Gantry machine tool three-dimensional modeling device and method based on multi-source data fusion

Abstract

The invention discloses a gantry machine tool three-dimensional modeling device and method based on multi-source data fusion, and relates to the technical field of machine tool control, and the device comprises the steps: carrying out the data sensing of a gantry machine tool through a plurality of data sources, and obtaining a multi-source data set; traversing the multi-source data set for analysis processing, and constructing a plurality of three-dimensional data elements; performing feature extraction on the plurality of three-dimensional data elements, and determining a multi-source data registration result; fusing the plurality of three-dimensional data elements to construct a three-dimensional model of the gantry machine tool; and introducing a plurality of actual application scenes of the gantry machine tool, generating deviation data to dynamically compensate the three-dimensional model, and determining a three-dimensional optimization model of the gantry machine tool. According to the method, the technical problems of low modeling precision, low data fusion efficiency and lack of a dynamic compensation mechanism of a gantry machine tool in the prior art are solved, and the technical effects of improving the modeling precision, improving the data processing efficiency and realizing dynamic compensation and optimization are achieved.

Description

Technical Field

[0001] This invention relates to the field of machine tool control technology, and specifically to a 3D modeling device and method for gantry milling machines based on multi-source data fusion. Background Technology

[0002] Gantry milling machines, as large CNC machine tools, are widely used in the processing and manufacturing of various heavy machinery parts. They are characterized by their large processing range, robust structure, and high precision. With the increasing demand from modern industry for complex curved surfaces, high precision, and large workpieces, traditional machining methods are no longer sufficient to meet the requirements of efficient and precise machining. Therefore, how to improve the processing efficiency and precision of gantry milling machines through more advanced technologies has become an urgent technical problem to be solved in the current machinery manufacturing field. With the development of digital twin technology, the application of 3D modeling-based digital technology in CNC machine tools has gradually matured. However, traditional 3D modeling methods mostly rely on a single data source and cannot dynamically reflect the changing characteristics of the machine tool during actual machining in real time. This type of model, lacking the integration and processing of multiple data sources, results in low model accuracy and cannot effectively adapt to complex machining environments.

[0003] Existing technologies suffer from low modeling accuracy for gantry milling machines, low data fusion efficiency, and a lack of dynamic compensation mechanisms. Summary of the Invention

[0004] This application provides a 3D modeling device and method for gantry milling machines based on multi-source data fusion, which is used to address the technical problems of low modeling accuracy, low data fusion efficiency, and lack of dynamic compensation mechanism in existing technologies for gantry milling machines.

[0005] In view of the above problems, this application provides a 3D modeling device and method for gantry milling machines based on multi-source data fusion.

[0006] The first aspect of this application provides a 3D modeling device for a gantry milling machine based on multi-source data fusion, the device comprising: The system includes: a multi-source dataset acquisition module, which senses data from multiple data sources to obtain a multi-source dataset for the gantry milling machine; a 3D data element construction module, which traverses the multi-source dataset for analysis and processing to construct multiple 3D data elements; a multi-source data registration result determination module, which extracts features from the multiple 3D data elements, registers them according to the data element features, and determines the multi-source data registration result; a 3D model construction module, which fuses the multiple 3D data elements based on the multi-source data registration result to construct a 3D model of the gantry milling machine; and a 3D optimization model determination module, which incorporates multiple practical application scenarios of the gantry milling machine, combines these scenarios with the 3D model for test feedback, generates deviation data to dynamically compensate the 3D model, and determines the 3D optimization model of the gantry milling machine.

[0007] A second aspect of this application provides a method for 3D modeling of a gantry milling machine based on multi-source data fusion, the method comprising: Data sensing of the gantry milling machine is performed through multiple data sources to obtain multi-source datasets. These datasets are then analyzed and processed to construct multiple 3D data elements. Features are extracted from these 3D data elements, and they are registered according to these features to determine the multi-source data registration result. Based on the registration result, the multiple 3D data elements are fused to construct a 3D model of the gantry milling machine. Multiple practical application scenarios of the gantry milling machine are introduced, and test feedback is conducted based on these scenarios and the 3D model to generate deviation data for dynamic compensation of the 3D model, thus determining the optimized 3D model of the gantry milling machine.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: The system includes a multi-source dataset acquisition module for sensing data from multiple data sources to obtain a multi-source dataset for the gantry milling machine; a 3D data element construction module for traversing and analyzing the multi-source dataset to construct multiple 3D data elements; a multi-source data registration result determination module for extracting features from the multiple 3D data elements to determine the multi-source data registration result; a 3D model construction module for fusing the multiple 3D data elements based on the multi-source data registration result to construct a 3D model of the gantry milling machine; and a 3D optimization model determination module for incorporating multiple practical application scenarios of the gantry milling machine, generating deviation data to dynamically compensate the 3D model, and determining the 3D optimization model of the gantry milling machine. This achieves the technical effects of improving modeling accuracy, enhancing data processing efficiency, and realizing dynamic compensation and optimization. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of the structure of the 3D modeling device for a gantry milling machine based on multi-source data fusion provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the 3D modeling method for gantry milling machines based on multi-source data fusion provided in this application embodiment.

[0011] Figure labeling: 10 Multi-source dataset acquisition module, 20 3D data element construction module, 30 Multi-source data registration result determination module, 40 3D model construction module, 50 3D optimization model determination module. Detailed Implementation

[0012] This application provides a 3D modeling device and method for gantry milling machines based on multi-source data fusion, which addresses the technical problems of low modeling accuracy, low data fusion efficiency, and lack of dynamic compensation mechanism in existing technologies for gantry milling machines.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] Example 1, as Figure 1 As shown, this application provides a 3D modeling device for gantry milling machines based on multi-source data fusion, the device comprising: Multi-source dataset acquisition module 10 is used to perform data sensing on the gantry machine tool through multiple data sources to obtain multi-source datasets.

[0015] Specifically, this module plays a crucial role in data collection. It senses the gantry milling machine through multiple data sources to acquire comprehensive and rich multi-source datasets. These data sources include various types of sensors, measuring devices, and data acquisition systems. For example, there are devices specifically designed to scan the external structure of the gantry milling machine to obtain point cloud data; devices that acquire images of the machine from multiple angles to obtain multiple image data sets; and devices that can probe the internal structure of the machine to determine multiple depth data points. By integrating this data from different sources, the module provides the foundational material for subsequent 3D modeling.

[0016] The three-dimensional data element construction module 20 is used to traverse the multi-source dataset for analysis and processing, and construct multiple three-dimensional data elements.

