cutting path; Step S4: performing real-time cutting control based on the adaptive laser power parameters and the multi-time point dynamic cutting path, and performing dynamic optical flow tracking of the cutting trajectory to construct a time-series laser cutting trajectory sequence; Step S5: performing real-time error calculation on the sequential laser cutting trajectory sequence and performing cutting path error compensation to construct a compensated optimized cutting path; Step S6: fine-tune the thermal abnormality laser power of the adaptive laser power parameters, and perform intelligent cutting control iterative optimization according to the compensation optimized cutting path to build an intelligent laser cutting control model.

[0007] By capturing real-time monitoring images of the steel workpiece to be cut and converting them into 3D point cloud data, the present invention accurately constructs a 3D morphological model of the workpiece. This real-time modeling method provides highly accurate workpiece geometric features for subsequent cutting path planning and parameter adjustment. Through 3D point cloud modeling, the system can analyze the shape and dimensions of the steel workpiece in detail and identify complex geometric features (such as holes and uneven surfaces), providing precise data support for subsequent cutting requirements analysis and path planning. A pre-set cutting control log provides historical cutting data that reflects the material's physical properties, cutting parameters, and historical issues. This data enables the system to adjust the cutting control strategy based on the characteristics of different steel materials, thereby improving cutting accuracy and efficiency. By analyzing the pre-set cutting control log, the system can extract specific cutting requirements from multiple stages (such as preheating, cutting, and cooling). This process helps the system evaluate parameter changes throughout the entire cutting process at a macro level, ensuring the stability and efficiency of the cutting process. The system dynamically adjusts the laser power based on the cutting requirements at each stage, avoiding the energy waste and instability associated with fixed laser power in traditional methods. Adaptive adjustment optimizes energy usage, minimizes heat-affected zones, and reduces material waste while ensuring cutting accuracy. By precisely analyzing the various requirements during the cutting phase, the system can adapt in real time to changes in steel thickness, material, and shape, ensuring that cutting accuracy and quality meet requirements at each stage. By analyzing the 3D workpiece morphology model, the system generates a dynamic cutting path, avoiding the accuracy and efficiency issues that can arise from fixed paths in traditional path planning. Path planning can be adjusted in real time based on the workpiece geometry and thermal effects, improving cutting flexibility and efficiency. Cutting requirements vary at each point in time. Based on multi-point analysis, the system accurately predicts the cutting requirements (such as cutting speed, cutting sequence, and power allocation) for each stage, laying the foundation for efficient and accurate cutting path generation. Timing analysis and dynamic planning enable the system to make subtle adjustments during the cutting process, such as dynamically correcting the path to address issues such as heat-affected zones and material deformation, thereby improving overall cutting quality. Adaptive laser power and dynamic cutting paths enable the system to precisely control laser power, focal length, and cutting speed at every moment. Real-time adjustment of laser parameters ensures cutting stability and accuracy. Using dynamic optical flow tracking technology, the system can capture and analyze cutting trajectories in real time, obtaining precise trajectory data. This provides a crucial basis for subsequent cutting quality analysis and error correction. Combined with adaptive power regulation and dynamic cutting paths, the system can flexibly adjust cutting strategies for different workpieces or cutting requirements, enhancing cutting diversity and flexibility. By monitoring the laser cutting trajectory sequence in real time and calculating errors, the system can promptly detect and correct errors in the cutting process, helping to improve cutting accuracy and ensure that the final workpiece meets design requirements.During the cutting process, errors are inevitable, particularly those caused by thermal effects and laser beam instability. By calculating errors in real time, the system can compensate for these errors, dynamically optimizing the cutting path and improving cutting quality. This error compensation technology effectively reduces cutting path deviation, ensuring the accuracy of each cutting point and enhancing the quality and consistency of the entire cutting process. By monitoring thermal anomalies during the laser cutting process, the system can detect problems such as overheating and material melting. Based on this anomaly information, the system can precisely fine-tune the laser power to minimize thermal damage and optimize cutting results. By optimizing the laser power at each cutting stage, the system effectively controls heat distribution, avoids excessive heat-affected zones, and reduces material deformation and uneven cutting caused by overheating. The system intelligently adjusts the cutting path based on real-time feedback and anomalies, making the cutting process more adaptive and achieving more stable and precise cutting results. Through continuous optimization and iteration, the system ultimately builds an efficient and precise laser cutting control model, significantly improving cutting accuracy and efficiency.

[0008] In this specification, an autonomous and controllable high-power laser two-dimensional cutting control system is provided, which is used to execute the autonomous and controllable high-power laser two-dimensional cutting control method described above, including: A three-dimensional morphology module obtains a real-time monitoring image of the steel structure workpiece to be cut and a preset cutting control log; performs three-dimensional morphology point cloud modeling on the real-time monitoring image of the steel structure workpiece to be cut to construct a three-dimensional workpiece morphology model; The laser power adjustment module mines the full-stage cutting requirements based on the preset cutting control log and performs adaptive laser power adjustment to obtain adaptive laser power parameters; The cutting path planning module performs multi-time point cutting shape demand analysis and dynamic cutting path planning on the 3D workpiece shape model, thereby generating a multi-time point dynamic cutting path; The cutting trajectory module performs real-time cutting control based on adaptive laser power parameters and multi-time point dynamic cutting paths, and performs dynamic optical flow tracking of the cutting trajectory to construct a time-series laser cutting trajectory sequence; The path error compensation module calculates the error of the sequential laser cutting trajectory in real time and compensates for the cutting path error to construct a compensated and optimized cutting path. The intelligent laser control module fine-tunes the thermal abnormality laser power of the adaptive laser power parameters, and performs iterative optimization of intelligent cutting control based on the compensation optimized cutting path to build an intelligent laser cutting control model.

[0009] The present invention uses a three-dimensional morphological model to accurately represent the shape and structural features of a workpiece, including complex curved surfaces, holes, and edges. Compared to traditional methods, this 3D modeling approach improves accuracy and reduces errors during the workpiece cutting process. 3D modeling technology can accurately identify irregularities or changing geometric features on the workpiece surface during real-time cutting, ensuring precise alignment of the cutting path with the workpiece shape. The 3D morphological model provides strong geometric data support for subsequent cutting path planning, laser power adjustment, and error compensation, ensuring the accuracy and efficiency of the cutting strategy. Through adaptive laser power adjustment, the system can adjust laser output in real time based on material properties and the cutting stage, avoiding energy waste and uneven cutting caused by fixed power. Dynamic adjustment of laser power reduces the extent of the heat-affected zone, effectively controls heat accumulation, and prevents deformation or burning of the workpiece due to overheating. Adaptive power adjustment automatically adapts to the physical properties and thickness of different steel materials, improving cutting quality and consistency across different workpieces. Traditional cutting path planning methods are mostly static, while this module dynamically adjusts the path based on real-time feedback from the workpiece, improving cutting flexibility and accuracy. Through precise path planning, the system effectively avoids redundant cuts and unnecessary pauses, optimizes the cutting process, and improves work efficiency. During the cutting process, steel may deform slightly due to heat. This module adjusts the path based on these deformations to ensure cutting accuracy. Real-time laser cutting control ensures that the laser beam precisely follows the planned path, avoiding deviations and inaccurate cutting. Using optical flow tracking technology, the system instantly captures the actual laser cutting trajectory and performs comparative analysis to ensure high consistency between the cutting path and the planned path. Real-time trajectory monitoring helps the system identify and correct deviations during the cutting process, significantly improving cutting stability and reliability. Real-time error calculation allows the system to promptly detect and compensate for path deviations, ensuring accurate correction at every cutting point. Error compensation effectively reduces cutting inconsistencies caused by factors such as equipment errors and thermal expansion, improving final cut quality. By continuously compensating and optimizing the cutting path, the system can adapt to various complex cutting situations and maintain an efficient and precise cutting process. By monitoring thermal anomalies generated during the laser cutting process, the system dynamically adjusts laser power to avoid material damage or deformation caused by overheating. The intelligent control system continuously optimizes the cutting process, automatically adjusting cutting parameters and paths based on real-time feedback to improve cutting stability. Iterative optimization improves cutting accuracy: The intelligent control module not only optimizes the cutting path and power regulation, but also continuously iterates and improves to ensure optimal performance at every stage of cutting, ultimately achieving efficient and precise steel structure cutting. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1This is a schematic flow chart of the steps of an autonomous and controllable high-power laser two-dimensional cutting control method of the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Schematic diagram of the detailed implementation steps of step S3. DETAILED DESCRIPTION

[0011] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0012] This application provides an autonomous and controllable high-power laser two-dimensional cutting control method and system. The execution entities of the autonomous and controllable high-power laser two-dimensional cutting control method and system include, but are not limited to, the following: mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc. equipped with the system, which can be regarded as general computing nodes of this 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.

[0013] See also Figures 1 to 4 The present invention provides an autonomous and controllable high-power laser two-dimensional cutting control method, which includes the following steps: Step S1: obtaining a real-time monitoring image of a steel structure workpiece to be cut and a preset cutting control log; performing three-dimensional morphological point cloud modeling on the real-time monitoring image of the steel structure workpiece to be cut to construct a three-dimensional workpiece morphological model; Step S2: mining the full-stage cutting requirements based on the preset cutting control log, and performing adaptive laser power adjustment to obtain adaptive laser power parameters; Step S3: performing multi-time point cutting shape requirement analysis and dynamic cutting path planning on the three-dimensional workpiece shape model, thereby generating a multi-time point dynamic cutting path; Step S4: performing real-time cutting control based on the adaptive laser power parameters and the multi-time point dynamic cutting path, and performing dynamic optical flow tracking of the cutting trajectory to construct a time-series laser cutting trajectory sequence; Step S5: performing real-time error calculation on the sequential laser cutting trajectory sequence and performing cutting path error compensation to construct a compensated optimized cutting path; Step S6: fine-tune the thermal abnormality laser power of the adaptive laser power parameters, and perform intelligent cutting control iterative optimization according to the compensation optimized cutting path to build an intelligent laser cutting control model.

