Automatic welding path optimization method based on BIM and point cloud data

Abstract

The present invention relates to the field of welding technology, and discloses an automated welding path optimization method based on BIM and point cloud data, comprising the following steps: acquiring point cloud data and BIM model data, wherein the point cloud data is acquired by laser scanning or other three-dimensional scanning technology, and the BIM model data is a design model before construction; preprocessing the point cloud data, including denoising, simplification and surface reconstruction, to improve the quality and processing efficiency of the point cloud data; and registering the point cloud data and the BIM model data. By integrating BIM and point cloud data, the accuracy and quality of the welding path are significantly improved, welding defects are reduced, a multi-objective optimization algorithm is adopted, path planning is optimized, operation time and material consumption are reduced, and welding efficiency is improved at the same time. The real-time feedback and dynamic adjustment mechanism enable the welding path to adapt to on-site geometric deformation, ensure welding quality and accuracy, and is particularly suitable for complex and dynamically changing construction environments.

Description

Technical Field

[0001] The present invention relates to the field of welding technology, and in particular to an automatic welding path optimization method based on BIM and point cloud data. Background Art

[0002] In modern welding technology, automated welding path planning and optimization have become key links in improving production efficiency and welding quality. Traditional welding path planning methods rely on manual design and static path calculation based on BIM models. These methods plan welding paths by calculating the shortest path or optimal path in the geometric model, but they cannot effectively reflect the geometric deformation, errors and complex environment in actual construction, resulting in welding quality and efficiency that are difficult to meet high-precision requirements. In recent years, the application of point cloud data has enabled welding path planning to better adapt to the complex conditions of the actual site, but how to effectively combine BIM models and point cloud data and make dynamic adjustments is still a difficult problem in current technology.

[0003] The existing technology mainly uses BIM models and point cloud data for separate path planning. The BIM model is an idealized design data, and the point cloud data is used to reflect the actual environment. Although point cloud technology plays a certain role in welding path optimization, the existing technology lacks the precise fusion of point cloud data and BIM models, and lacks an effective path optimization algorithm to reduce path errors and improve welding quality. In addition, the existing technology also has deficiencies in dynamic changes and real-time adjustments. It fails to automatically adjust the path according to changes such as thermal deformation and material expansion during the welding process, resulting in inaccurate path planning during the welding process, affecting welding quality and work efficiency. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides an automatic welding path optimization method based on BIM and point cloud data, which solves the problems in the prior art that BIM model and point cloud data cannot be effectively integrated, lacks dynamic real-time adjustment function and insufficient welding path accuracy.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an automated welding path optimization method based on BIM and point cloud data, comprising the following steps: Obtaining point cloud data and BIM model data, wherein the point cloud data is obtained by laser scanning or other three-dimensional scanning technology, and the BIM model data is a design model before construction; Preprocessing the point cloud data, including denoising, simplification and surface reconstruction, to improve the quality and processing efficiency of the point cloud data; Registering the point cloud data and the BIM model data, aligning the point cloud data with the BIM model, and eliminating differences in spatial coordinate systems and scales; Based on the registered point cloud data and BIM model data, the welding path is planned, and the preliminary welding path is generated by considering the spatial constraints of the welding area, welding sequence, and joint position. Perform multi-objective optimization on the preliminary welding path, optimize the path length, energy consumption, and welding time targets, and generate the optimal welding path; Smoothing the welding path to reduce sharp turns and unstable parts of the path; Track the welding path in real time and perform error detection, and correct the path through sensor data and visual system feedback; Monitor welding quality and adjust welding parameters based on real-time feedback to ensure welding quality meets predetermined standards; During the welding process, the path and welding parameters are dynamically updated to adapt to the geometric deformation in the construction environment and the thermal deformation during the welding process.

[0006] Preferably, the step of preprocessing the point cloud data includes: Use statistical outlier removal algorithm to remove noise points in the point cloud; Use Gaussian filter to smooth the point cloud data and remove high-frequency noise; Use Voxel Grid filter to simplify point cloud data to reduce computational complexity.

[0007] Preferably, the step of registering the point cloud data and the BIM model comprises: Based on the feature point extraction and matching method, the geometric features in the point cloud and BIM model are extracted and matched through the FPFH feature descriptor; Use the ICP algorithm for fine registration to minimize the geometric error between the point cloud and the BIM model; The registration results are further adjusted using the Levenberg-Marquardt optimization algorithm to improve accuracy.