[0017] Specifically, the 3D data element construction module 20 is a key step in the 3D modeling process of the gantry milling machine. This module first traverses the multi-source dataset, performing in-depth analysis and processing on different types of data during the traversal. For portions containing large amounts of point cloud data, noise reduction processing is performed to remove noise points caused by measurement errors, environmental interference, and other factors, in order to construct a type of 3D data element, namely 3D point matrix parameters. These 3D point matrix parameters can accurately describe the external geometry of the gantry milling machine, providing basic spatial structural information for subsequent 3D modeling. For multiple image data, conversion processing is performed, using image enhancement technology to improve image quality and transforming them into 3D image parameters that better reflect the characteristics of the gantry milling machine. These parameters can provide the machine tool's appearance details and texture information. For multiple depth data, correction processing is performed to ensure the accuracy of the depth data, thereby constructing three types of 3D data elements, namely 3D spatial parameters. These parameters help to accurately grasp the internal spatial layout and depth information of the gantry milling machine. Finally, these three different types of 3D data elements are integrated to form multiple complete 3D data elements, fully preparing for the construction of the 3D model of the gantry milling machine.

[0018] The multi-source data registration result determination module 30 is used to extract features from the multiple three-dimensional data elements, register the multiple three-dimensional data elements according to the data element features, and determine the multi-source data registration result.

[0019] Specifically, the multi-source data registration result determination module 30 plays a crucial connecting role in the 3D modeling of the gantry milling machine. This module first extracts features from multiple 3D data elements, mining features that represent their unique properties from different types of 3D data elements. For 3D point cloud parameters constructed from point cloud data, specific geometric shape features are extracted; for 3D image parameters, visual features such as color and texture are extracted; for 3D spatial parameters, depth and spatial layout features are extracted. Next, the multiple 3D data elements are registered according to these data element features. During the registration process, by comparing the features of different data elements, the correspondence between them is found, and their positions and orientations in space are adjusted so that different types of data can be accurately aligned in space. For example, it ensures that the external structure represented by the point cloud data matches the appearance presented by the image data and the internal space reflected by the depth data. After such feature extraction and registration, the multi-source data registration result is finally determined. This result provides an accurate spatial position and correspondence basis for the subsequent fusion of multiple 3D data elements to construct the 3D model of the gantry milling machine, ensuring the accuracy and integrity of the model.

[0020] The three-dimensional model construction module 40 is used to fuse the multiple three-dimensional data elements according to the multi-source data registration results to construct a three-dimensional model of the gantry machine tool.

[0021] Specifically, the 3D model construction module 40 is the core component for realizing the 3D modeling of the gantry milling machine. This module fuses multiple 3D data elements based on the multi-source data registration results. First, an impact analysis is performed on the multiple 3D data elements according to the multi-source data registration results. This means assessing the importance and scope of influence of each 3D data element in constructing the overall 3D model. For example, the contribution of 3D raster parameters to describing the external geometry of the gantry milling machine, the role of 3D image parameters in presenting appearance details, and the importance of 3D spatial parameters in reflecting the internal structure are analyzed. Based on the analysis results, the initial weights of the multiple 3D data elements are determined, and a data dictionary is constructed for subsequent management and adjustment of the weights. Next, based on the data dictionary, the initial weights of three types of 3D data elements (3D raster parameters), three types of 3D data elements (3D image parameters), and three types of 3D data elements (3D spatial parameters) are updated, generating multiple more accurate and reasonable weight coefficients. These weight coefficients can better reflect the relative importance of different data elements in the fusion process. Finally, the 3D raster parameters, 3D image parameters, and 3D spatial parameters are weighted and fused according to these weight coefficients. By combining different types of data elements according to their weights, the advantages of each are fully utilized while compensating for the shortcomings of single data types. Based on the fusion results, these parameters are reconstructed to further optimize the model's structure and details. Through this fusion and reconstruction process, a 3D model of the gantry milling machine is constructed. This 3D model integrates information from multiple data sources, accurately presenting the gantry milling machine's appearance, internal structure, and overall spatial form, providing an intuitive and effective tool for the design, analysis, and optimization of gantry milling machines.

[0022] The 3D optimization model determination module 50 is used to introduce multiple actual application scenarios of the gantry milling machine, and to conduct test feedback based on the multiple actual application scenarios and the 3D model to generate deviation data to dynamically compensate the 3D model and determine the 3D optimization model of the gantry milling machine.

[0023] Specifically, the 3D optimization model determination module 50 plays a crucial role in the 3D modeling of gantry milling machines, aiming to generate more accurate and practical 3D optimization models. This module first introduces several practical application scenarios of gantry milling machines. Among them, the machining path planning scenario simulates the machining path of the gantry milling machine based on the 3D model. This simulation can evaluate the accuracy of path planning, ensuring that the tool can accurately cut along the predetermined path during machining, and also assess the ability to avoid interference, preventing collisions between the tool and machine tool components or workpieces. In the load distribution analysis scenario, when machining different types of workpieces, the 3D model is used to analyze the structural deformation of the gantry milling machine under different loads. This helps to understand the stress state of the machine tool during actual operation, providing a basis for optimizing the machine tool's structural design. The thermal deformation monitoring scenario monitors the thermal deformation of the gantry milling machine under long-term working conditions using sensors, and combines this with the 3D model to evaluate the stability of the equipment. Understanding the impact of thermal deformation on machine tool accuracy allows for corresponding measures to reduce thermal deformation and improve machining accuracy. The vibration and dynamic performance evaluation scenario mainly analyzes the vibration characteristics of the gantry milling machine during high-speed machining or rapid movement. Optimizing the support structure and component layout using a 3D model can reduce vibration and improve the dynamic performance and machining quality of the machine tool. Next, testing and feedback are conducted based on these practical application scenarios and the constructed 3D model. In these scenarios, deviations between the 3D model and actual conditions are identified by comparing the model with actual measurement data or simulation results. Deviation data is then generated, reflecting the shortcomings of the 3D model under different practical application scenarios. Finally, the deviation data is used to dynamically compensate for the 3D model. By adjusting the model's parameters, structure, or weights, the model is made closer to the actual situation. After multiple rounds of testing, feedback, and dynamic compensation, the optimized 3D model of the gantry milling machine is determined. This optimized model better adapts to the needs of practical application scenarios, providing a more accurate and reliable reference for the design, manufacturing, debugging, and operation of the gantry milling machine.

[0024] In one possible implementation, the multi-source dataset acquisition module 10 further includes: Multiple data sources are traversed to identify multiple data sensing devices, including a first data sensing device, a second data sensing device, and a third data sensing device. The external structure of the gantry milling machine is scanned using the first data sensing device to determine multiple point cloud data. The gantry milling machine is then scanned from multiple angles using the second data sensing device to determine multiple image data. The internal structure of the gantry milling machine is scanned using the third data sensing device to determine multiple depth data. The multiple point cloud data, the multiple image data, and the multiple depth data are then added to the multi-source dataset.