[0014] By capturing real-time monitoring images of the steel workpiece to be cut and converting them into 3D point cloud data, the present invention accurately constructs a 3D morphological model of the workpiece. This real-time modeling method provides highly accurate workpiece geometric features for subsequent cutting path planning and parameter adjustment. Through 3D point cloud modeling, the system can analyze the shape and dimensions of the steel workpiece in detail and identify complex geometric features (such as holes and uneven surfaces), providing precise data support for subsequent cutting requirements analysis and path planning. A pre-set cutting control log provides historical cutting data that reflects the material's physical properties, cutting parameters, and historical issues. This data enables the system to adjust the cutting control strategy based on the characteristics of different steel materials, thereby improving cutting accuracy and efficiency. By analyzing the pre-set cutting control log, the system can extract specific cutting requirements from multiple stages (such as preheating, cutting, and cooling). This process helps the system evaluate parameter changes throughout the entire cutting process at a macro level, ensuring the stability and efficiency of the cutting process. The system dynamically adjusts the laser power based on the cutting requirements at each stage, avoiding the energy waste and instability associated with fixed laser power in traditional methods. Adaptive adjustment optimizes energy usage, minimizes heat-affected zones, and reduces material waste while ensuring cutting accuracy. By precisely analyzing the various requirements during the cutting phase, the system can adapt in real time to changes in steel thickness, material, and shape, ensuring that cutting accuracy and quality meet requirements at each stage. By analyzing the 3D workpiece morphology model, the system generates a dynamic cutting path, avoiding the accuracy and efficiency issues that can arise from fixed paths in traditional path planning. Path planning can be adjusted in real time based on the workpiece geometry and thermal effects, improving cutting flexibility and efficiency. Cutting requirements vary at each point in time. Based on multi-point analysis, the system accurately predicts the cutting requirements (such as cutting speed, cutting sequence, and power allocation) for each stage, laying the foundation for efficient and accurate cutting path generation. Timing analysis and dynamic planning enable the system to make subtle adjustments during the cutting process, such as dynamically correcting the path to address issues such as heat-affected zones and material deformation, thereby improving overall cutting quality. Adaptive laser power and dynamic cutting paths enable the system to precisely control laser power, focal length, and cutting speed at every moment. Real-time adjustment of laser parameters ensures cutting stability and accuracy. Using dynamic optical flow tracking technology, the system can capture and analyze cutting trajectories in real time, obtaining precise trajectory data. This provides a crucial basis for subsequent cutting quality analysis and error correction. Combined with adaptive power regulation and dynamic cutting paths, the system can flexibly adjust cutting strategies for different workpieces or cutting requirements, enhancing cutting diversity and flexibility. By monitoring the laser cutting trajectory sequence in real time and calculating errors, the system can promptly detect and correct errors in the cutting process, helping to improve cutting accuracy and ensure that the final workpiece meets design requirements.During the cutting process, errors are inevitable, particularly those caused by thermal effects and laser beam instability. By calculating errors in real time, the system can compensate for these errors, dynamically optimizing the cutting path and improving cutting quality. This error compensation technology effectively reduces cutting path deviation, ensuring the accuracy of each cutting point and enhancing the quality and consistency of the entire cutting process. By monitoring thermal anomalies during the laser cutting process, the system can detect problems such as overheating and material melting. Based on this anomaly information, the system can precisely fine-tune the laser power to minimize thermal damage and optimize cutting results. By optimizing the laser power at each cutting stage, the system effectively controls heat distribution, avoids excessive heat-affected zones, and reduces material deformation and uneven cutting caused by overheating. The system intelligently adjusts the cutting path based on real-time feedback and anomalies, making the cutting process more adaptive and achieving more stable and precise cutting results. Through continuous optimization and iteration, the system ultimately builds an efficient and precise laser cutting control model, significantly improving cutting accuracy and efficiency.

[0015] In the embodiment of the present invention, see Figure 1 , is a schematic flow chart of the steps of an autonomous and controllable high-power laser two-dimensional cutting control method of the present invention. In this example, the steps of the autonomous and controllable high-power laser two-dimensional cutting control method include: Step S1: obtaining a real-time monitoring image of a steel structure workpiece to be cut and a preset cutting control log; performing three-dimensional morphological point cloud modeling on the real-time monitoring image of the steel structure workpiece to be cut to construct a three-dimensional workpiece morphological model; In this embodiment, a suitable image acquisition device is selected, typically a high-resolution industrial camera or laser scanner. The device should have real-time data acquisition capabilities and be able to continuously capture images of the steel workpiece during the cutting process. Image acquisition parameters, including resolution, frame rate, and exposure time, are set. For example, a resolution of 1920x1080 and an acquisition rate of 30 frames per second are selected to ensure that subtle changes in the cutting process are captured. Before cutting begins, the device is calibrated to ensure accurate and consistent image capture. This process can include focus, white balance, and distortion correction, using standard objects for calibration to obtain accurate image data. A cutting control log is collected for the workpiece to be cut. This log should include cutting parameters (such as laser power, cutting speed, material thickness, etc.) and operator instructions. To ensure real-time update and accuracy of the log data, the cutting control system is typically used to automatically record relevant data. A timestamp can be set for each cutting operation to facilitate subsequent analysis. The captured cutting control log is organized into a structured dataset for subsequent analysis and comparison with the real-time monitoring image. Time synchronization between the real-time monitoring image and the pre-set cutting control log is ensured. Timestamps can be used to correlate images and log data to ensure that each frame corresponds to the correct cutting parameters. Using timestamps to match image and log data ensures accurate data for subsequent analysis. For example, during the cutting process, record the specific time of image acquisition in real time and compare it with the operation time in the log. Select an appropriate 3D modeling technology, such as laser scanning or structured light scanning, to acquire 3D point cloud data of the steel workpiece to be cut. Laser scanners can capture a large amount of point data in a short period of time, creating a dense point cloud. Set scanning parameters such as scanning range, resolution, and sampling frequency. Typically, a high resolution (such as 1mm) is selected to ensure that workpiece details are captured, and the scanning frequency should be fast enough to accommodate the dynamic changes in the cutting process. The captured point cloud data is preprocessed, including denoising, filtering, and resampling. Denoising can use statistical outlier removal to ensure point cloud quality. The processed point cloud data is then registered to ensure accurate alignment of point clouds acquired from different viewpoints. This can be achieved through feature matching and the Iterative Closest Point (ICP) algorithm, ensuring the generated point cloud model is continuous and consistent. Construct a 3D workpiece morphology model based on the processed point cloud data. Use 3D modeling software (such as MeshLab or Blender) to convert the point cloud into a mesh model, generating polygonal surfaces. During model construction, set an appropriate mesh subdivision level to ensure the model maintains detail while avoiding excessive complexity. The resulting 3D model should clearly reflect the workpiece's geometric features and the locations of cut surfaces. Verify the generated 3D workpiece morphology model to ensure its geometric features match the actual workpiece. This can be done by physically comparing it to the workpiece or using a CAD model.Perform visual presentation to facilitate subsequent cutting path planning and quality control. Use 3D visualization tools to display the model and ensure it is clearly visible from different viewing angles.

[0016] Step S2: mining the full-stage cutting requirements based on the preset cutting control log, and performing adaptive laser power adjustment to obtain adaptive laser power parameters; In this embodiment, the goal of cutting demand mining is to extract key parameters related to the cutting process from the pre-set cutting control log. These parameters include laser power, cutting speed, material type, thickness, and cutting path. The analytical framework for demand mining typically employs data mining methods, such as cluster analysis and association rule mining, to identify changes in demand across different cutting stages. These changes can be identified by analyzing the time series characteristics of the cutting control log. Data cleansing is performed on the cutting control log to remove duplicate and invalid data entries to ensure data integrity and accuracy. This process can include removing incomplete records and correcting incorrect timestamps. The cleaned data is formatted and converted into a structured database for subsequent analysis. The data can be divided into multiple dimensions, such as time, laser power, and cutting speed, to facilitate analysis from different perspectives. Data mining techniques are applied to the cutting control log. For example, cluster analysis methods (such as K-means) can be used to divide the cutting process into multiple stages and identify common requirements across these stages. By analyzing the power requirements, cutting speed, and material properties of each stage, the optimal laser power setting for different cutting conditions can be identified. This process can be accomplished by constructing a data model to explore the relationships between these parameters. Based on the results of the demand analysis, an adaptive laser power adjustment strategy is developed. A baseline power value and adjustment range are typically set based on the characteristics and thickness of the cutting material. For example, for steel cutting, a power range of 1kW to 4kW can be set. During the cutting process, the laser power is dynamically adjusted based on real-time monitoring data and the preset adjustment strategy. For example, if a decrease in cutting speed or an increase in material thickness is detected, the laser power is automatically increased. The laser power parameters after each adjustment are recorded and monitored in real time to ensure stable and consistent cutting quality.

