Skeleton technological procedure planning method and system based on clustering and storage medium

Abstract

The invention relates to the technical field of aircraft manufacturing, and discloses a clustering-based skeleton process procedure planning method and system and a storage medium, and the method comprises the following steps: constructing a skeleton part set, and setting an assembly threshold and a distance threshold for classifying skeleton parts; obtaining a centroid coordinate of each skeleton part in the skeleton part set; randomly selecting mass centers of k skeleton parts from the skeleton part set to form k clusters; distributing the skeleton parts into k clusters to form k cluster part subsets; obtaining a centroid coordinate of each cluster part subset; the contour coefficient of each skeleton part is calculated, the skeleton part with the largest distance in the corresponding cluster part subsets is obtained, the skeleton part is distributed to other cluster part subsets with the smallest distance from the skeleton part, and classification of the skeleton parts is completed. Based on the centroid coordinates of the skeleton parts in the MBD mathematical model, the skeleton parts are classified and grouped by adopting a clustering algorithm, and the problems of low classification efficiency and low accuracy during skeleton process procedure planning are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft manufacturing, and more specifically, to a clustering-based skeleton process planning method, system, and storage medium. Background Art

[0002] A process plan is a production process document prepared by the process department according to design requirements, process technical requirements, and quality requirements. The process plan includes not only process information but also production information and quality information, and is a specific work instruction for guiding workers to perform actual operations on the specified assembly process flow, including operation instructions, processes, assembly time sequences, change records, and other information.

[0003] The difficulty in assembling aircraft skeleton parts lies in the inability to accurately and efficiently determine the granularity of the assembled parts, that is, the inability to effectively complete the rapid classification and grouping of skeleton part assembly. For the assembly of aircraft skeleton parts, usually, process personnel rely on experience to complete the skeleton assembly plan. Due to the high degree of human intervention and inconsistent capabilities of process personnel, the results of the skeleton plan are inaccurate. How to effectively group and assemble skeleton parts requires repeated deduction and demonstration in the process design. This problem has always troubled process designers and increased their workload. In addition, in some existing technologies, a similarity calculation model is used based on part codes for process push, and the cost replacement accuracy constructed by the analytic hierarchy process model fails to effectively improve the work efficiency of the process.

[0004] Therefore, in the prior art, when performing skeleton process planning, there are technical problems of low classification efficiency and low accuracy due to high human intervention. Summary of the Invention

[0005] The purpose of the present invention is to provide a clustering-based skeleton process planning method, system, and storage medium, which solves the problems of low classification efficiency and low accuracy in skeleton process planning in the prior art.

[0006] The present invention is achieved by the following technical solutions: In the first aspect, a clustering-based skeleton process planning method for classifying skeleton parts in the preparation of a skeleton process plan includes the following steps: S01. Construct a set of skeleton parts and set an assembly threshold and a distance threshold for classifying the skeleton parts; S02. Obtain the centroid coordinates of each skeleton part in the set of skeleton parts; S03. Randomly select the centroids of k skeleton parts from the set of skeleton parts to form k clusters; S04. Obtain the minimum distance between each skeleton part and each cluster. Based on the principle that the minimum distance is less than the distance threshold and the number of parts in the current cluster part subset is less than the assembly threshold, allocate the skeleton parts into k clusters to form k cluster part subsets; S05. Obtain the centroid coordinates of each cluster part subset; S06. Calculate the silhouette coefficient of each skeleton part. When the silhouette coefficient is less than the preset value, obtain the skeleton part with the maximum distance in the corresponding cluster part subset, and allocate this skeleton part to the other cluster part subset with the minimum distance to it; S07. Repeat steps S05 and S06 to complete the classification of the skeleton parts.

[0007] To better implement the present invention, further, the method for constructing the skeleton part set includes: Construct a skeleton part rule base rule. Based on the MBD digital model data, identify the skeleton parts according to rule to form a skeleton part set , n is the number of skeleton parts.

[0008] To better implement the present invention, further, set the skeleton part assembly threshold smx , and the skeleton part assembly threshold is configured such that the number of parts assembled each time is less than smx ; Set the skeleton part distance threshold kmx , and the skeleton part distance threshold is configured such that the distance between parts is less than kmx .

[0009] To better implement the present invention, further, based on the MBD digital model, parse the digital model coordinates to obtain the centroid coordinates of each skeleton part.

[0010] To better implement the present invention, further, step S04 includes: Calculate the distance from the i th skeleton part to the cluster , and obtain the minimum distance dmin i , if , cparts i The number of skeleton parts in smx is less than cparts i , put this skeleton part into the cluster part subset s i is the centroid coordinate of the i th skeleton part, k i is the centroid coordinate of the i th cluster, ,j For the subset of cluster parts cparts i The number of skeleton parts in it; repeat the above steps to form k subsets of cluster parts.