[0025] Specifically, by traversing multiple data sources, several key data sensing devices were identified. The first data sensing device is a laser scanner. A laser scanner emits a laser beam and receives the reflected signal, accurately acquiring 3D point cloud data of the gantry milling machine's external structure by measuring the round-trip time or phase difference of the laser. It offers advantages such as high precision, high speed, and non-contact measurement, enabling it to quickly and accurately capture the complex external geometry of the gantry milling machine, providing detailed spatial information for subsequent 3D modeling. The second data sensing device is an RGB camera or multimodal camera. An RGB camera can capture color images of the gantry milling machine, providing rich visual information, including the machine's color, texture, and surface details. A multimodal camera can combine different imaging modes, such as infrared and ultraviolet, to acquire more feature information about the gantry milling machine. This image data can complement the point cloud data and depth data, further enhancing the accuracy and realism of the 3D modeling. The third data sensing device is a depth sensor. A depth sensor measures the distance between an object and the sensor, acquiring depth information of the gantry milling machine's internal structure. It helps determine the internal spatial layout, component positions, and depth relationships of a machine tool. Unlike laser scanners and cameras, depth sensors focus on providing information in the depth dimension, which is crucial for understanding the internal structure and spatial characteristics of a gantry milling machine. By identifying these three different types of data sensing devices, relevant data about the gantry milling machine can be collected from multiple angles and dimensions, laying a solid foundation for building a comprehensive and accurate 3D model of the gantry milling machine.

[0026] When a primary data sensing device (such as a laser scanner) scans the external structure of a gantry milling machine, it emits a specific signal (such as a laser beam) and receives the signal reflected from the machine's surface. By measuring parameters such as signal propagation time, phase difference, or intensity, the distance between each point along the signal propagation path and the scanner can be determined. As the scanner scans the gantry milling machine from different positions and angles, it gradually accumulates a large amount of point coordinate information, which together form multiple point cloud datasets. Each point in the point cloud dataset contains coordinate values ​​in three-dimensional space (usually coordinates in the x, y, and z directions), as well as possible other attribute information, such as reflection intensity. This point cloud data can very accurately reflect the shape and details of the gantry milling machine's external structure. For example, it can clearly present the machine's various surfaces, edges, corners, and other features. Through further analysis and processing of the point cloud data, a three-dimensional model of the gantry milling machine's external structure can be constructed, providing important basic data for subsequent design, inspection, and maintenance.

[0027] Secondary data sensing devices (such as RGB cameras or multimodal cameras) acquire comprehensive image data from multiple angles when collecting data on a gantry milling machine. Shooting from different angles covers all sides and details of the machine. RGB cameras capture color and texture information within the visible light range, revealing the machine's appearance, surface condition, and potential markings and decorations. Multimodal cameras utilize different imaging modes; for example, infrared imaging detects temperature distribution on the machine surface, while ultraviolet imaging reveals fluorescence reactions in specific materials, providing richer information. This multi-angle image data provides crucial visual references for the 3D modeling and analysis of the gantry milling machine. Processing this image data allows for the extraction of the machine's contours and texture features, which, combined with point cloud and depth data, enables a more accurate 3D model of the gantry milling machine. Furthermore, the image data can be used for machine appearance inspection and fault diagnosis; for instance, comparing images taken at different times can reveal wear, deformation, or damage on the machine surface.

[0028] Third-party data sensing devices (such as depth sensors) focus on determining multiple depth data points, especially the spatial position information of internal components, when collecting data on the internal structure of a gantry milling machine. Utilizing specific technologies, depth sensors can effectively penetrate the outer shell and other obstructions of the gantry milling machine to accurately measure the distance between the sensor and different internal locations. This depth data is crucial for the complex internal structure of a gantry milling machine. By performing a comprehensive scan of the gantry milling machine's interior, the spatial position of each internal component can be gradually determined. For example, the specific coordinates of different gears, shafts, and other components in the transmission device in three-dimensional space can be determined, the installation position and depth relationship of the guide rails can be understood, and the relative positions of various components in the electrical control system can be determined. This depth data provides a key basis for a deeper understanding of the internal structure of the gantry milling machine. In subsequent analysis and modeling processes, this data can be used to accurately construct a three-dimensional model of the gantry milling machine's interior, which is helpful for structural optimization design, fault diagnosis, and maintenance planning. For example, in fault diagnosis, depth data can be used to quickly locate the internal components that may have problems, thereby improving maintenance efficiency; in structural optimization design, the spatial layout of internal components can be analyzed to reasonably adjust the design scheme, thereby improving the performance and reliability of the machine tool.

[0029] Finally, multiple point cloud data, multiple image data, and multiple depth data obtained through these three data sensing devices were added to the multi-source dataset. This created a multi-source dataset containing various types of data, providing a rich data foundation for subsequent steps such as 3D metadata construction, data registration, 3D model construction, and optimization.

[0030] In one possible implementation, the three-dimensional data element construction module 20 further includes: The multi-source dataset is preprocessed to obtain a multi-source processed dataset, which includes an outlier set, an enhanced image set, and a depth data correction set. Noise is removed based on the multiple point cloud data and the outlier set to construct a first-class 3D data element, which is a 3D point raster parameter. The multiple image data are then transformed based on the enhanced image set to construct a second-class 3D data element, which is a 3D image parameter. The multiple depth data are then corrected based on the depth data correction set to construct a third-class 3D data element, which is a 3D spatial parameter. Finally, the first-class, second-class, and third-class 3D data elements are integrated to construct the multiple 3D data elements.

[0031] Specifically, preprocessing the multi-source dataset is a crucial step in the 3D modeling of gantry milling machines. Preprocessing yields a multi-source processed dataset, which includes outlier sets, augmented image sets, and depth data correction sets.

[0032] For multiple point cloud datasets, denoising is performed by combining outlier point sets. Point cloud data is affected by various factors during acquisition, resulting in outliers that can impact the accuracy of subsequent modeling. By removing outliers, a class of 3D data elements, namely 3D point lattice parameters, can be constructed. These 3D point lattice parameters can accurately describe the external geometry of the gantry milling machine, providing fundamental spatial structural information for subsequent 3D modeling.

[0033] For multiple image datasets, an enhanced image set is used for transformation. Image data may have quality issues such as insufficient contrast or blurriness. Processing the image data with an enhanced image set improves image quality and clarity, and then converts it into two types of 3D data elements, namely 3D image parameters. These 3D image parameters can provide details of the machine tool's appearance and texture, enriching the visual effect of the 3D model.

[0034] For multiple depth data sets, a depth data correction set is used for correction. Depth data may also contain errors during acquisition; correcting these errors using the depth data correction set allows for the construction of three types of 3D data elements, namely 3D spatial parameters. These 3D spatial parameters help to accurately grasp the spatial layout and depth information inside the gantry milling machine.

[0035] Finally, three types of 3D data elements (3D point matrix parameters, 3D image parameters, and 3D spatial parameters) are integrated. By integrating different types of 3D data elements, multiple complete 3D data elements can be constructed, fully preparing for building the 3D model of the gantry milling machine. These 3D data elements combine the advantages of point cloud data, image data, and depth data, enabling a more comprehensive and accurate description of the characteristics and structure of the gantry milling machine.

[0036] In one possible implementation, the three-dimensional data element construction module 20 further includes: Based on the multi-source dataset, a sensing time series analysis is performed to generate a time series; noise identification is performed based on the multiple point cloud data and the time series to determine the outlier set; wavelet transform is performed based on the multiple image data and the time series to determine the enhanced image set; error analysis is performed based on the multiple depth data and the time series to determine the depth data correction set; the outlier set, the enhanced image set, and the depth data correction set are added to the multi-source processing dataset.