[0017] Step S3: performing multi-time point cutting shape requirement analysis and dynamic cutting path planning on the three-dimensional workpiece shape model, thereby generating a multi-time point dynamic cutting path; In this embodiment, analysis criteria for a three-dimensional workpiece morphology model are defined to identify key features of interest during the cutting process. These features include the workpiece's geometry, the location of the cutting surface, the material thickness, and its physical properties. Analysis objectives are set, such as ensuring that the cutting path effectively covers all areas requiring cutting, while also considering cutting efficiency and the material's heat-affected zone. Consider using CAD software for preliminary geometric analysis to extract key model parameters. Based on cutting control logs and real-time monitoring data, cutting requirements information is collected at multiple points in time, including parameters such as cutting speed, laser power, and material properties, and combined with workpiece morphology data acquired from the three-dimensional model. During this process, a data collection interval is set, such as recording data every 500 milliseconds, to ensure sufficient cutting status information is obtained. The recorded data is organized into a structured dataset for subsequent analysis. A suitable dynamic cutting path planning algorithm is selected, such as the graph-based A* algorithm or the Dijkstra algorithm. These algorithms can effectively plan the optimal cutting path, taking into account the workpiece's geometric characteristics and dynamic changes during the cutting process. Algorithm parameters are set, including heuristic functions and path evaluation criteria, to ensure the optimization and effectiveness of the cutting path. Different path costs can be weighted based on the priority of the cutting target. Using a selected algorithm, dynamic cutting path planning is performed on the 3D workpiece model. Real-time cutting requirements are input and combined with the geometric features of the 3D model to generate a cutting path tailored to the current cutting state. During the planning process, factors such as cutting speed, material thickness, and laser power are considered to ensure that the cutting path not only covers all necessary cutting areas but also effectively minimizes thermal effects during the cutting process. The generated dynamic cutting path is optimized and simulated to ensure path feasibility and cutting performance. Visual analysis of the cutting path is performed using simulation software (such as MATLAB, Simulink, or a dedicated cutting simulation tool). During the simulation, the impact of the cutting path on cutting quality, production efficiency, and material utilization is evaluated. Fine-tune the path based on the simulation results to achieve optimal cutting results. The planned dynamic cutting path at multiple time points is organized into a comprehensive report, including the cutting path, cutting parameters, and corresponding status information at each time point. This ensures that the report accurately reflects changes during the cutting process. Visualizing the dynamic cutting path helps technicians intuitively understand the cutting process and its dynamic characteristics, providing a reference for subsequent implementation.

[0018] Step S4: performing real-time cutting control based on the adaptive laser power parameters and the multi-time point dynamic cutting path, and performing dynamic optical flow tracking of the cutting trajectory to construct a time-series laser cutting trajectory sequence; In this embodiment, the real-time parameters of the cutting control system are set based on the adaptive laser power parameters generated in the previous step. This ensures that the laser cutting machine can dynamically adjust its power based on real-time feedback to adapt to changes in the cutting path. The laser cutting machine's start and stop conditions are set, typically including information such as laser power, cutting speed, and cutting depth. These parameters must be updated in real time during the cutting process, for example, via a PLC (programmable logic controller) or computer control system. During the cutting process, the multi-point dynamic cutting path is input into the cutting control system. Based on the set cutting parameters and path, the laser cutting machine begins cutting. During cutting, the cutting status, including laser power, cutting speed, and material feedback, is monitored in real time to ensure that the cutting process remains within the preset optimal state. For example, the cutting speed is set to 1000 mm / min, and the laser power is automatically adjusted between 1 kW and 3 kW. To perform dynamic optical flow tracking of the cutting path, an appropriate image acquisition system must first be set up. A high-resolution camera is typically used to ensure that subtle changes in the cutting process can be captured. The camera must be installed to cover the entire cutting area and calibrated to obtain accurate image data. Set the camera's frame rate, for example, 30 frames per second, to ensure real-time capture of the cutting trajectory. Use an optical flow tracking algorithm (such as the Lucas-Kanade or Farneback algorithms) to analyze the real-time cutting images and extract motion information about the cutting trajectory. These algorithms calculate motion vectors between consecutive frames, thereby identifying the cutting path. In each frame, identify the cutting point and its velocity changes, and record the motion vectors to construct a dynamic cutting trajectory. This process provides real-time feedback on the cutting path, ensuring that the laser cutting machine can adjust according to actual conditions. Organize the cutting trajectory information obtained from optical flow tracking to form a time-series laser cutting trajectory sequence. Each trajectory point should include a timestamp, coordinate information, and motion vector for subsequent analysis. Generate a complete cutting trajectory sequence report, including the cutting status and corresponding parameters at each time point. This report provides basic data for subsequent cutting quality assessment and optimization. Analyze the constructed time-series laser cutting trajectory sequence to evaluate the effectiveness of the laser power and cutting speed during the cutting process. Check the cutting quality of each trajectory point to ensure that the smoothness of the cut edge and the cutting effect meet the expected standards. Based on the analysis results, feedback is provided to adjust the cutting system parameters, optimize the laser power and cutting speed settings, and improve cutting efficiency and quality. A series of feedback mechanisms can be set to update the cutting control strategy in real time.

[0019] Step S5: performing real-time error calculation on the sequential laser cutting trajectory sequence and performing cutting path error compensation to construct a compensated optimized cutting path; In this embodiment, before performing error calculation on a time-sequential laser cutting trajectory sequence, it is necessary to first define the error calculation criteria. Generally, the error can be defined as the distance between the actual cutting path and the preset cutting path. This distance can be calculated using the Euclidean distance formula. A baseline path for error calculation, namely the ideal cutting path, is determined. This can be generated using a three-dimensional workpiece morphology model to ensure that the path design meets the cutting requirements and workpiece characteristics. During the cutting process, the actual cutting trajectory information of the laser cutting machine is collected in real time. A high-frequency data recording device (such as a sensor or a high-resolution camera) records the position and status of the cutting head at each time point. A data collection interval is set, for example, every 100 milliseconds, to ensure sufficient trajectory data is obtained. This data should include timestamps, coordinate information (X, Y, Z), and parameters such as laser power and cutting speed. After obtaining the actual cutting trajectory data, the distance between the actual trajectory and the ideal cutting path is compared point by point using the defined error calculation criteria. The error value for each trajectory point is recorded to form an error dataset. This dataset should include the error value and corresponding cutting parameters at each time point for subsequent analysis. Based on the calculated error data, an error compensation strategy for the cutting path is formulated. The compensation value should be proportional to the error. Typically, a compensation coefficient is set to ensure the rationality of the compensation amount. Compensation is calculated for each trajectory point based on the established compensation strategy. The compensation value is applied to the ideal cutting path to generate a compensated, optimized cutting path. The compensated path is recorded at each point in time, ensuring the continuity and feasibility of the compensated path. During this process, synchronization with the actual cutting process is maintained to ensure the cutting machine can respond to the compensation in real time. The generated compensated, optimized cutting path is verified to ensure it effectively reduces cutting errors. The effectiveness of the compensation path can be evaluated by comparing the actual cutting results with the expected results.

[0020] Step S6: fine-tune the thermal abnormality laser power of the adaptive laser power parameters, and perform intelligent cutting control iterative optimization according to the compensation optimized cutting path to build an intelligent laser cutting control model.

[0021] In this embodiment, monitoring criteria for thermal anomalies are established, including temperature thresholds and the definition of the heat-affected zone. Thermal anomalies can typically be identified using a real-time temperature monitoring system, such as infrared thermal imaging equipment, which can capture the temperature distribution of the cutting area in real time. A reasonable temperature threshold is set, for example, temperatures above the material melting point are considered abnormal. Historical cutting data is collected to identify temperature variation characteristics under different cutting conditions, providing a reference for subsequent anomaly identification. During the cutting process, the temperature monitoring system collects temperature data from the cutting area in real time. The acquisition frequency is set at each time point, for example, recording temperature data once per second, and ensuring data accuracy and consistency. The collected temperature data is analyzed in real time to identify thermal anomalies that exceed the set threshold. Methods such as statistical control charts are used to monitor temperature trends and promptly identify potential thermal anomalies. Based on the detected thermal anomalies, a laser power fine-tuning strategy is developed. The magnitude and direction of fine-tuning are typically determined based on material characteristics and cutting speed. For example, if an abnormal temperature increase is detected, the laser power should be reduced to reduce heat input. Fine-tuning parameters are set, for example, the power can be fine-tuned by 5%-10% each time, and dynamically adjusted based on actual monitoring conditions. During the cutting process, the developed laser power fine-tuning strategy is applied in real time. The control system adjusts the laser power in real time to ensure that the cutting process can adapt to dynamically changing temperature conditions. The laser power parameters after each fine-tuning are recorded and their impact on cut quality is monitored. For example, the fine-tuning effect is evaluated by comparing the smoothness of the cut edge and the cutting speed to ensure the stability of the cutting process. The cutting path is optimized based on compensation, and intelligent cutting control iterative optimization is implemented. The inputs of the intelligent control model are set, including real-time monitoring data, laser power, and cutting path information. Machine learning algorithms (such as reinforcement learning or neural networks) are used to optimize the cutting process. A feedback mechanism is used to continuously update the model parameters to improve cutting quality and efficiency. All collected data and feedback information are integrated to construct an intelligent laser cutting control model. The model should be able to receive input in real time and dynamically adjust the laser power and cutting path to ensure optimal cutting quality and efficiency. The model is validated and tested to ensure its effectiveness under different cutting conditions. Based on the test results, the model structure and parameter settings are further optimized to achieve the best cutting results.

[0022] In this embodiment, refer to Figure 2 , is a flowchart of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: obtaining a real-time monitoring image of the steel structure workpiece to be cut and a preset cutting control log; Step S12: performing workpiece edge detail enhancement on the real-time monitoring image of the steel structure workpiece to be cut to generate an edge detail optimized monitoring image; Step S13: performing edge contour recognition on the edge detail optimization monitoring image to extract the edge contour line of the workpiece; Step S14: performing geometric analysis based on the edge contour of the workpiece to generate geometric features of the workpiece; Step S15: Perform three-dimensional point cloud modeling on the geometric features of the workpiece to construct a three-dimensional workpiece shape model.