[0008] Preferably, the welding path planning comprises the following steps: Use the A* algorithm or Dijkstra algorithm to calculate the preliminary welding path, taking into account the spatial limitations of the welding area and the welding sequence; The minimum spanning tree algorithm based on graph theory optimizes the welding path, making the path more concise and avoiding unnecessary repetition.

[0009] Preferably, the path optimization step includes: Use particle swarm optimization or genetic algorithm to perform multi-objective optimization of the path, with the optimization objectives including path length, welding time, and energy consumption; Based on the Pareto frontier theory, each objective is balanced in the multi-objective optimization to ensure the multi-dimensional optimization of the welding path.

[0010] Preferably, the path smoothing step includes: Use Bezier curve or spline interpolation method to smooth the welding path to reduce sharp turns and unstable parts in the path and ensure the smoothness of the welding process.

[0011] Preferably, the real-time tracking and error detection steps include: Use a visual perception system to obtain welding path data in real time through lidar or RGB-D camera; The Kalman filter algorithm is used to filter the welding path error and improve the stability of path tracking.

[0012] Preferably, the welding quality monitoring and feedback adjustment step includes: Use temperature sensors and stress sensors to monitor the temperature distribution, stress state and weld morphology generated during welding in real time; Based on real-time feedback data, an adaptive control algorithm is used to adjust the welding current and speed parameters to ensure that the welding quality meets the requirements.

[0013] Preferably, the step of dynamically updating the path and welding parameters comprises: The incremental ICP algorithm is used to perform real-time registration of the newly added point cloud data and adjust the welding path; Based on the multi-physics field coupling simulation model, the thermal effects and deformation during the welding process are simulated, and the welding path and welding parameters are dynamically adjusted.

[0014] Preferably, the real-time feedback and dynamic adjustment system includes: Integrate point cloud data acquisition, BIM model registration, welding path planning, real-time quality monitoring and feedback system into a unified platform to achieve closed-loop control; Through real-time feedback from sensor data and vision systems, welding paths and parameters are automatically adjusted to adapt to geometric changes and thermal deformation during construction.

[0015] The present invention provides an automated welding path optimization method based on BIM and point cloud data. It has the following beneficial effects: 1. By combining the BIM model and point cloud data, the present invention can accurately reflect the actual geometric shape and deformation of the on-site structure. The real-time acquisition of point cloud data and the precise registration of the BIM model ensure the high accuracy of the welding path, significantly reduce the welding defects and inconsistencies caused by path errors, and ensure the stability and uniformity of the weld quality.

[0016] 2. The present invention adopts a multi-objective optimization algorithm to optimize the welding path, which can minimize the path length, reduce unnecessary repeated welding operations, and shorten the operation time while maintaining the welding quality. This not only improves the welding efficiency and reduces working hours, but also effectively saves material and energy consumption.

[0017] 3. The present invention uses a real-time feedback and dynamic adjustment mechanism to enable the welding path to be adaptively adjusted according to the geometric deformation on site. By real-time monitoring of state changes during the welding process, the path can be corrected immediately to ensure the welding quality and accurate execution of the path, which has significant advantages, especially in complex and dynamically changing construction environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Please see attached Figure 1 -Attached Figure 2 The embodiment of the present invention provides an automated welding path optimization method based on BIM and point cloud data, which can optimize the welding path by efficiently integrating the BIM model and point cloud data, and realize dynamic adjustment in the actual welding process. The method can be widely used in industries such as construction, manufacturing, and aerospace, especially in complex structure welding scenarios, and can effectively improve welding accuracy, efficiency, and quality.

[0021] The specific implementation of the present invention includes the following main steps, and the principle, technical details and specific operation steps of each step are described in detail.

[0022] 1. Preprocessing and cleaning of point cloud data 1.1 Data Collection Point cloud data is collected through 3D measurement equipment such as laser scanners, lidar, and structured light scanners. Through these devices, we can obtain 3D spatial data of the on-site welding area, including surface shape, structural position and other information. Point cloud data may have problems such as noise, sparse areas, and measurement errors during collection, which will affect the planning and execution of subsequent welding paths. Therefore, point cloud data needs to be denoised, simplified, and reconstructed to ensure data quality.