[0037] Specifically, in the 3D modeling of gantry milling machines, sensor time-series analysis based on multi-source datasets is a crucial step. Multi-source datasets contain data from different data sources, such as point cloud data, image data, and depth data. By performing sensor time-series analysis on this data, the temporal order of data acquisition can be determined, generating a time series. The time series reflects the changes in data at different points in time. For the 3D modeling of gantry milling machines, time series analysis helps understand the changes in the machine tool under different operating conditions and the trends of data from different data sources over time. For example, by analyzing the time series, the acquisition status of point cloud data, image data, and depth data at different points in time, as well as the temporal relationships between these data, can be determined. The process of generating a time series typically involves extracting and organizing timestamps from the data in the multi-source dataset. Timestamps record the specific time of data acquisition; by sorting and analyzing the timestamps, the time series of the data can be obtained. Furthermore, time series analysis methods, such as trend analysis and periodic analysis, can be used to predict and analyze the changing trends of the data, providing more accurate and reliable data support for the 3D modeling of gantry milling machines.

[0038] In 3D modeling of gantry milling machines, the process of identifying noise and determining outlier sets by combining multiple point cloud data with time series data is as follows: First, it is clear that the time series reflects the order of data acquisition and the temporal changes. Each point in the multiple point cloud data has its corresponding acquisition timestamp. The point cloud data is arranged chronologically according to the time series. Next, the point cloud data at each time point is analyzed. The spatial distance between each point and its surrounding neighboring points is calculated. A reasonable distance threshold is set. If the distance between a point and most of its surrounding points exceeds this threshold, then this point is initially marked as a possible noise point. Then, over time, the changes of the points marked as possible noise points are observed in the time series. If a point is marked as a possible noise point at multiple consecutive time points, and its positional change is significantly different from the overall trend of the surrounding normal points, then this point can be identified as an outlier. For example, under normal circumstances, the shape change of a certain part of a gantry milling machine is relatively continuous and regular over a period of time. If the positional change of a certain point during this period is very abrupt and inconsistent with the trend of the surrounding points, then this point is an outlier caused by noise. Collecting all identified outliers forms an outlier set. These outliers will be further processed in subsequent steps to improve the quality of the point cloud data, ensuring accurate construction of 3D point matrix parameters and the 3D model of the gantry milling machine.

[0039] In the 3D modeling of a gantry milling machine, an enhanced image set is determined based on wavelet transform using multiple image data combined with time series data. First, multiple image data are arranged sequentially according to the time series, and wavelet transform is used to perform multi-scale analysis on the image at each time point. Wavelet transform decomposes the image into sub-bands of different frequencies. The low-frequency sub-band covers the main contours and overall information of the image, and histogram equalization enhances contrast and brightness; the high-frequency sub-band contains details and edge information, and edge enhancement algorithms are used to highlight edge and texture features. Considering the time series factor, if the image changes little at consecutive time points, a conservative enhancement method is used to avoid distortion; if there are significant changes, corresponding enhancement processing is performed according to its characteristics to reflect the dynamic changes of the gantry milling machine. After processing the images at each time point, all optimized images are combined into an enhanced image set, providing higher-quality image data for constructing 3D image parameters and the 3D model of the gantry milling machine.

[0040] In the 3D modeling of gantry milling machines, it is crucial to determine the depth data correction set by combining multiple depth data with time series analysis. Firstly, depth data may be affected by various factors during acquisition, such as sensor accuracy and environmental interference, resulting in inherent errors. Introducing time series analysis provides a basis for analyzing the trends in these depth data. Multiple depth data points are arranged according to the time series, and a preliminary evaluation of the depth data at each time point is performed. By comparing the differences in depth data at adjacent time points, abnormal fluctuations can be identified. Simultaneously, the rationality of the depth data is analyzed in conjunction with the actual working state and motion patterns of the gantry milling machine. For depth data points exhibiting significant deviations, the causes of the errors are further analyzed. These errors may be due to factors such as momentary sensor malfunctions or object occlusion. Statistical methods and data fitting techniques are used to correct these abnormal data points. After error analysis and correction of the depth data at each time point, all corrected depth data are combined into a depth data correction set. This correction set provides a reliable depth data foundation for constructing accurate 3D spatial parameters and a high-quality 3D model of the gantry milling machine.

[0041] Finally, the outlier set, the enhanced image set, and the depth data correction set were added to the multi-source processing dataset. This makes the multi-source processing dataset richer and more accurate, providing a more reliable foundation for subsequently constructing different types of 3D data elements and the final 3D model of the gantry milling machine.

[0042] In one possible implementation, the 3D model building module 40 further includes: An impact analysis is performed on the multiple 3D data elements according to the multi-source data registration results. Based on the analysis results, the initial weights of the multiple 3D data elements are determined, and a data dictionary is constructed. Based on the data dictionary, the initial weights of the first-class, second-class, and third-class 3D data elements are updated to generate multiple weight coefficients. The 3D point matrix parameters, 3D image parameters, and 3D spatial parameters are weighted and fused according to the multiple weight coefficients. Based on the fusion results, the 3D point matrix parameters, 3D image parameters, and 3D spatial parameters are reconstructed to construct the 3D model of the gantry milling machine.

[0043] Specifically, in the process of 3D modeling of gantry milling machines, a crucial step is to conduct an impact analysis on multiple 3D data elements based on the multi-source data registration results. The multi-source data registration results provide spatial correspondences and references for different types of 3D data elements. First, each 3D data element is evaluated individually. For the first type of 3D data element, namely 3D raster parameters, their accuracy and completeness in describing the external geometry of the gantry milling machine are considered. If the point cloud data acquisition is comprehensive and accurate, its influence on the 3D model construction may be significant. For the second type of 3D data element, namely 3D image parameters, the degree to which the provided appearance details and texture information enhance the model's realism is analyzed. Vivid colors and clear textures in image data can add more visual effects to the model. For the third type of 3D data element, namely 3D spatial parameters, their importance in reflecting the internal structure of the gantry milling machine is assessed. Accurate depth data can help determine the position and spatial relationships of internal components. Based on the analysis results, the initial weights of the multiple 3D data elements are determined. These weights reflect the relative importance of each 3D data element in the overall model. For example, if the 3D raster parameters are crucial to building the basic framework of the model, they are assigned a higher weight; conversely, if some image data is of poor quality or contributes relatively little to the model, its weight will be reduced accordingly. Simultaneously, a data dictionary is constructed. This dictionary records in detail the type, description, initial weight, and related attribute information of each 3D data element. Such a dictionary structure facilitates subsequent weight adjustments and management, and provides a clear reference for the model construction and optimization process. By conducting influence analysis on multiple 3D data elements and constructing a data dictionary, a solid foundation can be laid for subsequent weight updates and model construction.