[0023] In this embodiment, first, select appropriate monitoring equipment, such as a high-resolution camera, to ensure it can capture clear images under various lighting conditions. The equipment should have real-time data transmission capabilities to facilitate immediate image acquisition and processing. Ensure that the camera is mounted in a position that fully covers the steel workpiece to be cut. Typically, multiple angles are selected to enhance monitoring accuracy and comprehensiveness. The camera is used for real-time image acquisition, with a set acquisition frequency, for example, one image per second, to ensure that dynamic changes in the workpiece are captured. Images should be saved in a high-quality format (such as PNG or TIFF) for easy subsequent processing. The monitoring system should include preliminary edge detection and image preprocessing capabilities to automatically remove blur and noise during data transmission, ensuring image quality. While acquiring real-time images, the cutting control log is extracted from the cutting control system. This log should include information such as cutting parameters (such as cutting speed, temperature, and pressure), operation time, and cutting path. Ensure that the log data is formatted in a unified format to facilitate subsequent data integration and analysis. CSV or JSON formats are typically used to facilitate association with image data. Before enhancing edge details in real-time monitoring images, image preprocessing is performed. This includes adjusting the image's brightness and contrast to improve image clarity and detail visibility. A filter (such as a high-pass filter) is used to remove noise and blur from the image, ensuring that subsequent processing accurately captures the artifact's edge details. Edge enhancement algorithms (such as the Canny edge detector or the Sobel operator) are then used to enhance edge details in the preprocessed image. The Canny algorithm is widely used for its excellent edge detection capabilities, effectively identifying high-frequency signals in images. When performing edge enhancement, algorithm parameters (such as the threshold setting) are adjusted to optimize edge detection and ensure that the extracted edge details accurately reflect the artifact's geometry. The enhanced edge details are then fused with the original image to generate an edge-detail-optimized monitoring image. This image should clearly display the artifact's edges to facilitate subsequent edge contour detection. The resulting optimized monitoring image is stored in a high-quality format for subsequent processing and analysis. Select an appropriate edge contour detection algorithm, such as the Hough transform or a contour detection algorithm (such as the findContours function in OpenCV). The Hough transform is suitable for detecting straight lines, while contour detection is more suitable for complex shapes. Based on the workpiece's geometric characteristics, algorithm parameters are determined to improve recognition accuracy. For example, the Hough transform threshold parameter can be set to control detection sensitivity. The selected recognition algorithm is applied to the edge detail-optimized monitoring image to extract the workpiece's edge contours. During the recognition process, the coordinates of each contour are recorded to form a contour description. Ensure that multiple contours are properly processed during the recognition process, especially when there are overlapping or similar edges on the workpiece, to avoid false detections.The extracted edge contours are stored in a structured format (such as SVG or DXF) to facilitate subsequent geometric analysis and 3D modeling. Visualization tools are used to display the identified contours to ensure their accuracy and completeness, facilitating subsequent verification and adjustment. The geometric features to be analyzed are determined, including the contour's area, perimeter, shape factor, center point, and principal axis orientation. These features provide information about the workpiece's geometric properties and stability. Based on the analysis objectives, a feature extraction algorithm is set, such as using a shape analysis algorithm (such as Hu invariant features) to extract shape features. Geometric analysis is then performed using the extracted edge contours. Geometric features are generated by calculating the contour's geometric properties, including the contour's length, center point coordinates, and related shape indices. For complex shapes, segmented analysis can be performed to ensure that the geometric features of each component are accurately captured. The resulting geometric features are structured and recorded to form a feature database. Each workpiece's geometric features should include their corresponding contour information and analysis results. A geometric feature analysis report is generated, containing the feature values, analysis methods, and significance, to facilitate subsequent 3D modeling and decision support. Select an appropriate 3D modeling method, such as laser scanning, structured light scanning, or photogrammetry. Assume that existing edge contours can be converted into a 3D point cloud using photogrammetry. Determine the required point cloud density and accuracy, which are usually set based on the complexity of the workpiece and subsequent application requirements. For example, more complex workpieces may require a higher point cloud density. Generate a 3D point cloud based on the extracted geometric features and edge contours. By combining the 2D coordinate information of the contours with the depth information, point cloud data with spatial location is formed. During the generation process, ensure the integrity and accuracy of the point cloud to avoid model deviations caused by data loss or errors. Perform 3D modeling on the generated point cloud data and use modeling software (such as Meshlab or Blender) to process and optimize the point cloud to generate a high-quality 3D workpiece morphological model. During the modeling process, smoothing, noise reduction, and refinement are performed to ensure that the final model can truly reflect the geometry and details of the workpiece.

[0024] In this embodiment, refer to Figure 3 , is a flowchart of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: extracting material properties of the steel structure to be cut based on a preset cutting control log; Step S22: Calculating the target laser cut width based on the preset cutting control log to obtain the target cut width; Step S23: performing heat-affected zone size limitation analysis according to a preset cutting control log to generate a heat-affected zone limitation range; Step S24: performing a multi-stage target cutting shape analysis on the preset cutting control log, thereby obtaining target cutting shapes for multiple stages; Step S25: mining the full-stage cutting requirements based on the target incision width, heat-affected zone restriction range, and target cutting shapes of multiple stages, thereby generating full-stage cutting requirement features; Step S26: Adaptively adjust the laser power based on the cutting requirement characteristics of all stages, thereby obtaining adaptive laser power parameters.

[0025] In this example, log data containing cutting parameters is collected. This data typically includes information such as material type, thickness, cutting speed, laser power, gas type, and pressure. Ensure data integrity and accuracy to facilitate subsequent analysis. Data should be stored in a consistent format, typically in CSV or Excel format, for easy extraction and analysis. Determine the material properties to be extracted, such as thermal conductivity, melting point, density, and mechanical strength. These properties are critical for thermal management and cut quality during the cutting process. Use appropriate data parsing tools (such as the Python Pandas library) to extract material properties from the log data. For example, for steel, its type (e.g., low-alloy steel or stainless steel) can be identified from the cutting parameters. Analyze the extracted material properties to ensure their suitability for the cutting process. For example, assess the impact of different materials on laser cutting efficiency. Record the extracted material properties in a database to create a material property profile to facilitate subsequent cutting parameter optimization and cutting requirements analysis. Determine the target kerf width, typically based on the material properties and cutting objectives. The target width should meet engineering requirements to ensure workpiece quality. The target kerf width may be affected by factors such as laser power, cutting speed, and material thickness. Selecting an appropriate calculation method is typically based on heat conduction models and the physical principles of laser cutting. For example, empirical formulas or simulations can be used to derive the target kerf width. Calculation parameters should be set, including material thermophysical properties (such as thermal conductivity and specific heat capacity) and laser parameters during the cutting process (such as power and speed). Apply the selected method to calculate the target kerf width based on the cutting control log. Record the target width values ​​for each material and cutting condition for comparison and analysis. The calculated results should be compared with actual cutting results to verify the algorithm's accuracy and cut quality. Define the heat-affected zone (HAZ), which generally refers to the area where material properties change due to heat input during laser cutting. The size of the HAZ is closely related to cutting parameters. HAZ characteristics, including width and depth, directly affect the material's mechanical properties and subsequent processing. Select an appropriate calculation model, such as a finite element analysis (FEA) model or empirical formula, to estimate the size of the HAZ. FEA models provide more accurate results but are computationally more complex. Key parameters that influence HAZ size should be set, including laser power, cutting speed, material thickness, and heat conduction characteristics. Use data from the cutting control log to perform a heat-affected zone (HAZ) size limit analysis. Enter relevant parameters, calculate the HAZ range, and record the analysis results. Generate a HAZ limit report containing the possible HAZ variation range under different cutting conditions to facilitate subsequent cutting parameter adjustments. Identify the multiple stages that may be involved in the cutting process, including initial cut, transition cut, and final cut. The target cut shape may vary at each stage. Set cutting targets and shape standards for each stage based on the geometry and functional requirements of the workpiece.Select appropriate analysis methods, such as geometric shape analysis or contour recognition algorithms, to effectively extract cutting shape information for each stage from the cutting log. Set analysis parameters to ensure accurate identification of the target shape for each stage. Conduct multi-stage target cutting shape analysis based on the cutting control log. Record the cutting parameters for each stage and generate the corresponding target cutting shape. Compare the target shape for each stage with the actual cutting results to ensure controllability and accuracy of the cutting process. Identify the characteristics of the cutting requirements for all stages, including the target kerf width, heat-affected zone (HAZ) limits, and the cutting shape for each stage. These characteristics will help optimize the laser cutting process. Identify the impact of each characteristic on cutting quality for comprehensive analysis. Integrate the target kerf width, HAZ limits, and target cutting shapes for multiple stages to form a complete set of cutting requirement characteristics. Use data analysis tools (such as data mining algorithms) to extract potential cutting requirement patterns and characteristics from the integrated data. Analyze the integrated data to identify the cutting requirement characteristics for all stages. Record changes in requirements under different cutting conditions to facilitate subsequent process optimization. Generate a cutting requirement characteristic report containing analysis results, key characteristics, and their impact, providing a basis for subsequent adjustments. Determine the mechanism for adaptive laser power regulation and dynamically adjust the laser power according to the characteristics of the cutting requirements at all stages. The regulation mechanism should have real-time response capabilities to adapt to the requirements of different cutting stages. Set the regulation strategy, including the range and speed of power adjustment, to ensure the stability and safety of the cutting process. Select a suitable laser power calculation model, usually based on cutting efficiency and material properties. The model should be able to consider the thermophysical properties of the material and the cutting conditions. Set key parameters, such as the relationship between laser power and cutting speed, to ensure the accuracy of the calculation. Implement adaptive laser power regulation according to the characteristics of the cutting requirements at all stages. Monitor the cutting status in real time and dynamically adjust the laser power according to changes in demand. Record the adjusted laser power parameters and evaluate the effects to ensure the stability and consistency of the cutting quality.

[0026] In this embodiment, refer to Figure 4 , is a flowchart of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: marking the two-dimensional cutting surface of the three-dimensional workpiece morphology model based on the preset cutting control log, and extracting the two-dimensional surface of the workpiece to be cut; Step S32: Predicting the thermal deformation of the two-dimensional surface of the workpiece to be cut, thereby generating prediction data of the thermal deformation of the two-dimensional surface; Step S33: identifying cutting avoidance points on the two-dimensional surface of the workpiece to be cut, and marking multiple cutting avoidance points; Step S34: performing a multi-time point cutting shape demand analysis based on the full-stage cutting demand characteristics and the two-dimensional surface cutting thermal deformation prediction data, thereby generating cutting shape demands at multiple time points; Step S35: Dynamic cutting path planning is performed based on the cutting shape requirements at multiple time points and multiple cutting avoidance points, thereby generating a multi-time point dynamic cutting path.