[0023] 1.2 Noise Removal The noise in point cloud data usually comes from measurement errors or interference from environmental factors. To remove these noises, the statistical outlier removal (SOR) algorithm is used. This algorithm identifies and removes outliers that are far away by calculating the statistical information of other points in the neighborhood of each point. In this way, the valid geometric information in the point cloud data is retained. A Gaussian filter is further used to smooth the point cloud to eliminate high-frequency noise caused by sensor errors, reflection interference, etc., thereby improving the quality of the point cloud data.

[0024] 1.3 Point cloud simplification Since the acquisition of point cloud data usually generates millions of data points, directly processing this data will result in low computational efficiency. Therefore, the Voxel Grid filter is used to simplify the point cloud data. This method divides the point cloud space into multiple cubic voxels, and each voxel replaces the original point with a representative point (the center point of the voxel). The size of the voxel directly affects the degree of data simplification. The larger the voxel size, the lower the data accuracy and the higher the computational efficiency.

[0025] In order to further improve the computing efficiency and storage efficiency, the octree structure is used to store and query point cloud data. The octree structure recursively divides the space into 8 sub-areas, making it possible to operate and process large-scale point cloud data more efficiently.

[0026] 1.4 Surface reconstruction Since point cloud data may have sparse areas or missing surfaces, the Poisson Surface Reconstruction method is used for surface reconstruction. This method generates a continuous and smooth three-dimensional surface by smoothing the point cloud data, which can effectively fill the holes in the point cloud and restore the integrity of the surface structure. Through this process, it can be ensured that the planning of the welding path is based on complete geometric data.

[0027] 2. Registration and fusion of point cloud and BIM model 2.1 Registration Principle The BIM model is a design model before construction, usually used to describe idealized structural geometry data, while the point cloud data is the geometry information actually obtained from the site. Due to the differences between the two in terms of spatial coordinate system, scale, posture, etc., the point cloud and BIM model need to be registered. The purpose of registration is to align the point cloud data with the BIM model so that the two share the same geometry information in the same coordinate system, providing an accurate basis for path planning.

[0028] 2.2 Coarse Registration The initial registration process is to match the geometric features of the point cloud and BIM model. The FPFH (Fast Point Feature Histograms) feature descriptor is used to extract local geometric features from the point cloud and BIM model, and the RANSAC (Random Sample Consensus) algorithm is used to remove mismatched feature points, and then a rough registration is performed based on this. The rough registration provides an initial spatial relationship estimate for the subsequent fine registration.

[0029] 2.3 Fine Registration Based on the rough registration, the ICP (Iterative Closest Point) algorithm is used for fine registration. The ICP algorithm iteratively finds the nearest neighbor of each point and minimizes the distance between the point cloud and the BIM model to gradually optimize the registration results. In addition, the Levenberg-Marquardt algorithm is used for nonlinear optimization to improve the registration accuracy and eliminate possible registration errors.

[0030] 2.4 Error Correction and Geometric Alignment During the welding process, factors such as thermal deformation and material expansion can cause geometric deviations. The thermal stress simulation of the welding process is carried out through the finite element analysis (FEA) method to simulate the thermal effects generated during the welding process and predict the deformation of the welding area. Based on this deformation model, the geometric error between the BIM model and the point cloud data is corrected to ensure that the welding path optimization is based on accurate geometric data.

[0031] 3. Welding path planning and optimization 3.1 Path Planning Welding path planning needs to comprehensively consider factors such as the geometry of the welding area, joint location, welding sequence, robot accessibility, and obstacles. Preliminary path planning is performed using the A* algorithm or the Dijkstra algorithm. These algorithms search for paths and calculate the shortest or optimal path, which can take into account work efficiency and mechanical operability while ensuring path accuracy.

[0032] 3.2 Multi-objective optimization The optimization of welding path is not only the selection of the shortest path, but also requires the optimization of multiple objectives such as path length, welding time, energy consumption, etc. Particle swarm optimization (PSO) and genetic algorithm (GA) are used for multi-objective optimization of the path. Specifically, PSO finds the global optimal solution by simulating particles moving in the search space; GA optimizes the path by simulating natural selection and crossover mutation. These optimization algorithms consider multiple dimensions in welding path planning and ultimately generate the optimal path.