[0011] To better implement the present invention, further, the centroid coordinates of the subset of cluster parts in step S05 are .

[0012] To better implement the present invention, further, step S06 includes: Calculate the silhouette coefficient of the skeleton parts , where a(i) is cparts i in, the i th skeleton part and the average distance of all other skeleton parts in its subset of cluster parts, b(i) is the i th skeleton part and the average distance of all skeleton parts in the nearest subset of cluster parts; When F(i) is less than the preset value, obtain cparts i the skeleton part with the largest distance in, use the Euclidean distance, calculate the minimum distance from this skeleton part to other subsets of cluster parts and assign it to the subset of cluster parts corresponding to the minimum distance.

[0013] In a second aspect, the present invention also provides a clustering-based skeleton process planning system for classifying skeleton parts by using the clustering-based skeleton process planning method, including: A skeleton part set construction module, which is used to identify skeleton parts based on the MBD digital model data according to the skeleton part rule library and construct a skeleton part set; A skeleton part cluster establishment module, which is used to assign skeleton parts to each cluster to form subsets of cluster parts; A skeleton part classification module, which is used to reassign the skeleton part with the largest distance in each subset of cluster parts to other subsets of cluster parts according to the silhouette coefficient of each skeleton part.

[0014] In a third aspect, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the clustering-based skeleton process planning method described above.

[0015] Compared with the prior art, the present invention has the following advantages and beneficial effects: Based on the centroid coordinates of the skeleton parts in the MBD digital model, the present invention classifies and groups the skeleton parts using a clustering algorithm. By setting a distance threshold, it ensures that there is an assembly relationship between the skeleton parts assigned to the same group, improving the accuracy of classification. By setting an assembly threshold, it controls the number of skeleton parts in each group, ensuring the efficiency of skeleton part classification and the compilation of the skeleton process plan, as well as the efficiency of subsequent assembly. By using the silhouette coefficient, it ensures the rationality of clustering and improves the accuracy of classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described in conjunction with the following drawings and embodiments. All creative concepts of the present invention should be regarded as the disclosed content and the protection scope of the present invention.

[0017] Figure 1 It is a flow chart of the skeleton process plan planning method in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The skeleton of an aircraft mainly includes: The fuselage skeleton, such as the fuselage frames, fuselage stringers, etc., which constitute the main structure of the aircraft, support the weight of the aircraft and withstand various forces during flight; The wing skeleton, such as the wing beams, wing ribs, etc., which constitute the structure of the wing, support the shape of the wing and withstand the aerodynamic loads during flight; The empennage skeleton, such as the frames and stringers of the horizontal and vertical tails, which constitute the structure of the empennage, control the flight direction and stability of the aircraft; The landing gear skeleton, such as the landing gear struts, wheels, brakes, etc., which constitute the landing gear system of the aircraft, enabling the aircraft to take off and land and move on the ground; Other skeleton parts, such as engine mounts, flaps, slats, ailerons, elevators, rudders, etc., which are various control surfaces and components on the aircraft and play an important role in the flight performance and safety of the aircraft.

[0019] The clustering-based skeleton process plan planning method of the present invention is used for the automatic classification of skeleton parts in the compilation of the skeleton process plan. By automatically classifying the skeleton parts, it facilitates the automatic compilation of the skeleton process plan and improves the efficiency of the skeleton process plan compilation.

[0020] In some embodiments of the present invention, the clustering-based skeleton process plan planning method, with reference to Figure 1 , includes the following steps: S01. Construct a set of skeleton parts and set an assembly threshold and a distance threshold for classifying the skeleton parts; S02. Obtain the centroid coordinates of each skeleton part in the set of skeleton parts; S03. Randomly select the centroids of k skeleton parts from the set of skeleton parts to form k clusters; S04. Obtain the minimum distance between each skeleton part and each cluster. Based on the principle that the minimum distance is less than the distance threshold and the number of parts in the current cluster part subset is less than the assembly threshold, allocate the skeleton parts into k clusters to form k cluster part subsets; S05. Obtain the centroid coordinates of each cluster part subset; S06. Calculate the silhouette coefficient of each skeleton part. When the silhouette coefficient is less than the preset value, obtain the skeleton part with the largest distance in the corresponding cluster part subset, and allocate this skeleton part to the other cluster part subset with the smallest distance to it; S07. Repeat steps S05 and S06 to complete the classification of the skeleton parts.

[0021] In this embodiment, the method for constructing the skeleton part set includes: Construct a skeleton part rule base rule. Based on the MBD digital model data, identify the skeleton parts according to the rule to form a skeleton part set , n is the number of skeleton parts.