[0027] In this embodiment, ensure that a complete 3D workpiece morphological model has been generated. This model should include detailed geometric information. 3D models are typically stored in point cloud data or CAD file formats. Import the model using 3D modeling software (such as Blender or SolidWorks) and check its completeness and accuracy. Identify the area to be cut in the model. Based on the parameters in the cutting control log (such as cutting angle and cutting position), define the geometric characteristics of the cutting surface in the 3D model. These characteristics may include the shape, size, and relative position of the cutting surface. Mark the cutting surface in the software to ensure that the marked cutting surface meets the design requirements. The marking process may involve using a projection tool to project the 3D cutting surface onto a 2D plane to facilitate subsequent extraction and analysis. Extract the marked cutting surface from the 3D model as a 2D surface. This process typically involves converting the 3D data into plane coordinates to generate a 2D view of the cutting surface. The extracted 2D surface should be saved in a standard format (such as DXF or SVG) to facilitate subsequent thermal deformation prediction and avoidance point identification. Determine the thermal physics model required for cutting thermal deformation prediction. A finite element analysis (FEA) model can be used to simulate the temperature distribution during the cutting process and its impact on material deformation. Model parameters, including the material's thermal conductivity, specific heat capacity, melting point, cutting power, and cutting speed, should be set; these parameters directly affect the thermal deformation prediction results. The extracted 2D cut surface data should be input into the thermal deformation prediction model for numerical calculations. Thermal analysis should be performed using appropriate computing tools (such as ANSYS or COMSOL) to assess the heat generated during the cutting process and its impact on the workpiece. The temperature changes at each point and the corresponding thermal deformation data should be recorded. This data should include the magnitude, direction, and distribution of deformation for subsequent analysis. The predicted results should be organized into a 2D thermal deformation prediction map. This map should display the deformation areas and extent after cutting, facilitating subsequent avoidance point identification and path planning. The generated thermal deformation prediction data should be correlated with the cutting control log to ensure consistent and reliable analysis. Criteria for defining cutting avoidance points should be established. These points are typically located near the predicted thermal deformation area, material defects, or critical structural components; cutting at these locations should be avoided. Set the recognition threshold of avoidance points to ensure that the marked avoidance points can effectively reduce the risks in the cutting process. The threshold can be based on factors such as thermal deformation amplitude and material strength. Use thermal deformation prediction data to identify avoidance points on the two-dimensional surface to be cut. Image processing algorithms (such as edge detection or region segmentation) can be used to identify potential avoidance points. Combined with the analysis results of material properties and heat-affected zones, multiple cutting avoidance points are automatically marked. These points should be clearly marked on the two-dimensional surface to facilitate subsequent path planning. The marked avoidance points are recorded and saved to form an avoidance point list. This list should contain the coordinates of each avoidance point and its related attributes (such as priority and degree of impact).Use visualization tools to display the 2D surface to be cut and its avoidance points, ensuring designers can intuitively understand the distribution of avoidance points and their impact on the cutting path. Integrate the full-stage cutting requirements with the 2D surface thermal deformation prediction data. Ensure that the integrated data includes information such as cut width, heat-affected zone, cutting shape, and time factors. Establish an analysis model, typically using time series analysis or a multivariate regression model, to identify changes in cutting requirements at each time point. Perform a multi-point cutting shape requirement analysis on the integrated data to identify cutting requirements at different time points, including cutting width, depth, speed, and shape. Record the analysis results and generate cutting shape requirement data at multiple time points to ensure that the data reflects dynamic changes during the cutting process. Organize the cutting shape requirements analyzed at multiple time points into a structured report. The report should include the cutting requirement characteristics and their changing trends at each time point to facilitate subsequent path planning and cutting strategy optimization. Ensure that the analysis results are consistent with the cutting control log to improve data reliability and operability. Design a dynamic cutting path planning model based on cutting shape requirements and avoidance points. This model should be able to respond to changes in cutting requirements in real time and flexibly adjust the cutting path. Determine path planning algorithms, such as the A* algorithm or the Dijkstra algorithm, which can effectively find the shortest path and avoid obstacles. Input the cutting shape requirements and cutting avoidance points at multiple time points into the path planning model to implement dynamic cutting path planning. Ensure that the cutting efficiency and heat-affected zone limitations are taken into account during the planning process. Record the cutting path at each stage and adjust the path based on real-time feedback to ensure the continuity and safety of the cutting process. Visualize the generated multi-time point dynamic cutting path to ensure the rationality and feasibility of the path. Use CAD tools or simulation software to display the cutting path and its relationship with the avoidance points. Perform path verification to ensure the accuracy of the path planning and the consistency of the cutting effect. Make necessary adjustments based on the verification results to optimize the cutting path.

[0028] In this embodiment, step S4 includes the following steps: Step S41: using a laser transmitter to perform real-time cutting control on the steel structure workpiece to be cut according to the adaptive laser power parameters and the multi-time point dynamic cutting path, and collecting real-time cutting monitoring video; Step S42: Optimizing the timing frame delay of the real-time cut monitoring video to obtain a delay-optimized monitoring video; Step S43: Perform cutting trajectory dynamic optical flow tracking on the delay optimization monitoring video to construct a time-series laser cutting trajectory sequence.

[0029] In this embodiment, a suitable laser emitter is selected, typically a fiber laser, with a power and wavelength that meet the requirements for steel structure cutting. Before cutting, the laser must be calibrated to ensure that its output power, beam quality, and cutting accuracy meet preset standards. Based on the adaptive laser power parameters, the laser operating mode, including cutting speed, laser power, and gas-assisted cutting parameters, is set. Ensure that the parameter settings are suitable for different materials and cutting shapes. Based on the multi-point dynamic cutting path, the laser emitter is activated for real-time cutting control. The laser should dynamically adjust the cutting power and path based on real-time monitoring feedback to ensure the stability and efficiency of the cutting process. A real-time monitoring system must be connected to the laser emitter to quickly respond to changes during the cutting process and ensure the continuity and accuracy of the cutting process. During the cutting process, a high-resolution camera is used to monitor the cutting area in real time and capture cutting monitoring video. The video should be captured at a high frame rate (e.g., 30 frames per second) to ensure clear details of the cutting process. Ensure that the video capture system is synchronized with the laser emitter to facilitate subsequent data analysis and cutting effect evaluation. The captured video should be saved in a high-quality format (such as MP4 or AVI) for subsequent processing. Perform time-series frame delay analysis on the captured real-time cutting monitoring video to identify delayed frames. Delays can occur due to factors such as camera acquisition, data transmission, and processing, impacting real-time monitoring of the cutting process. Use video analysis software to extract the timestamp information for each frame, calculate inter-frame delay, and identify frames with delays exceeding a set threshold (for example, set the threshold to 200 milliseconds). Select an appropriate optimization algorithm, such as interpolation or motion compensation, to address delayed frames. Interpolation algorithms generate new frames between delayed frames, smoothing video playback. Use optical flow to analyze motion between adjacent frames to identify the trajectory of moving objects and generate interpolated frames based on motion patterns to reduce the perceived delay in the video. Process the real-time monitoring video to generate an optimized delayed video. The processed monitoring video should have smoother playback and significantly reduced delay, ensuring clearer visualization of the cutting process. Save the optimized video and ensure its quality meets the requirements for subsequent analysis to facilitate cutting effect evaluation and path adjustment. Select a suitable dynamic optical flow tracking algorithm, such as the Lucas-Kanade optical flow method or the Horn-Schunck optical flow method. These algorithms can effectively capture the motion trajectory of objects in the video and are suitable for cutting trajectory analysis. According to the quality and content of the video frames, set the parameters of the optical flow algorithm, such as the window size and number of iterations, to improve the tracking accuracy. Perform optical flow tracking on the monitoring video after delay optimization. By analyzing the pixel motion between consecutive video frames, the cutting trajectory during the laser cutting process is extracted. The optical flow vector of each frame is recorded to form a cutting trajectory dataset. Each optical flow vector should contain the direction and speed of motion information to facilitate subsequent analysis.The extracted optical flow data is organized into a time-series laser cutting trajectory sequence to ensure that the trajectory sequence reflects the dynamic changes during the cutting process. The generated trajectory sequence should include the spatial coordinates and time information of the cutting path to facilitate subsequent analysis and optimization. The trajectory sequence is visualized to help technicians better understand the dynamic characteristics of the cutting process.

[0030] In this embodiment, the specific steps of step S42 are: Perform laser reflection analysis on real-time cutting monitoring video to generate reflection characteristics of laser cutting parts; Perform overall video brightness distribution analysis based on the reflection characteristics of the laser cutting area to generate overall video brightness distribution data; Detect overly bright areas based on the overall brightness distribution data of the video and extract overly bright areas of the video; Calculate the brightness mean of the surrounding images based on the overly bright areas of the video to generate the brightness mean of the surrounding images; Adaptively stretch and adjust the contrast of overly bright areas of the video based on the average brightness of the surrounding images, thereby generating an adaptive contrast-optimized video; Performing temporal frame decomposition on the adaptive contrast optimization video to obtain multiple temporal image frames; Performing inter-frame delay detection on multiple time-sequential image frames and extracting delayed image frames; The delayed image frames are subjected to adjacent frame delay deviation calculation to obtain an inter-frame delay deviation value; Image frame interpolation processing is performed according to the inter-frame delay deviation value, and time correction is performed to obtain a delay-optimized monitoring video.