[0033] 3.3 Path Smoothing In order to improve the motion accuracy and stability of the welding robot, Bezier curve or spline interpolation method is used to smooth the welding path. These methods smooth the turning part of the path through interpolation calculation, thereby reducing the sharp turns and unstable parts in the path and ensuring the smooth progress of the welding process.

[0034] 4. Real-time feedback and dynamic adjustment 4.1 Path tracking and error detection The precise execution of the welding path directly affects the welding quality. In order to monitor the execution of the path in real time, a visual system such as LiDAR or RGB-D camera is used to obtain the welding progress in real time and compare it with the planned path. By monitoring the deviation between the actual path and the preset path, the Kalman filter algorithm is used to dynamically correct the path error and reduce the noise and instability in the path execution.

[0035] 4.2 Welding quality monitoring and feedback The quality of welding is closely related to the precise execution of the path. During the welding process, temperature sensors, stress sensors, etc. are used to monitor the temperature, stress distribution and weld shape of the welding area in real time. By real-time monitoring of various quality parameters in the welding process, adaptive control algorithms are used to dynamically adjust welding parameters such as welding current and speed to ensure that the welding quality always meets the predetermined standards.

[0036] 4.3 Dynamic Path Update During the welding process, as the welding area is gradually completed, the geometry of the site may change (for example, due to material deformation caused by thermal effects). The present invention uses an incremental ICP algorithm to perform real-time registration of the newly collected point cloud data and dynamically updates the welding path according to the actual deviation. In addition, combined with the multi-physics field coupling simulation method, the thermal effect and geometric deformation during the welding process are predicted to ensure that the welding path can be adjusted in real time to cope with the deformation.

[0037] 5. System integration and real-time control 5.1 System Integration The present invention integrates modules such as point cloud data acquisition, BIM model registration, welding path planning, real-time quality monitoring and feedback into a unified automation platform. The platform has multiple functions such as data collection, path planning, real-time monitoring, quality feedback and dynamic adjustment. Each module communicates and collaborates in real time through data interfaces and control instructions to ensure the smooth progress of the entire welding process.

[0038] 5.2 Real-time Control Through the integrated control system, the path execution, welding parameter adjustment, welding quality monitoring, etc. in the welding process can be fed back and adjusted in real time. The system dynamically adjusts the welding path and welding parameters based on the real-time feedback information, so that the welding operation is always kept in the optimal state, ensuring the accuracy and quality of the welding process. Example 1: Welding path optimization of complex steel structures Specific steps: Point cloud data collection Use a laser scanner to perform 3D scanning of complex steel structures and generate point cloud data.

[0039] The scanning resolution was set to 0.2 mm, and the point cloud data contained approximately 5 million points.

[0040] Point cloud preprocessing Statistical Outlier Removal (SOR) algorithm was used to remove 10% of the outliers.

[0041] Voxel Grid filtering was used to simplify the point cloud, and the voxel size was set to 5 mm.

[0042] Point cloud and BIM model registration Rough registration: Extract feature points of point cloud and BIM model through FPFH feature and perform preliminary registration.

[0043] Fine registration: The ICP algorithm is used to perform fine registration between the point cloud and the BIM model, with the final registration error <0.5 mm.

[0044] Path Planning The A* algorithm is used to calculate the welding path. The welding area is the connection area between the steel beam and the steel column, and obstacle avoidance and welding sequence are considered.

[0045] The preliminary path length is 150 m.

[0046] Path Optimization The path is optimized using the particle swarm optimization (PSO) algorithm with the goal of minimizing the path length and welding time.

[0047] After optimization, the path length was reduced to 135 m and the welding time was expected to be reduced by 10%.

[0048] Real-time feedback and dynamic path adjustment Real-time tracking of welding paths using an RGB-D camera.

[0049] The geometric deviation caused by thermal deformation during welding is monitored (with a deviation amplitude of 0.2 mm) and the path is adjusted dynamically.

[0050] Process parameters Welding current: 120 A Welding speed: 10 mm / s Shielding gas flow rate: 15 L / min Comparative Example 1: Welding path planning without using point cloud and BIM model registration Manually measure the size of the welding area and rely on experience to plan the welding path.

[0051] There is no point cloud data or BIM model reference, resulting in low path accuracy.

[0052] The path planning length is 160 m, and no optimization is performed, resulting in a long welding time.

[0053] The thermal deformation was not monitored in real time, and deviations of up to 2 mm were found after welding was completed.