[0044] In 3D modeling of gantry milling machines, updating the weights of different types of 3D data elements based on the data dictionary is a crucial step. First, initial weight information for three types of 3D data elements (3D point cloud parameters), three types (3D image parameters), and three types (3D spatial parameters) is extracted from the data dictionary. For the first type of 3D data elements, their performance in describing the external geometry of the gantry milling machine is analyzed. If the point cloud data shows high accuracy in certain key areas and plays an important role in constructing the overall external structure of the model, its weight can be gradually increased. For example, comparing the point cloud density and accuracy of different regions, if the point cloud data in a certain region is very dense and can clearly delineate complex shapes, the weight of the corresponding 3D point cloud parameter in that region can be considered for increase. For the second type of 3D data elements, the quality of the image data and its contribution to the details of the model's appearance are carefully evaluated. If certain image data has vibrant colors and clear textures, significantly improving the realism of the model, its weight can be appropriately increased. For example, observing whether there are unique identifiers or features in the image; if these identifiers or features are very important for the identification and differentiation of the gantry milling machine, the weight of the corresponding image parameter can be increased. For the three types of 3D data elements, a thorough analysis of the value of depth data in revealing the internal structure of the gantry milling machine is conducted. If the accuracy of the internal structure is crucial in a specific application scenario, its weight can be increased. For example, checking whether the depth data can accurately reflect the positional relationships and spatial layout of internal components; if the depth data performs well in this regard, the weight of the 3D spatial parameters can be increased. By analyzing and comparing each type of 3D data element in this way, the weight values ​​are gradually adjusted using an iterative approach. First, the weight of one type of data element is adjusted, and then its impact on the overall model is observed before adjusting other types of data elements accordingly. Ultimately, multiple more accurate and reasonable weight coefficients are generated. These weight coefficients can better reflect the relative importance of different 3D data elements in the current modeling task, providing a reliable basis for subsequent weighted fusion and model reconstruction.

[0045] In constructing the 3D model of a gantry milling machine, a crucial step is the weighted fusion of 3D point parameters, 3D image parameters, and 3D spatial parameters using multiple weighting coefficients. First, based on the generated weighting coefficients, the 3D point parameters, 3D image parameters, and 3D spatial parameters are processed separately. For the 3D point parameters, which primarily reflect the external geometry of the gantry milling machine, the coordinates of each point in 3D space are adjusted according to its corresponding weighting coefficients, allowing important points to play a greater role in the fusion process. This ensures a more accurate and clearer external contour of the gantry milling machine. For the 3D image parameters, which provide rich appearance details and texture information, the color, brightness, and contrast attributes of the image are adjusted using weighting coefficients to better integrate them with other parameters. Simultaneously, the texture information in the image is mapped onto the surface of the 3D model, enhancing the model's realism. For the 3D spatial parameters, the internal spatial layout and depth information of the gantry milling machine are accurately captured. Weighting coefficients are used to optimize the representation of the internal structure, ensuring a more rational position and spatial relationship between various components. During the weighted fusion process, the three parameters are combined according to their respective weights. For example, for a specific region, different weights are assigned to different parameters based on its characteristics. If the region needs to emphasize its external shape, the weight of the 3D raster parameters will be relatively high; if the surface texture needs to be highlighted, the weight of the 3D image parameters will increase; if the focus is on the internal structure, the weight of the 3D spatial parameters will be even greater. Based on the fusion results, the 3D raster parameters, 3D image parameters, and 3D spatial parameters are reconstructed. This reconstruction process further optimizes the model's structure and details. By continuously adjusting the parameters and weights, the 3D model of the gantry milling machine becomes more accurate, complete, and realistic. Ultimately, a 3D model of the gantry milling machine was successfully constructed, providing a powerful tool for its design, analysis, and optimization.

[0046] In one possible implementation, the three-dimensional optimization model determination module 50 further includes: The three-dimensional model is simulated and tested across multiple real-world application scenarios to generate multiple simulation test results. Based on the multiple simulation test results, deviations are calculated for the multiple real-world application scenarios to generate initial deviation data. The initial deviation data is matched with the multiple real-world application scenarios, and a comprehensive analysis is performed based on the matching results to construct a deviation distribution map. Cluster analysis is performed on the initial deviation data according to the deviation distribution map to determine multiple deviation types. Feedback is then performed based on the multiple deviation types combined with the initial deviation data to obtain the deviation data.

[0047] Specifically, in the evaluation phase of the 3D model of the gantry milling machine, comprehensive simulation testing across multiple real-world application scenarios is crucial. The first step is machining path planning testing. The 3D model is used to simulate the machining paths for different workpieces. During this process, the model simulates the tool's movement trajectory on the workpiece and plans the optimal machining path. Then, actual machining is performed, closely monitoring path deviation information. For example, path errors may arise due to differences between model calculations and actual machine tool motion control. If the actual tool path deviates from the model-planned path, the specific error value is recorded. Simultaneously, workpiece boundary interference must be monitored. If the tool exceeds the workpiece boundary during machining, this is also considered deviation data and recorded. Next, load analysis testing is conducted. The machine tool is run under different load conditions, and stress and strain data are collected in real time using various sensors. This data reflects the machine tool's structural response under different loads. The structural deformation of the machine tool is evaluated using the 3D model, and the actual collected structural deformation data is compared with the model's predicted deformation, recording the deviation between the two. For example, if the model predicts a certain deformation of the machine tool under a certain load, but the actual measured deformation differs, this difference will be recorded as deviation data. In thermal deformation monitoring tests, infrared sensors monitor the machine tool's temperature changes. Temperature changes cause thermal deformation of the machine tool structure, and the difference between the actual and predicted thermal deformation data is analyzed using the 3D model. If the model's predicted thermal deformation is inconsistent with the actual monitored thermal deformation, the thermal deformation deviation is recorded, including information such as the direction and magnitude of the deformation. Finally, vibration testing is conducted. Under high-speed operation or complex machining conditions, accelerometers or vibration sensors are used to monitor the machine tool's vibration frequency and amplitude. The actual monitored vibration data is compared with the dynamic simulation results in the 3D model, and vibration deviations are recorded. For example, if the model's predicted vibration frequency and amplitude differ from the actual measured values, this difference will be recorded as a vibration deviation. Through these multiple simulation tests, multiple simulation test results are generated, which will provide important basis for subsequent analysis and optimization of the 3D model.

[0048] In the 3D model evaluation of gantry milling machines, calculating deviations for multiple practical application scenarios based on simulation test results is a crucial step. Results obtained from various simulation tests, including machining path planning, load analysis, thermal deformation monitoring, and vibration testing, are compared with the expected states in each practical application scenario. For machining path planning tests, the error between the actual machining path and the model-planned path is calculated, including positional and angular deviations, while also analyzing the degree of workpiece boundary interference. In load analysis tests, the actual collected stress and strain data are compared with the structural deformation predicted by the 3D model to determine the deviation in structural deformation. For thermal deformation monitoring tests, the difference between the actual thermal deformation data and the model-predicted thermal deformation is calculated, covering aspects such as the magnitude and direction of deformation. In vibration tests, the actual monitored vibration frequency and amplitude are compared with the dynamic simulation results in the model to determine vibration deviations. By comprehensively analyzing these comparisons, a complete deviation calculation is performed for multiple practical application scenarios, generating initial deviation data. This provides the foundation for subsequent deviation distribution mapping, deviation type identification, and feedback optimization.