[0031] In this embodiment, criteria for defining the reflective features of laser-cut areas are determined. Reflective features typically include changes in reflection intensity, reflection angle, and reflection area; these features can reflect cutting quality and material condition. Analysis parameters, such as a reflection intensity threshold, are set to distinguish valid reflections from background noise. Image data for each frame is extracted from the real-time cutting monitoring video. Image processing software (such as OpenCV) performs preprocessing, such as denoising and contrast enhancement, to improve the visibility of the reflective features. Edge detection algorithms (such as the Canny algorithm) are used to identify the laser-cut areas and extract possible reflective areas. The extracted reflective areas are analyzed to calculate the reflection intensity and angle. Reflection intensity can be calculated using image brightness information and quantified using the laser incident angle. The generated reflective features of the laser-cut areas are saved as structured data for subsequent analysis and comparison. Parameters for the overall brightness distribution of the video are defined, including the mean, variance, and distribution range. These parameters can reflect changes in lighting in the video and the impact of the cutting process. A time window for analysis is determined, for example, calculating the brightness distribution every 10 frames to capture brightness changes during the cutting process. Perform brightness analysis on each frame of the live cut video to extract the image's grayscale value information. Use image processing tools to calculate the average brightness and standard deviation of each frame. Organize the brightness data for each frame into a time series for subsequent analysis and visualization. Based on the extracted brightness information, generate overall brightness distribution data, which should include statistical features such as the average brightness, maximum, and minimum values ​​at each time point. Visualize the brightness distribution data and generate a brightness change curve chart to intuitively analyze brightness change trends during the cutting process. Set the detection threshold for overly bright areas based on the overall brightness distribution data.

[0032] Typically, the mean brightness plus two standard deviations is used as the upper limit for overbright areas. Determine the characteristics of overbright areas, including their location, size, and duration, to facilitate subsequent processing and analysis. Detect overbright areas in each frame using a set threshold. Use image processing tools (such as OpenCV) to perform binarization to distinguish overbright areas from normal areas. Mark overbright areas in each frame and record their location and area to form a list of overbright areas. Extract the detected overbright areas to generate an overbright area dataset containing location and characteristic information. Ensure that the data reflects lighting anomalies during video cutting. Visualize the extracted overbright areas to help technicians intuitively understand the distribution and impact of lighting issues. Determine the surrounding area of ​​the overbright area. This can typically be defined as the extended area of ​​the overbright area, for example, extending a certain pixel range (e.g., 10-20 pixels) from the center of the overbright area. Ensure that the definition of the surrounding area effectively captures the effects of lighting variations. For each detected overbright area, extract image data from the surrounding area and calculate the average brightness of that area. This can be done using the mean calculation function in the image processing tool. Record the average ambient brightness value for each overly bright area and create a dataset for subsequent processing. Organize the calculated average ambient brightness values ​​into a report that includes each overly bright area and its corresponding average ambient brightness value. This report will provide a reference for subsequent contrast adjustments. Visualize the average ambient brightness values ​​to facilitate understanding of changes in lighting distribution.

[0033] Determine an adaptive contrast stretching strategy to adjust the brightness of overly bright areas based on the average brightness of the surrounding areas. The goal is to reduce the brightness of overly bright areas and make them more harmonious with their surroundings. Set a stretch factor, typically calculated based on the difference between the average brightness of the surrounding areas and the brightness of the overly bright areas. Apply the adaptive contrast stretching algorithm to each overly bright area, adjusting its brightness based on the calculated stretch factor. This can be achieved using either histogram equalization or linear stretching. Ensure that the adjusted brightness changes effectively reduce overly bright areas while maintaining the integrity of image details. Reassemble the adjusted image frames to generate an adaptive contrast-optimized video. Ensure smooth video playback and a more balanced brightness distribution. Save the optimized video and evaluate the results to ensure that the stretching adjustment achieved the desired effect. Determine the goal of time-series frame decomposition, typically splitting the optimized video into individual frames for subsequent analysis. Set the frame rate for the decomposition, typically keeping it consistent with the original video (e.g., 30 frames per second) to ensure data integrity. Use video processing tools to decompose the adaptive contrast-optimized video and extract the image data for each frame. Ensure that the extracted frames are saved in chronological order. Save each frame in a high-quality format (such as PNG or JPEG) to facilitate subsequent analysis and processing. Organize the extracted image frames into a structured dataset. Each frame's file name should include a timestamp to facilitate subsequent analysis and traceability. Record relevant attributes of each frame, such as time and brightness value, to ensure data integrity. Determine the criteria for detecting inter-frame delay, typically based on timestamp information, and set a maximum allowable delay threshold (for example, 200 milliseconds). Use inter-frame comparison to analyze the temporal differences between video frames and identify delayed frames. Analyze the extracted time-series image frames frame by frame, calculate the time difference between adjacent frames, and identify delayed frames exceeding the set threshold. Record the timestamp and corresponding image frame number of each delayed frame for subsequent processing. Extract the identified delayed frames to form a delayed frame dataset. This dataset should include both the delayed frame's temporal information and image data to facilitate subsequent analysis. Visualize delayed frames to help technicians intuitively understand the distribution of delay phenomena and their impact.

[0034] Determine the calculation method for inter-frame delay deviation. The deviation is typically expressed as the difference between the actual frame time and the expected time. Delay deviation = actual time − expected time. Delay deviation is calculated for the extracted delayed frames, and the actual time of each delayed frame and its corresponding expected time are recorded. The delay deviation value for each frame is calculated and stored in the dataset for subsequent analysis. The calculated delay deviation values ​​are compiled into a report containing the time information and deviation value for each delayed frame. This report provides a reference for subsequent interpolation processing. Visualizing the delay deviation data helps technicians intuitively understand the severity of the delay issue. Select an appropriate interpolation method, such as linear interpolation, spline interpolation, or optical flow interpolation. Consider the nature of the delay and the characteristics of the cut surveillance video. Interpolation parameters are set to ensure that the generated interpolated frames effectively fill the gaps caused by the delay. Interpolate the delayed image frames based on the calculated inter-frame delay deviation value. Generate new frames to compensate for the gaps caused by the delay and ensure the smoothness and continuity of the entire video. Record the frames generated during the interpolation process and their time information to ensure that the timestamps of the new frames are consistent with those of the original video. Temporally correct the interpolated frames to ensure that the timing information of each frame accurately reflects the actual playback order of the video. Reassemble all processed frames to generate a latency-optimized monitoring video. Save the video in a format and quality suitable for subsequent analysis.

[0035] In this embodiment, the specific steps of step S43 are: Perform continuous frame dynamic optical flow tracking on the delay optimization monitoring video to extract the motion vector field between continuous frames; Perform dynamic cutting trajectory highlighting on the motion vector field between consecutive frames to extract the actual laser cutting trajectory; Calculate the actual laser cutting trajectory point by point in time to generate the cutting trajectory coordinates for each time point; The cutting trajectory coordinates at each time point are fitted with a time series trajectory to construct a time series laser cutting trajectory sequence.

[0036] In this embodiment, a suitable optical flow tracking algorithm is selected, such as the Lucas-Kanade optical flow method or the Farneback optical flow method. These algorithms can effectively capture the motion of objects in the video and are particularly suitable for motion vector extraction in dynamic scenes. Algorithm parameters, such as the window size, number of pyramid levels, and number of iterations of the optical flow calculation, are set according to the characteristics of the video to ensure high-precision motion estimation. Each frame of the delay-optimized monitoring video is preprocessed, including denoising and contrast enhancement, to improve the accuracy of the optical flow calculation. Gaussian filtering can be used to reduce image noise. The quality of each image frame is ensured to support the effective operation of the optical flow algorithm to facilitate subsequent motion vector field extraction. Using the selected optical flow algorithm, optical flow analysis is performed on consecutive frames to extract the motion vector between each pair of adjacent frames. Each vector in the motion vector field represents the direction and speed of motion of a small area in the image. The motion vector data between each frame is recorded and organized into a motion vector field to facilitate subsequent cutting trajectory analysis. The highlight recognition criteria for the laser cutting trajectory are generally determined based on the intensity and direction of the motion vector. During laser cutting, the direction of the motion vector should be consistent with the direction of movement of the laser beam. A threshold is set to filter out areas with motion vector intensities greater than the set value to identify highlighted trajectory areas. The extracted motion vector field is analyzed to identify trajectory points that meet the highlight criteria. Cluster analysis can be used to cluster similar motion vectors to form a complete cutting trajectory. The identified cutting trajectory points are recorded and connected to form the actual laser cutting trajectory. The extracted laser cutting trajectory is organized into a structured dataset, including the coordinates of each trajectory point and its corresponding timestamp. The extracted trajectory is visualized and compared with the actual cutting position to ensure the accuracy and validity of the extraction results. A method is determined to calculate the cutting trajectory coordinates at each time point. The position of each time point is typically calculated based on the actual laser cutting speed and time interval. A time sampling interval is set, such as calculating trajectory coordinates every 100 milliseconds, to ensure data integrity. Based on the extracted actual laser cutting trajectory, the trajectory coordinates are calculated for each time point. The coordinate data for each time point is recorded, ensuring accuracy and consistency for subsequent trajectory fitting. The calculated cutting trajectory coordinates for each time point are organized into a time series dataset. Ensure that the data includes timestamps, coordinate information, and their changing trends. Save the coordinate data to a file for later analysis and visualization. Select an appropriate trajectory fitting method, such as polynomial fitting, spline fitting, or Bezier curve fitting, which effectively captures trajectory trends and smooths the trajectory data. Depending on the complexity of the trajectory, set the polynomial order or number of spline nodes to ensure accurate fitting. Perform time series fitting on the cutting trajectory coordinates at each time point to generate a smooth trajectory line. Use the fitting algorithm to process the coordinate data to eliminate noise and irregular fluctuations.Record the fitted trajectory data, including the fitting equation and fitting error, for subsequent analysis and evaluation. Organize the fitted trajectory data into a time-series laser cutting trajectory sequence. Ensure that the sequence fully reflects the trajectory changes during the cutting process.

