A feature-based method for characterizing multi-source heterogeneous data in assembly processes

Abstract

This invention relates to the field of aircraft assembly technology, specifically to a feature-based method for representing multi-source heterogeneous data in the assembly process. The method includes the following steps: pre-reading and analyzing the assembly process model to identify key assembly features; extracting attribute information corresponding to the key assembly features in the relevant assembly process; prioritizing the key assembly features; developing a measurement scheme for the entire assembly process based on the prioritized key assembly features and attribute information, combined with the geometric error propagation rules throughout the assembly process; matching the measured data from the assembly site with the attribute information according to the measurement scheme; determining the data representation strategy based on the type of measured data from the assembly site, and fitting geometric data and labeling non-geometric data. The purpose of this feature-based method for representing multi-source heterogeneous data in the assembly process is to address the problem of assembly deviations and defects easily arising under the combined effect of multi-source heterogeneous data generated during aircraft assembly.

Description

Technical Field

[0001] This invention relates to the field of aircraft assembly technology, and more specifically to a feature-based method for characterizing multi-source heterogeneous data in the assembly process. Background Technology

[0002] Aircraft assembly consists of several assembly sections (such as the forward, mid-, and aft fuselage assembly sections), each of which can be further subdivided into several assembly stations. Typically, the specific tasks performed at each assembly station are defined by the Assembly Outline (AO). The AO details the assembly operations, and critical inspections are performed after each assembly operation. The main elements of aircraft assembly technical status can be summarized as: assembly section, assembly station, AO, assembly operations, and inspection. In actual assembly operations, due to tooling errors, assembly deformation, and manufacturing deviations of parts, assembled aircraft products often fail to meet design requirements, leading to rework and scrap incidents.

[0003] Currently, domestic aircraft manufacturers primarily focus their quality control measures on the assembly process design stage. They utilize digital assembly tolerance analysis software to scale component design tolerances and adjust assembly processes. Virtual assembly is repeatedly performed in a digital software environment to gradually adjust component tolerances and assembly processes, aiming to control assembly precision. However, in the actual aircraft assembly production stage, effective precision prediction and control methods are lacking. Aside from allowing for adjustments during assembly in certain processes, most assembly processes are in a trial state, leading to unnecessary rework and tooling adjustments, resulting in decreased assembly productivity. Due to structural characteristics and rigidity, aircraft assembly extensively uses riveting and bolting connections. Furthermore, to ensure assembly coordination and shape accuracy, and to guarantee structural rigidity of components and parts during assembly, aircraft assembly employs numerous complex and highly accurate assembly jigs. All of this demonstrates that aircraft assembly has very distinct characteristics compared to the assembly process of general mechanical products.

[0004] The multi-source heterogeneous data generated during the assembly process includes heterogeneous laminated structures (CFRP, GFRP, aluminum alloy, titanium alloy, etc.), assembly forces (positioning, drilling, connection, springback), geometric errors from multiple sources (part errors, tooling errors, deformation errors), and variable assembly environments (temperature changes, humidity changes), etc. Under the combined effect of these multi-source heterogeneous data, assembly deviations and assembly defects are very likely to occur.

[0005] Therefore, the inventors provide a feature-based method for characterizing multi-source heterogeneous data in assembly processes. Summary of the Invention

[0006] (1) Technical problems to be solved

[0007] This invention provides a feature-based method for characterizing multi-source heterogeneous data in the assembly process, which solves the technical problem that assembly deviations and defects are easily generated under the combined effect of multi-source heterogeneous data generated during aircraft assembly.

[0008] (2) Technical solution

[0009] This invention provides a feature-based method for characterizing multi-source heterogeneous data in an assembly process, comprising the following steps:

[0010] Pre-reading analysis of the assembly process model is performed to identify key assembly features;

[0011] Based on the actual assembly process requirements, extract the attribute information corresponding to the key assembly features of the corresponding assembly process;

[0012] Based on the impact of the technical indicators of each part and each assembly unit in the entire assembly process on the final top-level technical requirements, the key assembly features are prioritized.

[0013] Based on the sorted assembly key features and attribute information, and combined with the geometric error propagation law in the entire assembly process, a measurement scheme for the entire assembly process is formulated.

[0014] According to the measurement scheme, the actual measured data at the assembly site are matched with the attribute information;

[0015] The data representation strategy is determined based on the type of measured data from the assembly site, and geometric data is fitted and non-geometric data is labeled.

[0016] Furthermore, the pre-reading analysis of the assembly process model to screen out key assembly features specifically includes:

[0017] Based on the key and important characteristics defined in the design phase, the fit relationships of components, and the inspection requirements of each process, the assembly process model is pre-read and analyzed to identify the key assembly features.

[0018] Furthermore, the key assembly features include geometric features and non-geometric features.

[0019] Furthermore, the non-geometric features include combined features, associated features, and attribute features.