[0049] In the optimization process of the 3D model of a gantry milling machine, matching the initial deviation data with multiple actual application scenarios and constructing a deviation distribution map is a crucial step. First, the initial deviation data is categorized and organized according to different actual application scenarios. Each scenario has unique working conditions and requirements, resulting in deviation data with specific characteristics. By matching the initial deviation data with the corresponding scenarios, the deviation situation in each scenario can be understood more accurately. Next, a comprehensive analysis is performed based on the matching results. The deviation data generated from multiple application scenarios are processed uniformly; the deviation of each scenario cannot be viewed in isolation but must be considered holistically. For example, when calculating the comprehensive deviation of each scenario, the deviations from load analysis and thermal deformation monitoring are combined. The structural deformation deviation obtained from load analysis and the thermal deformation deviation obtained from thermal deformation monitoring will influence each other. Under complex working conditions, these deviations work together to form the overall deviation distribution of the machine tool. To more intuitively display the deviation situation, a deviation distribution map is constructed. The deviation distribution map can graphically present the magnitude, direction, and distribution pattern of deviations under different scenarios, using different colors or symbols to represent different types of deviations, such as red for larger positive deviations and blue for larger negative deviations. The deviation distribution map clearly shows the concentrated areas of deviation and weak points of the gantry milling machine in various practical application scenarios. This constructed deviation distribution map provides an important reference for subsequent cluster analysis of the initial deviation data and determination of deviation types, and also provides a clear direction for further optimization of the gantry milling machine's 3D model.

[0050] In the process of perfecting the 3D model of the gantry milling machine, cluster analysis of the initial deviation data according to the deviation distribution map plays a crucial role. First, observing the deviation distribution map, which visually displays the deviation distribution of the gantry milling machine under different practical application scenarios, allows for the classification of the initial deviation data based on different characteristics and manifestations. For geometric deviations, the focus is primarily on deviations in the shape and dimensions of the machine tool. This includes workpiece shape deviations caused by path errors during machining path planning tests, or geometric shape changes caused by structural deformation during load analysis tests. By analyzing the deviation areas related to geometric shape in the deviation distribution map, these data can be categorized as geometric deviations. Thermal deformation deviations originate from thermal deformation monitoring tests. When the machine tool's temperature changes during operation, thermal deformation occurs. The deviation areas related to temperature changes shown in the deviation distribution map can be identified as thermal deformation deviations, which affect the machining accuracy and stability of the machine tool. Vibration deviations come from vibration tests. Under high-speed operation or complex machining conditions, machine tool vibration can lead to deviations. The areas in the deviation distribution map related to vibration frequency and amplitude can be identified as vibration deviations. Vibration deviations affect the surface quality and machining accuracy of the machine tool. By performing cluster analysis on the initial deviation data, multiple deviation types were identified. These types provide clear directions for further analysis and resolution of problems encountered in the practical application of gantry milling machines. Corresponding optimization and improvement measures can be taken for different deviation types to enhance the accuracy and reliability of the gantry milling machine's 3D model.

[0051] Finally, based on multiple deviation types and the initial deviation data, feedback is provided to the construction and optimization process of the 3D model. The model is adjusted and improved according to the feedback information to improve its accuracy and reliability in different practical application scenarios. Ultimately, more accurate deviation data is obtained, providing a strong basis for further optimization of the gantry machine tool 3D model.

[0052] In one possible implementation, the three-dimensional optimization model determination module 50 further includes: Based on the aforementioned multiple practical application scenarios, features are extracted from the 3D model of the gantry milling machine to obtain multiple feature parameters, including geometric feature parameters, material feature parameters, and structural feature parameters. A compensation and adjustment parameter set is then established based on these geometric, material, and structural feature parameters. A feedback control mechanism is constructed to synchronize the deviation data to the feedback control mechanism. The 3D model is then iteratively tested through the feedback control mechanism to determine the iterative test results. Based on the iterative test results and the compensation and adjustment parameter set, the 3D model is validated according to the aforementioned multiple practical application scenarios, generating a validation token. Finally, the optimized 3D model is generated based on the validation token.

[0053] Specifically, in the optimization of the 3D model of a gantry milling machine, feature extraction based on multiple practical application scenarios is a crucial step. Geometric feature parameters are primarily obtained through corresponding point cloud data. Point cloud data accurately reflects the external geometry of the gantry milling machine, including the dimensions, angles, and curvature of various components. By analyzing point cloud data from different practical application scenarios, the geometric feature parameters of the gantry milling machine under different working states can be extracted. For example, when machining large workpieces, some parts of the gantry milling machine may undergo slight deformation. By analyzing point cloud data, these deformations can be accurately captured, thereby determining the corresponding changes in geometric feature parameters. Material feature parameters correspond to image data, especially enhanced images. Enhanced images have clearer textures and edges, which can help in material identification. Different materials exhibit different textures, colors, and glosses in images. By analyzing image data from multiple practical application scenarios, the feature parameters of the materials used in the gantry milling machine can be extracted. For example, certain materials exhibit unique color changes under specific lighting conditions. By analyzing image data, the characteristic parameters of these materials can be determined, thereby providing a better understanding of the material performance of gantry milling machines under different working environments. Structural characteristic parameters correspond to depth data. Depth data accurately reflects the internal structure and spatial layout of gantry milling machines. By analyzing depth data from different practical application scenarios, structural characteristic parameters of gantry milling machines can be extracted, including information such as the connection methods between various components, the strength and stability of the supporting structure, etc. For example, the structure of a gantry milling machine may change under heavy loads. By analyzing depth data, these changes in structural characteristic parameters can be determined, thus providing a basis for optimizing the structural design. Through comprehensive analysis of point cloud, image, and depth data from multiple practical application scenarios, multiple characteristic parameters, including geometric characteristic parameters, material characteristic parameters, and structural characteristic parameters, are extracted, providing important basic data for subsequent model optimization and verification.

[0054] In the optimization process of the 3D model of a gantry milling machine, setting a set of compensation and adjustment parameters based on geometric, material, and structural feature parameters is a crucial step. For geometric feature parameters, a careful analysis of the gantry milling machine's shape, dimensions, and other characteristics is conducted. If geometric deviations are found in the actual application scenario, such as inaccurate dimensions or inappropriate angles in certain parts, corresponding geometric adjustment parameters are set to address these issues, ensuring that the machine's external geometry better meets actual requirements. For material feature parameters, based on material properties derived from image analysis, such as hardness, toughness, and wear resistance, if material performance is unsatisfactory in a specific application scenario, material compensation parameters are set. For example, adjusting material strength or selecting a more suitable material alternative can improve the machine's material performance under different working conditions. For structural feature parameters, based on the internal structural information of the machine based on depth data, if insufficient structural stability or unreasonable connection methods are found, structural adjustment parameters are set. For example, strengthening the support structure or optimizing component connections can enhance the overall structural performance of the machine. By comprehensively considering these three feature parameters and carefully setting the set of compensation and adjustment parameters, specific adjustment directions and numerical basis are provided for the optimization of the gantry milling machine's 3D model.