[0037] In this embodiment, step S5 includes the following steps: Step S51: performing real-time error calculation on the sequential laser cutting trajectory sequence to obtain the cutting accuracy error of each trajectory point; Step S52: performing cumulative error prediction on the cutting accuracy error of each trajectory point to generate a cutting trajectory cumulative prediction error; Step S53: performing advance error compensation calculation according to the cumulative prediction error of the cutting trajectory to generate a pre-trajectory error compensation value; Step S54: performing cutting path error compensation on the multi-time point dynamic cutting path based on the pre-set trajectory error compensation value to construct a compensated optimized cutting path.

[0038] In this embodiment, the calculation standard for determining the cutting accuracy error usually includes the distance between the actual trajectory and the expected trajectory. The benchmark for error calculation can be set, and a preset ideal cutting path can be used as a reference. Select a suitable error measurement method, commonly used ones include Euclidean distance and Manhattan distance, to quantify the deviation of each trajectory point. For the trajectory coordinates of each time point, calculate the error between it and the corresponding ideal trajectory point. Organize the calculated cutting accuracy errors of each trajectory point to form a time series data set. Ensure that the data contains timestamps, error values ​​and their changing trends. Generate an error statistical report containing information such as error mean, maximum value and standard deviation to provide basic data for subsequent error prediction. Determine the calculation method of the cumulative error, usually by adding the error value of each time point to the cumulative error of the previous time point. Set the initial conditions for the cumulative error calculation, usually the error value of the first time point. According to the characteristics of the cutting process, set a prediction model, such as using a simple cumulative model or a weighted model to reflect the error impact of different time points. Cumulative calculation of the cutting accuracy error at each time point is performed. Cumulative error t = cumulative error (t − 1) + error (t). The generated cumulative predicted errors of the cutting trajectory are organized into a report containing the cumulative errors and their changing trends at each time point. Compensation value t = k × cumulative error t. This report provides a basis for subsequent error compensation. Determine the construction criteria for the compensated optimized cutting path, which typically involves applying the compensation value to the original cutting path to correct the actual cutting trajectory. Set the calculation method for the compensated path to ensure that the compensated path effectively reflects the error correction during the actual cutting process. Compensate the original cutting path coordinates at each time point and apply the pre-defined trajectory error compensation value. Optimized path t = original path t + compensation value t. Record the compensated optimized cutting path at each time point and ensure the path's continuity and feasibility. Organize the calculated compensated optimized cutting path into a report containing the optimized path coordinates and their changing trends at each time point. This report supports subsequent cutting execution.

[0039] In this embodiment, the specific steps of step S6 are: Step S61: identifying the temperature value of the cutting track based on the time-sequential laser cutting track sequence; Step S62: performing a trajectory point heat distribution analysis on the temperature value to generate a trajectory point heat distribution feature; Step S63: identifying abnormal heat distribution based on the heat distribution characteristics of the trajectory points based on the heat affected zone limit range, and marking abnormal heat trajectory points; Step S64: locating the cutting stage according to the abnormal heat trajectory points and extracting the abnormal heat cutting stage; Step S65: dynamically fine-tuning the adaptive laser power parameters based on the abnormal heat cutting stage, thereby obtaining the optimized laser cutting power parameters; Step S66: Perform intelligent cutting control iterative optimization according to the laser cutting optimized power parameters and the compensation optimized cutting path to construct an intelligent laser cutting control model.

[0040] In this embodiment, temperature monitoring standards are determined during the cutting process. Generally, during laser cutting, temperature is closely related to laser power, material properties, and cutting speed. A benchmark for temperature calculation is typically set based on thermophysics models and experimental data. A suitable temperature measurement method, such as infrared thermal imaging or thermocouples, is selected to ensure accurate reflection of temperature changes at each trajectory point during the cutting process. During the cutting process, infrared thermal imaging equipment is used to monitor the temperature of the cutting trajectory in real time. The equipment should have high resolution and fast response capabilities to capture instantaneous temperature changes. Temperature data is collected for each trajectory point, and the data is denoised and calibrated to ensure accuracy and consistency. The collected temperature data is organized into a time series dataset. Each data point should include a timestamp, trajectory point coordinates, and its corresponding temperature value. A method for calculating heat distribution is typically determined by combining temperature values ​​with a heat conduction model. Analysis criteria are set, including the mean, variance, and extreme values ​​of the heat distribution. Appropriate analysis tools, such as MATLAB or Python, are selected for data calculation and visualization. Heat distribution characteristics are calculated based on the temperature values ​​of each trajectory point. Temperature values ​​can be converted to calorific values ​​using the heat conduction equation, taking into account parameters such as the thermal conductivity of the material and cutting speed. Heat distribution data for each trajectory point is recorded, and statistical features, such as the mean and standard deviation, are generated to facilitate subsequent identification of abnormal distributions. The calculated heat distribution features are compiled into a report containing the heat distribution feature data for each trajectory point. This report provides a basis for subsequent anomaly identification. Visualizing the heat distribution feature data helps technicians intuitively understand the thermal changes during the cutting process. The heat-affected zone (HAZ) is defined, typically referring to the area where material properties change due to heat input during laser cutting. Limits for the HAZ are set to facilitate subsequent anomaly identification. Based on material properties and cutting parameters, a temperature threshold for the HAZ is set, such as the portion exceeding the material's melting point. The heat distribution features of each trajectory point are analyzed, and abnormal heat trajectory points are identified within the set HAZ limits. Statistical methods, such as Z-score analysis, can be used to identify abnormal points. Trajectory points with heat distribution exceeding the set threshold are marked, and their locations and calorific values ​​are recorded to create a list of abnormal heat trajectory points. The identified abnormal heat trajectory points are organized to form a structured data set, including the coordinates, heat value, and impact of each abnormal point. The positioning criteria for the cutting stage are determined, usually based on the changes in laser power and material properties during the cutting process. The basis for dividing the stages is set, such as changes in cutting speed, changes in material thickness, etc. The key time points in the cutting process are selected as the basis for dividing the cutting stages. Based on the marked abnormal heat trajectory points, their position and time in the cutting process are analyzed to determine the cutting stage corresponding to the abnormal heat. The cutting stages under the influence of abnormal heat are extracted, and the characteristics of these stages, such as duration, power changes, etc., are recorded.The extracted abnormal heat cutting stages are organized into a report containing the timing, abnormal heat trajectory points, and characteristics of each stage. This report will support subsequent power fine-tuning. A strategy for laser power fine-tuning is determined, and laser power is adjusted based on the information from the abnormal heat cutting stages. The amplitude and direction of fine-tuning should be set. Generally, reducing heat input should be considered to prevent overheating. A power adjustment range should be set, such as reducing laser power by 5%-10% depending on the severity of the abnormal heat. Laser power is adjusted in real time based on the characteristics of the abnormal heat cutting stages. This can be achieved through a feedback control system to ensure dynamic power adjustment during the cutting process. Laser power parameters are recorded after each fine-tuning, ensuring the real-time and accuracy of the adjustment process. The framework of the intelligent laser cutting control model is determined, typically consisting of an input layer (laser power, cutting path), a processing layer (algorithm model), and an output layer (cutting results). Appropriate control algorithms, such as fuzzy control, PID control, or machine learning, are selected to enable dynamic adjustment. The intelligent cutting control model is implemented based on optimized power parameters and compensation-optimized cutting paths. Feedback information from the cutting process is monitored in real time and iteratively optimized based on the feedback. The parameters and results of each cutting process are recorded to facilitate subsequent analysis and model adjustment. Evaluate the effectiveness of the constructed intelligent laser cutting control model, analyzing indicators such as cutting quality, speed, and power consumption to ensure the model's effectiveness and reliability. Feedback the evaluation results into the model for continuous optimization to ensure the stability and efficiency of the cutting process.

[0041] In this embodiment, an autonomous and controllable high-power laser two-dimensional cutting control system is provided, which is used to execute the autonomous and controllable high-power laser two-dimensional cutting control method described above, including: A three-dimensional morphology module obtains a real-time monitoring image of the steel structure workpiece to be cut and a preset cutting control log; performs three-dimensional morphology point cloud modeling on the real-time monitoring image of the steel structure workpiece to be cut to construct a three-dimensional workpiece morphology model; The laser power adjustment module mines the full-stage cutting requirements based on the preset cutting control log and performs adaptive laser power adjustment to obtain adaptive laser power parameters; The cutting path planning module performs multi-time point cutting shape demand analysis and dynamic cutting path planning on the 3D workpiece shape model, thereby generating a multi-time point dynamic cutting path; The cutting trajectory module performs real-time cutting control based on adaptive laser power parameters and multi-time point dynamic cutting paths, and performs dynamic optical flow tracking of the cutting trajectory to construct a time-series laser cutting trajectory sequence; The path error compensation module calculates the error of the sequential laser cutting trajectory in real time and compensates for the cutting path error to construct a compensated and optimized cutting path. The intelligent laser control module fine-tunes the thermal abnormality laser power of the adaptive laser power parameters, and performs iterative optimization of intelligent cutting control based on the compensation optimized cutting path to build an intelligent laser cutting control model.