[0022] Specifically, the skeleton part rule base rule may include skeleton parts such as frames, spars, engine pylons, flaps, slats, ailerons, etc., effectively supporting the planning according to clustering.

[0023] Furthermore, the method for setting the skeleton part assembly threshold and the distance threshold includes: Set the skeleton part assembly threshold smx , and the skeleton part assembly threshold is configured such that the number of parts assembled each time is less than smx ; Set the skeleton part distance threshold kmx , and the skeleton part distance threshold is configured such that the distance between parts is less than kmx .

[0024] Adopting this embodiment, the number of parts assembled is constrained by the skeleton part distance threshold kmx , effectively controlling the assembly efficiency. By setting the threshold between parts kmx , it is ensured that there is an assembly relationship between parts, improving the reliability and accuracy of the classification planning.

[0025] Furthermore, based on the MBD digital model, parse the digital model coordinates to obtain the centroid coordinates of each skeleton part.

[0026] Furthermore, step S04 includes: Calculate the distance from the i th skeleton part to the cluster , and obtain the minimum distance dmin i , if ,cparts i The number of skeleton parts in smx is less than cparts i , and put the skeleton part into the subset of cluster parts s i . Among them, i is the centroid coordinate of the k i -th skeleton part, i is the centroid coordinate of the -th cluster, j , and cparts i is the number of skeleton parts in the subset of cluster parts

[0027] . Repeat the above steps to form k subsets of cluster parts. i . Further, the centroid coordinate of the -th subset of cluster parts in step S05 can be expressed as j , where cparts i is the number of skeleton parts in the subset of cluster parts s i , and i is the centroid coordinate of the

[0028] . Further, step S06 includes: Calculating the silhouette coefficient of the skeleton part , where a(i) is cparts i in, the average distance between the i -th skeleton part and all other skeleton parts in its subset of cluster parts, b(i) is the average distance between the i -th skeleton part and all skeleton parts in the nearest subset of cluster parts; When F(i) is less than the preset value, obtain cparts i the skeleton part with the largest distance in, calculate the minimum distance from this skeleton part to other subsets of cluster parts and allocate it to the subset of cluster parts corresponding to the minimum distance.

[0029] Adopting this embodiment, on the basis of effectively improving the classification efficiency, the irrationality of clustering is avoided and the classification accuracy is improved.

[0030] The method for planning the skeleton process specification based on clustering of the present invention will be described in detail below in conjunction with specific embodiments. The method includes the following steps: Step S101: Construct a part set. Specifically, construct a skeleton part rule base, denoted as rule = (frame, beam...). Based on the MBD digital model data, identify the skeleton parts according to rule to form a skeleton part set , n is the number of skeleton parts; Step S102: Set the skeleton part assembly threshold. Specifically, denote the skeleton part assembly threshold as smx , that is, the number of parts assembled each time is less than smx ; Step S103: Set the skeleton part distance threshold. Specifically, denote the skeleton part distance threshold as kmx , that is, the distance between parts is less than kmx ; Step S104: Based on the MBD digital model, parse the digital model coordinates and calculate the centroid coordinates of each skeleton part. Specifically, according to parts in Step S101, extract the coordinates of each skeleton part and denote them as ; Step S105: Randomly select k centroid points of skeleton parts to form k clusters. Specifically, randomly select k centroid points of the skeleton k =( k 1 …… k k ), to form k clusters; Step S106: Calculate the distance i from the th skeleton part to the cluster, and obtain the minimum distance dmin i . If , cparts i the number of skeleton parts in smx is less than cparts i , put this skeleton part into the cluster part subset s i . Among them, i is the centroid coordinate of the k i th skeleton part, i is the centroid coordinate of the th cluster, j , cparts i is the number of skeleton parts in the cluster part subset Step S107: The centroid coordinates of each cluster part subset. The centroid coordinates of the cluster part subset are expressed as ; Step S108, calculate the silhouette coefficient, specifically, according to calculate the silhouette coefficient, where a(i) is cparts i in, the i th skeleton part and the average distance between all other skeleton parts in its cluster part subset, that is, the within-cluster distance; b(i) is the average distance between the i th skeleton part and all skeleton parts in the nearest cluster part subset; If , obtain the i th cluster part subset cparts i in the skeleton part with the largest distance, use the Euclidean distance to calculate the minimum distance from the part with the largest distance to other cluster part subsets, and assign the part with the largest distance to the cluster corresponding to the minimum distance; Step S109, recalculate Step S107 and Step S108, recalculate the centroid coordinates of the cluster part subsets after reallocation, and calculate the silhouette coefficient until the silhouette coefficient is greater than the preset value to complete the classification of the skeleton parts.