[0020] Furthermore, the step of extracting attribute information corresponding to the key assembly features of the corresponding assembly process according to the actual assembly process needs specifically includes:

[0021] For the aircraft component assembly process, the top-level technology of the product is decomposed into specific assembly features, and the attribute information contained in each of the key assembly features is extracted.

[0022] Furthermore, the matching of the measured data from the assembly site with the attribute information according to the measurement scheme specifically involves:

[0023] By comparing the sorting results of the key assembly features, the corresponding attributes, the dimensional geometric measurement data generated at the assembly site, and the multi-dimensional heterogeneous data from multiple sources, the measured data at the assembly site are matched with the attribute information.

[0024] Furthermore, after matching the measured data from the assembly site with the attribute information, the process further includes: filtering out noise data.

[0025] Furthermore, the fitting of geometric data specifically includes:

[0026] When representing planar data, the planar equation is fitted based on the first point cloud data obtained from the measurement, the planar normal is calculated, and the planar attribute information is labeled according to the measurement results.

[0027] Furthermore, the fitting of geometric data further includes:

[0028] When characterizing the hole data, the cylindrical equation is fitted based on the measured second point cloud data to calculate the hole center point, hole normal vector and hole depth, and the surface attribute information is labeled according to the measurement results.

[0029] Furthermore, the annotation of non-geometric data specifically includes:

[0030] When characterizing the gap / step difference data, the surface of the part associated with the gap / step difference is fitted according to the measured third point cloud data, the gap value, step difference value and included angle value are calculated, and the reference surface information is marked according to the gap / step difference associated part information.

[0031] (3) Beneficial effects

[0032] In summary, this invention comprehensively considers dimensional and geometric measurement data generated at the assembly site, as well as multi-source heterogeneous data from multiple dimensions such as personnel, machines, materials, methods, and environment. Combined with assembly process requirements, it reconstructs and characterizes multi-source heterogeneous data based on key assembly features for assembly simulation and precision control, guiding process optimization and improving assembly accuracy and efficiency. Attached Figure Description

[0033] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the 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.

[0034] Figure 1 This is a flowchart illustrating a feature-based method for characterizing multi-source heterogeneous data in an assembly process, as provided in an embodiment of the present invention. Detailed Implementation

[0035] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The following detailed description of the embodiments and the accompanying drawings are used to illustrate the principles of the present invention by way of example, but should not be used to limit the scope of the present invention, that is, the present invention is not limited to the described embodiments.

[0036] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0037] Figure 1 This is a flowchart illustrating a feature-based method for characterizing multi-source heterogeneous data in an assembly process, as provided in an embodiment of the present invention. Figure 1 As shown, the method may include the following steps:

[0038] S100. Perform pre-reading analysis on the assembly process model and screen out key assembly features;

[0039] S200. Based on the actual assembly process requirements, extract the attribute information corresponding to the key assembly features of the corresponding assembly process.

[0040] S300: Based on the impact of the technical indicators of each part and each assembly unit in the entire assembly process on the final top-level technical requirements, prioritize the key features of the assembly.

[0041] S400. Based on the sorted assembly key features and attribute information, and combined with the geometric error propagation law in the entire assembly process, formulate a measurement scheme for the entire assembly process.

[0042] S500. Based on the measurement plan, match the actual measured data at the assembly site with the attribute information;

[0043] S600. Determine the data representation strategy based on the type of measured data at the assembly site, and fit geometric data and label non-geometric data.

[0044] In the above embodiments, the characterization method is mainly applied to the assembly process of aircraft parts. By adopting this data characterization method, the dimensional geometric measurement data generated on the assembly site, as well as multi-source heterogeneous data from multiple dimensions such as personnel, machines, materials, methods, and environment, can be fully utilized to effectively improve assembly quality and efficiency, and has considerable economic and social benefits.

[0045] As an optional implementation, in step S100, a pre-reading analysis is performed on the assembly process model to screen out key assembly features, specifically:

[0046] Based on the key and important characteristics defined in the design phase, the fit relationships of components, and the inspection requirements of each process, the assembly process model is pre-read and analyzed to identify key assembly features.

[0047] As an optional implementation method, the key features of the assembly include geometric features (such as planes, curved surfaces, curves, etc.) and non-geometric features (such as associated attributes, connection attributes, equipment, environment, etc.). Non-geometric features include combination features, associated features, and attribute features.

[0048] Specifically, the attribute information corresponding to the listed key assembly features is shown in Table 1:

[0049]

[0050] As an optional implementation, in step S200, according to the actual assembly process requirements, the attribute information corresponding to the key assembly features of the corresponding assembly process is extracted. Specifically, for the aircraft parts assembly process, the top-level technology of the product is decomposed into specific assembly features, and the attribute information contained in each key assembly feature is extracted.

[0051] As an optional implementation method, in step S500, according to the measurement scheme, the measured data from the assembly site are matched with the attribute information, specifically as follows:

[0052] By comparing the sorting results of key assembly features, corresponding attributes, dimensional geometric measurement data generated at the assembly site, and multi-dimensional heterogeneous data from multiple sources, the measured data at the assembly site are matched with the attribute information.