[0055] In the optimization process of the 3D model of the gantry milling machine, the construction of a feedback control mechanism plays a crucial role. First, collected deviation data is synchronized to the feedback control mechanism. This data reflects the differences between the 3D model and the actual application scenario, including geometric deviations, material property inconsistencies, and unreasonable structural designs. Upon receiving this deviation data, the feedback control mechanism begins iterative testing of the 3D model. In each iteration, the model is adjusted and optimized based on the deviation data; for example, geometric parameters are adjusted to reduce shape deviations, material selection is optimized to improve performance, and structural design is improved to enhance stability. Then, the adjusted model is tested again in the actual application scenario, collecting new deviation data. This process is repeated, continuously improving the model based on feedback information. Through multiple iterative tests, the final iterative test results are determined. These results reflect the performance of the 3D model in various aspects and its fit with the actual application scenario after a series of adjustments, providing an important basis for further optimization and improvement of the gantry milling machine's 3D model.

[0056] In the optimization process of the 3D model of a gantry milling machine, a crucial step is to verify the 3D model under multiple practical application scenarios based on iterative test results and a compensation adjustment parameter set. First, the iterative test results are analyzed to examine the performance of the 3D model after multiple adjustments under different practical application scenarios, including the accuracy of geometry, the reliability of material properties, and the stability of the structure. Then, the model is further refined using the compensation adjustment parameter set. Based on the requirements of different practical application scenarios, geometric adjustment parameters, material compensation parameters, and structural adjustment parameters from the compensation adjustment parameter set are used to optimize the model specifically. Next, the adjusted 3D model undergoes rigorous verification in multiple practical application scenarios. Through comparison and evaluation with actual conditions, it is determined whether the model meets the requirements of each scenario. If the model passes verification, a verification token is generated. This verification token identifies the model's validity in a specific practical application scenario. Finally, based on the verification token and the combined verification results from various practical application scenarios, an optimized 3D model is generated. This optimized model is more in line with the needs of practical application scenarios in terms of geometry, materials, and structure, providing more accurate and reliable support for the design, production, and operation of the gantry milling machine.

[0057] Example 2 is based on the same inventive concept as the gantry milling machine 3D modeling device based on multi-source data fusion in the previous examples, such as... Figure 2 As shown, this application provides a 3D modeling method for gantry milling machines based on multi-source data fusion. The method and apparatus embodiments in this application are based on the same inventive concept. The method includes: Step S100: Perform data sensing on the gantry machine tool through multiple data sources to obtain a multi-source dataset.

[0058] Step S200: Traverse the multi-source dataset for analysis and processing to construct multiple three-dimensional data elements.

[0059] Step S300: Extract features from the multiple three-dimensional data elements, register the multiple three-dimensional data elements according to the data element features, and determine the multi-source data registration result.

[0060] Step S400: Based on the multi-source data registration results, the multiple three-dimensional data elements are fused to construct a three-dimensional model of the gantry machine tool.

[0061] Step S500: Introduce multiple practical application scenarios of the gantry milling machine, conduct test feedback based on the multiple practical application scenarios and the three-dimensional model, generate deviation data to dynamically compensate the three-dimensional model, and determine the three-dimensional optimization model of the gantry milling machine.

[0062] Furthermore, step S100 also includes: Step S110: Traverse the multiple data sources to determine multiple data sensing devices, the multiple data sensing devices including a first data sensing device, a second data sensing device, and a third data sensing device.

[0063] Step S120: Scan the external structure of the gantry milling machine based on the first data sensing device to determine multiple point cloud data.

[0064] Step S130: Based on the second data sensing device, the gantry machine tool is collected from multiple angles to determine multiple image data.

[0065] Step S140: Collect data on the internal structure of the gantry milling machine based on the third data sensing device to determine multiple depth data.

[0066] Step S150: Add the multiple point cloud data, the multiple image data, and the multiple depth data to the multi-source dataset.

[0067] Furthermore, step S200 also includes: Step S210: Preprocess the multi-source dataset to obtain a multi-source processed dataset, which includes an outlier set, an augmented image set, and a depth data correction set.

[0068] Step S220: Based on the multiple point cloud data and the outlier point set, denoise is performed to construct a class of three-dimensional data elements, wherein the class of three-dimensional data elements is three-dimensional point matrix parameters.

[0069] Step S230: Based on the multiple image data and the enhanced image set, perform conversion to construct two types of three-dimensional data elements, where the two types of three-dimensional data elements are three-dimensional image parameters.

[0070] Step S240: Based on the multiple depth data and the depth data correction set, perform correction to construct three types of three-dimensional data elements, which are three-dimensional spatial parameters.

[0071] Step S250: Integrate the first type of three-dimensional data elements, the second type of three-dimensional data elements, and the third type of three-dimensional data elements to construct the plurality of three-dimensional data elements.

[0072] Furthermore, step S210 also includes: Step S211: Perform sensing time series analysis based on the multi-source dataset to generate a time series.

[0073] Step S212: Based on the multiple point cloud data and the time series, noise identification is performed to determine the outlier set.

[0074] Step S213: Perform wavelet transform based on the multiple image data and the time series to determine the enhanced image set.

[0075] Step S214: Based on the multiple depth data and the time series, perform error analysis to determine the depth data correction set.

[0076] Step S215: Add the outlier set, the enhanced image set, and the depth data correction set to the multi-source processing dataset.

[0077] Furthermore, step S400 also includes: Step S410: Perform an impact analysis on the multiple three-dimensional data elements according to the multi-source data registration results, determine the initial weights of the multiple three-dimensional data elements based on the analysis results, and construct a data dictionary.

[0078] Step S420: Based on the data dictionary, update the initial weights of the first type of 3D data element, the second type of 3D data element, and the third type of 3D data element to generate multiple weight coefficients.

[0079] Step S430: The three-dimensional point matrix parameters, the three-dimensional image parameters, and the three-dimensional spatial parameters are weighted and fused according to the multiple weight coefficients. Based on the fusion result, the three-dimensional point matrix parameters, the three-dimensional image parameters, and the three-dimensional spatial parameters are reconstructed to construct the three-dimensional model of the gantry milling machine.

[0080] Furthermore, step S500 also includes: Step S510: Traverse the multiple real-world application scenarios to perform simulation tests on the 3D model and generate multiple simulation test results.

[0081] Step S520: Based on the multiple simulation test results, perform deviation calculations on the multiple actual application scenarios to generate initial deviation data.

[0082] Step S530: Match the initial deviation data with the multiple actual application scenarios, perform comprehensive analysis based on the matching results, and construct a deviation distribution map.

[0083] Step S540: Perform cluster analysis on the initial deviation data according to the deviation distribution map to determine multiple deviation types.

[0084] Step S550: Based on the multiple deviation types and the initial deviation data, feedback is performed to obtain the deviation data.