[0054] Example 2: Pipeline welding path optimization Specific steps: Point cloud data collection Use a 3D scanner to collect point cloud data of the pipe joint area.

[0055] The resolution was set to 0.1 mm and the point cloud contained 2 million points.

[0056] Point cloud preprocessing Use a Gaussian filter to remove small noises.

[0057] The surface reconstruction uses the Poisson reconstruction algorithm to restore the complete surface of the pipe joint area.

[0058] Point cloud and BIM model registration Rough registration: Use SHOT features to extract key geometric points for preliminary alignment.

[0059] Fine registration: The Levenberg-Marquardt optimization algorithm is used to improve the registration accuracy, and the final error is controlled within 0.3 mm.

[0060] Path Planning The Dijkstra algorithm is used to calculate the welding path, and the total length of the path is 20 m.

[0061] Path Optimization Genetic algorithm is used to optimize the path, with the goal of minimizing welding time.

[0062] After optimization, the path length was reduced by 5% and the welding time was shortened by 8%.

[0063] Real-time feedback and adjustments Use RGB-D sensors to monitor the welding process in real time.

[0064] Combined with the thermal deformation compensation model, the welding parameters are dynamically adjusted to ensure the welding quality.

[0065] Process parameters Welding current: 90 A Welding speed: 8 mm / s Protective gas flow rate: 12 L / min Comparative Example 2: Using only the BIM model for welding path planning No point cloud data is collected, and path planning is done only based on the BIM model.

[0066] The geometric deviation of the actual pipeline is not considered, and the path planning length is 22 m.

[0067] Due to the large deformation of the actual pipeline, the welding joint deviated from the designed position with a deviation of up to 1.5 mm.

[0068] The welding quality is poor and some welds are defective.

[0069] Example 3: Optimization of welding path for large equipment Specific steps: Point cloud data collection Scanning large equipment welding areas with LiDAR.

[0070] The scanning range is 500 m², the point cloud resolution is 0.5 mm and the point cloud contains approximately 10 million points.

[0071] Point cloud preprocessing Use Octree partitioning to process point clouds and reduce memory usage.

[0072] The SOR algorithm is used to remove outliers with a removal rate of 5%.

[0073] Point cloud and BIM model registration Rough registration: Use the FPFH feature matching algorithm for preliminary registration.

[0074] Fine registration: ICP algorithm and finite element simulation are combined to consider the thermal effect during welding. Path planning The A* algorithm was used to generate a preliminary welding path with a path length of 300 m.

[0075] Path Optimization PSO was combined with Pareto frontier theory to optimize path length, welding time and energy consumption.

[0076] After optimization, the path length was reduced to 280 m and the welding time was shortened by 15%.

[0077] Real-time feedback and adjustments Monitor welding deviation in real time (deviation range 0.1 mm) and dynamically adjust welding path.

[0078] Adjust welding current and speed based on temperature sensor data.

[0079] Process parameters Welding current: 140 A Welding speed: 12 mm / s Shielding gas flow rate: 18 L / min Comparative Example 3: Large Equipment Welding Relying Only on Manual Path Planning No point cloud or BIM model was used and the welding path was planned entirely manually.

[0080] The path length is 320 m, the welding time is long, and there are repeated moves in the path planning.

[0081] Thermal deformation was not monitored and the welds showed deviations with a maximum deviation of 2.5 mm.

[0082] Comparative experiment Experimental design The performance difference between the automated welding path optimization based on the method of the present invention and the traditional welding path planning is compared. The experiment takes the welding of complex steel structures as an example, focusing on comparing the path length, welding time, and welding quality (deviation and defect rate).

[0083] Experimental results: Experimental groups Path length (m) Welding time (hours) Average deviation(mm) Defect rate (%) Example 1 (method of the present invention) 135 8.5 0.2 1 Comparative Example 1 (Traditional Manual Planning) 160 10.5 2 5 Example 2 (method of the present invention) 19 1.5 0.1 0.5 Comparative Example 2 (BIM model only) 22 2 1.5 3.5 Example 3 (method of the present invention) 280 14 0.1 0.8 Comparative Example 3 (Manual Planning) 320 17 2.5 6 According to the table data, the inventive method significantly optimized the welding path length (reduced by 10-15%), welding time (reduced by 15-20%), average deviation (reduced to 0.1-0.2 mm) and defect rate (reduced to less than 1%) in all embodiments. Compared with the traditional method, the inventive method significantly improves the welding efficiency and quality, reflecting the advantages of point cloud and BIM data fusion, path optimization, multi-objective optimization and real-time dynamic adjustment.