[0042] The present invention uses a three-dimensional morphological model to accurately represent the shape and structural features of a workpiece, including complex curved surfaces, holes, and edges. Compared to traditional methods, this 3D modeling approach improves accuracy and reduces errors during the workpiece cutting process. 3D modeling technology can accurately identify irregularities or changing geometric features on the workpiece surface during real-time cutting, ensuring precise alignment of the cutting path with the workpiece shape. The 3D morphological model provides strong geometric data support for subsequent cutting path planning, laser power adjustment, and error compensation, ensuring the accuracy and efficiency of the cutting strategy. Through adaptive laser power adjustment, the system can adjust laser output in real time based on material properties and the cutting stage, avoiding energy waste and uneven cutting caused by fixed power. Dynamic adjustment of laser power reduces the extent of the heat-affected zone, effectively controls heat accumulation, and prevents deformation or burning of the workpiece due to overheating. Adaptive power adjustment automatically adapts to the physical properties and thickness of different steel materials, improving cutting quality and consistency across different workpieces. Traditional cutting path planning methods are mostly static, while this module dynamically adjusts the path based on real-time feedback from the workpiece, improving cutting flexibility and accuracy. Through precise path planning, the system effectively avoids redundant cuts and unnecessary pauses, optimizes the cutting process, and improves work efficiency. During the cutting process, steel may deform slightly due to heat. This module adjusts the path based on these deformations to ensure cutting accuracy. Real-time laser cutting control ensures that the laser beam precisely follows the planned path, avoiding deviations and inaccurate cutting. Using optical flow tracking technology, the system instantly captures the actual laser cutting trajectory and performs comparative analysis to ensure high consistency between the cutting path and the planned path. Real-time trajectory monitoring helps the system identify and correct deviations during the cutting process, significantly improving cutting stability and reliability. Real-time error calculation allows the system to promptly detect and compensate for path deviations, ensuring accurate correction at every cutting point. Error compensation effectively reduces cutting inconsistencies caused by factors such as equipment errors and thermal expansion, improving final cut quality. By continuously compensating and optimizing the cutting path, the system can adapt to various complex cutting situations and maintain an efficient and precise cutting process. By monitoring thermal anomalies generated during the laser cutting process, the system dynamically adjusts laser power to avoid material damage or deformation caused by overheating. The intelligent control system continuously optimizes the cutting process, automatically adjusting cutting parameters and paths based on real-time feedback to improve cutting stability. Iterative optimization improves cutting accuracy: The intelligent control module not only optimizes the cutting path and power regulation, but also continuously iterates and improves to ensure optimal performance at every stage of cutting, ultimately achieving efficient and precise steel structure cutting.

[0043] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0044] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. An autonomous and controllable high-power laser two-dimensional cutting control method, characterized in that: The following steps are involved: Step S1: obtaining a real-time monitoring image of a steel structure workpiece to be cut and a preset cutting control log; Performing three-dimensional point cloud modeling on the real-time monitoring image of the steel structure workpiece to be cut to construct a three-dimensional workpiece morphology model; Step S2: mining the full-stage cutting requirements based on the preset cutting control log, and performing adaptive laser power adjustment to obtain adaptive laser power parameters; Step S3: performing multi-time point cutting shape requirement analysis and dynamic cutting path planning on the three-dimensional workpiece shape model, thereby generating a multi-time point dynamic cutting path; Step S4: performing real-time cutting control based on the adaptive laser power parameters and the multi-time point dynamic cutting path, and performing dynamic optical flow tracking of the cutting trajectory to construct a time-series laser cutting trajectory sequence; Step S5: performing real-time error calculation on the sequential laser cutting trajectory sequence and performing cutting path error compensation to construct a compensated optimized cutting path; Step S6: fine-tune the thermal abnormality laser power of the adaptive laser power parameters, and perform intelligent cutting control iterative optimization according to the compensation optimized cutting path to build an intelligent laser cutting control model.

2. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: obtaining a real-time monitoring image of the steel structure workpiece to be cut and a preset cutting control log; Step S12: performing workpiece edge detail enhancement on the real-time monitoring image of the steel structure workpiece to be cut to generate an edge detail optimized monitoring image; Step S13: performing edge contour recognition on the edge detail optimization monitoring image to extract the edge contour line of the workpiece; Step S14: performing geometric analysis based on the edge contour of the workpiece to generate geometric features of the workpiece; Step S15: Perform three-dimensional point cloud modeling on the geometric features of the workpiece to construct a three-dimensional workpiece shape model.

3. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: extracting material properties of the steel structure to be cut based on a preset cutting control log; Step S22: Calculating the target laser cut width based on the preset cutting control log to obtain the target cut width; Step S23: performing heat-affected zone size limitation analysis according to a preset cutting control log to generate a heat-affected zone limitation range; Step S24: performing a multi-stage target cutting shape analysis on the preset cutting control log, thereby obtaining target cutting shapes for multiple stages; Step S25: mining the full-stage cutting requirements based on the target incision width, heat-affected zone restriction range, and target cutting shapes of multiple stages, thereby generating full-stage cutting requirement features; Step S26: Adaptively adjust the laser power based on the cutting requirement characteristics of all stages, thereby obtaining adaptive laser power parameters.

4. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: marking the two-dimensional cutting surface of the three-dimensional workpiece morphology model based on the preset cutting control log, and extracting the two-dimensional surface of the workpiece to be cut; Step S32: Predicting the thermal deformation of the two-dimensional surface of the workpiece to be cut, thereby generating prediction data of the thermal deformation of the two-dimensional surface; Step S33: identifying cutting avoidance points on the two-dimensional surface of the workpiece to be cut, and marking multiple cutting avoidance points; Step S34: performing a multi-time point cutting shape demand analysis based on the full-stage cutting demand characteristics and the two-dimensional surface cutting thermal deformation prediction data, thereby generating cutting shape demands at multiple time points; Step S35: Dynamic cutting path planning is performed based on the cutting shape requirements at multiple time points and multiple cutting avoidance points, thereby generating a multi-time point dynamic cutting path.

5. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: using a laser transmitter to perform real-time cutting control on the steel structure workpiece to be cut according to the adaptive laser power parameters and the multi-time point dynamic cutting path, and collecting real-time cutting monitoring video; Step S42: Optimizing the timing frame delay of the real-time cut monitoring video to obtain a delay-optimized monitoring video; Step S43: Perform cutting trajectory dynamic optical flow tracking on the delay optimization monitoring video to construct a time-series laser cutting trajectory sequence.

6. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 5, characterized in that: The specific steps of step S42 are: Perform laser reflection analysis on real-time cutting monitoring video to generate reflection characteristics of laser cutting parts; Perform overall video brightness distribution analysis based on the reflection characteristics of the laser cutting area to generate overall video brightness distribution data; Detect overly bright areas based on the overall brightness distribution data of the video and extract overly bright areas of the video; Calculate the brightness mean of the surrounding images based on the overly bright areas of the video to generate the brightness mean of the surrounding images; Adaptively stretch and adjust the contrast of overly bright areas of the video based on the average brightness of the surrounding images, thereby generating an adaptive contrast-optimized video; Performing temporal frame decomposition on the adaptive contrast optimization video to obtain multiple temporal image frames; Performing inter-frame delay detection on multiple time-sequential image frames and extracting delayed image frames; The delayed image frames are subjected to adjacent frame delay deviation calculation to obtain an inter-frame delay deviation value; Image frame interpolation processing is performed according to the inter-frame delay deviation value, and time correction is performed to obtain a delay-optimized monitoring video.

7. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 5, characterized in that: The specific steps of step S43 are: Perform continuous frame dynamic optical flow tracking on the delay optimization monitoring video to extract the motion vector field between continuous frames; Perform dynamic cutting trajectory highlighting on the motion vector field between consecutive frames to extract the actual laser cutting trajectory; Calculate the actual laser cutting trajectory point by point in time to generate the cutting trajectory coordinates for each time point; The cutting trajectory coordinates at each time point are fitted with a time series trajectory to construct a time series laser cutting trajectory sequence.

8. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: performing real-time error calculation on the sequential laser cutting trajectory sequence to obtain the cutting accuracy error of each trajectory point; Step S52: performing cumulative error prediction on the cutting accuracy error of each trajectory point to generate a cutting trajectory cumulative prediction error; Step S53: performing advance error compensation calculation according to the cumulative prediction error of the cutting trajectory to generate a pre-trajectory error compensation value; Step S54: performing cutting path error compensation on the multi-time point dynamic cutting path based on the pre-set trajectory error compensation value to construct a compensated optimized cutting path.

9. The autonomous and controllable high-power laser two-dimensional cutting control method according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: identifying the temperature value of the cutting track based on the time-sequential laser cutting track sequence; Step S62: performing a trajectory point heat distribution analysis on the temperature value to generate a trajectory point heat distribution feature; Step S63: identifying abnormal heat distribution based on the heat distribution characteristics of the trajectory points based on the heat affected zone limit range, and marking abnormal heat trajectory points; Step S64: locating the cutting stage according to the abnormal heat trajectory points and extracting the abnormal heat cutting stage; Step S65: dynamically fine-tuning the adaptive laser power parameters based on the abnormal heat cutting stage, thereby obtaining the optimized laser cutting power parameters; Step S66: Perform intelligent cutting control iterative optimization according to the laser cutting optimized power parameters and the compensation optimized cutting path to construct an intelligent laser cutting control model.

10. An autonomous and controllable high-power laser two-dimensional cutting control system, characterized in that: The method for executing the autonomous and controllable high-power laser two-dimensional cutting control method according to claim 1 comprises: A three-dimensional morphology module obtains a real-time monitoring image of the steel structure workpiece to be cut and a preset cutting control log; performs three-dimensional morphology point cloud modeling on the real-time monitoring image of the steel structure workpiece to be cut to construct a three-dimensional workpiece morphology model; The laser power adjustment module mines the full-stage cutting requirements based on the preset cutting control log and performs adaptive laser power adjustment to obtain adaptive laser power parameters; The cutting path planning module performs multi-time point cutting shape demand analysis and dynamic cutting path planning on the 3D workpiece shape model, thereby generating a multi-time point dynamic cutting path; The cutting trajectory module performs real-time cutting control based on adaptive laser power parameters and multi-time point dynamic cutting paths, and performs dynamic optical flow tracking of the cutting trajectory to construct a time-series laser cutting trajectory sequence; The path error compensation module calculates the error of the sequential laser cutting trajectory in real time and compensates for the cutting path error to construct a compensated and optimized cutting path. The intelligent laser control module fine-tunes the thermal abnormality laser power of the adaptive laser power parameters, and performs iterative optimization of intelligent cutting control based on the compensation optimized cutting path to build an intelligent laser cutting control model.

Images