[0085] Furthermore, step S500 also includes: Step S560: Based on the multiple practical application scenarios, perform feature extraction on the three-dimensional model of the gantry milling machine to obtain multiple feature parameters, including geometric feature parameters, material feature parameters, and structural feature parameters.

[0086] Step S570: Set a set of compensation adjustment parameters based on the geometric feature parameters, the material feature parameters, and the structural feature parameters.

[0087] Step S580: Construct a feedback control mechanism, synchronize the deviation data to the feedback control mechanism, perform iterative testing on the three-dimensional model through the feedback control mechanism, and determine the iterative test results.

[0088] Step S590: Based on the iterative test results and the compensation adjustment parameter set, the three-dimensional model is verified according to the multiple actual application scenarios, a verification token is generated, and the three-dimensional optimized model is generated according to the verification token.

[0089] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0090] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0091] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A 3D modeling device for gantry milling machines based on multi-source data fusion, characterized in that, The device includes: A multi-source dataset acquisition module is used to perform data sensing on the gantry machine tool through multiple data sources to obtain a multi-source dataset. A 3D data element construction module is used to traverse the multi-source dataset for analysis and processing, and to construct multiple 3D data elements. A multi-source data registration result determination module is used to extract features from the multiple three-dimensional data elements, register the multiple three-dimensional data elements according to the data element features, and determine the multi-source data registration result. A 3D model construction module is used to fuse multiple 3D data elements according to the multi-source data registration results to construct a 3D model of the gantry machine tool. The three-dimensional optimization model determination module is used to introduce multiple actual application scenarios of the gantry milling machine, and to conduct test feedback based on the multiple actual application scenarios and the three-dimensional model to generate deviation data for dynamic compensation of the three-dimensional model, thereby determining the three-dimensional optimization model of the gantry milling machine.

2. The gantry milling machine 3D modeling device based on multi-source data fusion as described in claim 1, characterized in that, The multi-source dataset acquisition module also includes: The plurality of data sources are traversed to determine a plurality of data sensing devices, the plurality of data sensing devices including a first data sensing device, a second data sensing device, and a third data sensing device; The external structure of the gantry milling machine is scanned based on the first data sensing device to determine multiple point cloud data. Based on the second data sensing device, multiple image data are collected from the gantry machine tool from multiple angles to determine multiple image data; The internal structure of the gantry milling machine is collected based on the third data sensing device to determine multiple depth data. The multiple point cloud data, the multiple image data, and the multiple depth data are added to the multi-source dataset.

3. The gantry milling machine 3D modeling device based on multi-source data fusion as described in claim 2, characterized in that, The three-dimensional data element construction module also includes: The multi-source dataset is preprocessed to obtain a multi-source processed dataset, which includes an outlier set, an augmented image set, and a depth data correction set. Denoising is performed based on the multiple point cloud data and the outlier point set to construct a class of three-dimensional data elements, which are three-dimensional point matrix parameters. Based on the multiple image data combined with the enhanced image set, two types of three-dimensional data elements are constructed, and the two types of three-dimensional data elements are three-dimensional image parameters. Based on the multiple depth data and the depth data correction set, three types of three-dimensional data elements are constructed, which are three-dimensional spatial parameters. The first type of three-dimensional data element, the second type of three-dimensional data element, and the third type of three-dimensional data element are integrated to construct the multiple three-dimensional data elements.

4. The gantry milling machine 3D modeling device based on multi-source data fusion as described in claim 3, characterized in that, The three-dimensional data element construction module also includes: Based on the multi-source dataset, a sensing time series analysis is performed to generate a time series. Based on the multiple point cloud data and the time series, noise identification is performed to determine the outlier set; Wavelet transform is performed on the multiple image data and the time series to determine the enhanced image set; Error analysis is performed based on the multiple depth data and the time series to determine the depth data correction set; The outlier set, the enhanced image set, and the depth data correction set are added to the multi-source processing dataset.

5. The gantry milling machine 3D modeling device based on multi-source data fusion as described in claim 3, characterized in that, The 3D model construction module also includes: An impact analysis is performed on the multiple three-dimensional data elements based on the multi-source data registration results. The initial weights of the multiple three-dimensional data elements are determined based on the analysis results, and a data dictionary is constructed. Based on the data dictionary, the initial weights of the first type of 3D data element, the second type of 3D data element, and the third type of 3D data element are updated to generate multiple weight coefficients; The three-dimensional point matrix parameters, the three-dimensional image parameters, and the three-dimensional spatial parameters are weighted and fused according to the multiple weighting coefficients. Based on the fusion result, the three-dimensional point matrix parameters, the three-dimensional image parameters, and the three-dimensional spatial parameters are reconstructed to construct the three-dimensional model of the gantry milling machine.

6. The gantry milling machine 3D modeling device based on multi-source data fusion as described in claim 1, characterized in that, The three-dimensional optimization model determination module also includes: The three-dimensional model is simulated and tested through multiple real-world application scenarios to generate multiple simulation test results. Based on the multiple simulation test results, deviation calculations are performed on the multiple practical application scenarios to generate initial deviation data; The initial deviation data is matched with the multiple actual application scenarios, and a comprehensive analysis is performed based on the matching results to construct a deviation distribution map; Cluster analysis is performed on the initial deviation data according to the deviation distribution map to determine multiple deviation types; The deviation data is obtained by combining the multiple deviation types with the initial deviation data.

7. The 3D modeling device for gantry milling machines based on multi-source data fusion as described in claim 6, characterized in that, The three-dimensional optimization model determination module also includes: Based on the aforementioned multiple practical application scenarios, feature extraction is performed on the three-dimensional model of the gantry milling machine to obtain multiple feature parameters, including geometric feature parameters, material feature parameters, and structural feature parameters. A set of compensation and adjustment parameters is set based on the geometric feature parameters, the material feature parameters, and the structural feature parameters; A feedback control mechanism is constructed to synchronize the deviation data to the feedback control mechanism. The 3D model is iteratively tested through the feedback control mechanism to determine the iterative test results. Based on the iterative test results and the compensation adjustment parameter set, the 3D model is verified according to the multiple practical application scenarios, a verification token is generated, and the 3D optimized model is generated based on the verification token.

8. A 3D modeling method for gantry milling machines based on multi-source data fusion, characterized in that, The method is applied to the apparatus according to any one of claims 1 to 7, the method comprising: Data sensing of the gantry milling machine tool is performed through multiple data sources to obtain multi-source datasets; The multi-source dataset is traversed for analysis and processing to construct multiple three-dimensional data elements; Feature extraction is performed on the multiple three-dimensional data elements, and the multiple three-dimensional data elements are registered according to the data element features to determine the multi-source data registration result; The multiple three-dimensional data elements are fused based on the multi-source data registration results to construct a three-dimensional model of the gantry milling machine. Multiple practical application scenarios of gantry milling machines are introduced. Based on these scenarios and the three-dimensional model, test feedback is conducted to generate deviation data for dynamic compensation of the three-dimensional model, thereby determining the optimized three-dimensional model of the gantry milling machine.

Images