[0084] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The automated welding path optimization method based on BIM and point cloud data is characterized by: The following steps are involved: Obtaining point cloud data and BIM model data, wherein the point cloud data is obtained by laser scanning or other three-dimensional scanning technology, and the BIM model data is a design model before construction; Preprocessing the point cloud data, including denoising, simplification and surface reconstruction, to improve the quality and processing efficiency of the point cloud data; Registering the point cloud data and the BIM model data, aligning the point cloud data with the BIM model, and eliminating differences in spatial coordinate systems and scales; Based on the registered point cloud data and BIM model data, the welding path is planned, and the preliminary welding path is generated by considering the spatial constraints of the welding area, welding sequence, and joint position. Perform multi-objective optimization on the preliminary welding path, optimize the path length, energy consumption, and welding time targets, and generate the optimal welding path; Smoothing the welding path to reduce sharp turns and unstable parts of the path; Track the welding path in real time and perform error detection, and correct the path through sensor data and visual system feedback; Monitor welding quality and adjust welding parameters based on real-time feedback to ensure welding quality meets predetermined standards; During the welding process, the path and welding parameters are dynamically updated to adapt to the geometric deformation in the construction environment and the thermal deformation during the welding process.

2. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The preprocessing steps of the point cloud data include: Use statistical outlier removal algorithm to remove noise points in the point cloud; Use Gaussian filter to smooth the point cloud data and remove high-frequency noise; Use Voxel Grid filter to simplify point cloud data to reduce computational complexity.

3. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The registration steps of the point cloud data and the BIM model include: Based on the feature point extraction and matching method, the geometric features in the point cloud and BIM model are extracted and matched through the FPFH feature descriptor; Use the ICP algorithm for fine registration to minimize the geometric error between the point cloud and the BIM model; The registration results are further adjusted using the Levenberg-Marquardt optimization algorithm to improve accuracy.

4. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The welding path planning comprises the following steps: Use the A* algorithm or Dijkstra algorithm to calculate the preliminary welding path, taking into account the spatial limitations of the welding area and the welding sequence; The minimum spanning tree algorithm based on graph theory optimizes the welding path, making the path more concise and avoiding unnecessary repetition.

5. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The path optimization step comprises: Use particle swarm optimization or genetic algorithm to perform multi-objective optimization of the path, with the optimization objectives including path length, welding time, and energy consumption; Based on the Pareto frontier theory, each objective is balanced in the multi-objective optimization to ensure the multi-dimensional optimization of the welding path.

6. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The path smoothing step comprises: Use Bezier curve or spline interpolation method to smooth the welding path to reduce sharp turns and unstable parts in the path and ensure the smoothness of the welding process.

7. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The real-time tracking and error detection steps include: Use a visual perception system to obtain welding path data in real time through lidar or RGB-D camera; The Kalman filter algorithm is used to filter the welding path error and improve the stability of path tracking.

8. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The welding quality monitoring and feedback adjustment steps include: Use temperature sensors and stress sensors to monitor the temperature distribution, stress state and weld morphology generated during welding in real time; Based on real-time feedback data, an adaptive control algorithm is used to adjust the welding current and speed parameters to ensure that the welding quality meets the requirements.

9. The automatic welding path optimization method based on BIM and point cloud data according to claim 1 is characterized in that: The steps of dynamically updating the path and welding parameters include: The incremental ICP algorithm is used to perform real-time registration of the newly added point cloud data and adjust the welding path; Based on the multi-physics field coupling simulation model, the thermal effects and deformation during the welding process are simulated, and the welding path and welding parameters are dynamically adjusted.

10. The automatic welding path optimization method based on BIM and point cloud data according to claim 1, characterized in that: The real-time feedback and dynamic adjustment system includes: Integrate point cloud data acquisition, BIM model registration, welding path planning, real-time quality monitoring and feedback system into a unified platform to achieve closed-loop control; Through real-time feedback from sensor data and vision systems, welding paths and parameters are automatically adjusted to adapt to geometric changes and thermal deformation during construction.

Images