[0031] On the other hand, some embodiments of the present invention relate to a clustering-based skeleton process planning system for implementing the skeleton process planning method in the above embodiments, including: A skeleton part set construction module, which is used to identify skeleton parts based on the MBD digital model data according to the skeleton part rule library and construct a skeleton part set; A skeleton part cluster establishment module, which is used to assign skeleton parts to each cluster to form cluster part subsets; A skeleton part classification module, which is used to reassign the skeleton part with the largest distance in each cluster part subset to other cluster part subsets according to the silhouette coefficient of each skeleton part.

[0032] On the other hand, some embodiments of the present invention relate to a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the clustering-based skeleton process planning method in the above embodiments.

[0033] The above is only a preferred embodiment of the present invention, and does not impose any form of limitation on the present invention. Any simple modification or equivalent change made to the above embodiments based on the technical essence of the present invention falls within the protection scope of the present invention.

Claims

1. A clustering-based skeleton process planning method, characterized in that, For the classification of skeleton parts in the preparation of the skeleton process plan, the following steps are included: S01. Construct a set of skeleton parts and set an assembly threshold and a distance threshold for classifying the skeleton parts; S02. Obtain the centroid coordinates of each skeleton part in the set of skeleton parts; S03. Randomly select the centroids of k skeleton parts from the set of skeleton parts to form k clusters; S04. Obtain the minimum distance between each skeleton part and each cluster. Based on the principle that the minimum distance is less than the distance threshold and the number of parts in the current cluster part subset is less than the assembly threshold, allocate the skeleton parts to the k clusters to form k cluster part subsets; S05. Obtain the centroid coordinates of each cluster part subset; S06. Calculate the silhouette coefficient of each skeleton part. When the silhouette coefficient is less than the preset value, obtain the skeleton part with the largest distance in the corresponding cluster part subset and allocate this skeleton part to the other cluster part subset with the smallest distance to it; S07. Repeat steps S05 and S06 to complete the classification of the skeleton parts.

2. The clustering-based skeleton process planning method according to claim 1, wherein The method for constructing a set of skeleton parts includes: Build a rule library for skeleton parts, based on MBD digital model data, identify skeleton parts according to the rules, and form a set of skeleton parts , n is the number of skeleton parts.

3. The clustering-based skeleton process planning method according to claim 2, wherein Set the assembly threshold for the skeleton parts smx , the assembly threshold for the skeleton parts is configured such that the number of parts assembled each time is less than smx ; Set the skeleton part distance threshold kmx , the skeleton part distance threshold is configured such that the distance between parts is less than kmx .

4. The clustering-based skeleton process planning method according to claim 1, characterized in that, Based on the MBD digital model, analyze the digital model coordinates to obtain the centroid coordinates of each skeleton part.

5. The clustering-based skeleton process planning method according to claim 3, wherein Step S04 includes: Calculate the distance from the i th skeleton part to the cluster , and obtain the minimum distance dmin i . If , cparts i the number of skeleton parts in smx is less than cparts i , put this skeleton part into the subset of cluster parts s i where i is the centroid coordinate of the k i th skeleton part, i is the centroid coordinate of the th cluster, j and cparts i is the number of skeleton parts in the subset of cluster parts . Repeat the above steps to form k subsets of cluster parts.

6. The clustering-based skeleton process planning method according to claim 5, wherein The centroid coordinates of the subset of cluster parts in step S05 are .

7. The clustering-based skeleton process planning method according to claim 5, wherein Step S06 includes: Calculate the silhouette coefficient of the skeleton parts , where a(i) is cparts i in, the i th skeleton part and the average distance between it and all other skeleton parts in the subset of cluster parts where it is located, b(i) is the average distance between the i th skeleton part and all skeleton parts in the nearest subset of cluster parts; When F(i) is less than a preset value, obtain cparts i the skeletal part with the largest distance in i . Using the Euclidean distance, calculate the minimum distance from this skeletal part to other subsets of cluster parts and assign it to the subset of cluster parts corresponding to the minimum distance.

8. A clustering-based skeleton process planning system, characterized in that For classifying skeleton parts by using the clustering-based skeleton process plan planning method described in any one of claims 1-7, it includes: A skeleton part set construction module, which is used to identify skeleton parts and construct a set of skeleton parts; A skeleton part cluster establishment module, which is used to allocate skeleton parts to each cluster to form cluster part subsets; A skeleton part classification module, which is used to re-allocate the skeleton part with the largest distance in each cluster part subset to other cluster part subsets according to the silhouette coefficient of each skeleton part.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the clustering-based skeleton process plan planning method described in any one of claims 1-7.

Images