[0053] As an optional implementation method, after matching the measured data from the assembly site with the attribute information, the method further includes: filtering out noise data.

[0054] As an optional implementation method, fitting geometric data specifically includes:

[0055] When representing planar data, based on the first point cloud data obtained from the measurement, the plane equation is fitted using the least squares method (random consistency sampling method, eigenvalue method, etc.), the plane normal is calculated, and the plane attribute information is labeled according to the measurement results; and,

[0056] When characterizing the hole data, based on the measured second point cloud data, the cylinder equation is fitted using the least squares method (ellipse method or random consistency sampling method can also be used for cylindrical surfaces, and Poisson reconstruction method or Delaunay triangulation reconstruction method can also be used for curved surfaces), the hole center point, hole normal vector and hole depth are calculated, and the curved surface attribute information is labeled according to the measurement results.

[0057] It should be noted that the above specifically describes the fitting and labeling process for two types of geometric data: planar and curved surfaces.

[0058] As an optional implementation method, non-geometric data is labeled, specifically as follows:

[0059] When characterizing the gap / step difference data, the surface of the part associated with the gap / step difference is fitted according to the measured third point cloud data, the gap value, step difference value and included angle value are calculated, and the reference surface information is marked according to the gap / step difference associated part information.

[0060] It should be noted that the above specifically describes the fitting and labeling process for non-geometric data with related attributes.

[0061] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. The present invention is not limited to the specific steps and structures described above and shown in the figures. Furthermore, for the sake of brevity, detailed descriptions of known methods and techniques are omitted here.

[0062] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art without departing from the scope of the invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A feature-based method for characterizing multi-source heterogeneous data in an assembly process, characterized in that, The method includes the following steps: Pre-reading analysis of the assembly process model is performed to identify key assembly features; Based on the actual assembly process requirements, extract the attribute information corresponding to the key assembly features of the corresponding assembly process; Based on the impact of the technical indicators of each part and each assembly unit in the entire assembly process on the final top-level technical requirements, the key assembly features are prioritized. Based on the sorted assembly key features and attribute information, and combined with the geometric error propagation law in the entire assembly process, a measurement scheme for the entire assembly process is formulated. According to the measurement scheme, the actual measured data at the assembly site are matched with the attribute information; The data representation strategy is determined based on the type of measured data from the assembly site, and geometric data is fitted and non-geometric data is labeled.

2. The feature-based multi-source heterogeneous data representation method for assembly processes according to claim 1, characterized in that, The pre-reading analysis of the assembly process model and the screening of key assembly features are as follows: Based on the key and important characteristics defined in the design phase, the fit relationships of components, and the inspection requirements of each process, the assembly process model is pre-read and analyzed to identify the key assembly features.

3. The feature-based multi-source heterogeneous data characterization method for assembly processes according to claim 1, characterized in that, The key features of the assembly include geometric features and non-geometric features.

4. The feature-based multi-source heterogeneous data characterization method for assembly processes according to claim 3, characterized in that, The non-geometric features include combination features, association features, and attribute features.

5. The feature-based multi-source heterogeneous data characterization method for assembly processes according to claim 1, characterized in that, The step of extracting attribute information corresponding to the key assembly features of the assembly process according to the actual assembly process requirements is as follows: For the aircraft component assembly process, the top-level technology of the product is decomposed into specific assembly features, and the attribute information contained in each of the key assembly features is extracted.

6. The feature-based multi-source heterogeneous data characterization method for assembly processes according to claim 1, characterized in that, The step of matching the measured data from the assembly site with the attribute information according to the measurement scheme is as follows: By comparing the sorting results of the key assembly features, the corresponding attributes, the dimensional geometric measurement data generated at the assembly site, and the multi-dimensional heterogeneous data from multiple sources, the measured data at the assembly site are matched with the attribute information.

7. The feature-based method for characterizing multi-source heterogeneous data in assembly processes according to claim 1, characterized in that, After matching the measured data from the assembly site with the attribute information, the process further includes: filtering out noise data.

8. The feature-based method for characterizing multi-source heterogeneous data in assembly processes according to claim 1, characterized in that, The fitting of geometric data specifically includes: When representing planar data, the planar equation is fitted based on the first point cloud data obtained from the measurement, the planar normal is calculated, and the planar attribute information is labeled according to the measurement results.

9. The feature-based method for characterizing multi-source heterogeneous data in assembly processes according to claim 1, characterized in that, The fitting of geometric data further includes: When characterizing the hole data, the cylindrical equation is fitted based on the measured second point cloud data to calculate the hole center point, hole normal vector and hole depth, and the surface attribute information is labeled according to the measurement results.

10. The feature-based multi-source heterogeneous data characterization method for assembly processes according to claim 1, characterized in that, The annotation of non-geometric data specifically includes: When characterizing the gap / step difference data, the surface of the part associated with the gap / step difference is fitted according to the measured third point cloud data, the gap value, step difference value and included angle value are calculated, and the reference surface information is marked according to the gap / step difference associated part information.

Images