Aircraft conduit part identification method

Abstract

The invention relates to the technical field of computer aided design, and discloses an aircraft conduit part identification method, which adopts a multi-stage screening and contrast identification mode, and specifically comprises the following steps of: screening all parts in an aircraft from two aspects of part name keywords and the proportion of the sum of curved surface areas in the parts in the surface area; filtering irrelevant part digifax to reduce the calculation amount of conduit digifax recognition, and obtaining a part candidate set; in the part candidate set, whether other parts assembled on each part have pipe sleeve parts or not is judged through a shape similarity calculation method, the influence of shape uncertainty of the guide pipe parts on the recognition process can be avoided, and then successful recognition of the guide pipe parts is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer-aided design, and specifically, to a method for identifying aircraft duct parts. Background Art

[0002] Aircraft products usually have a high degree of complexity, and the number of their parts is often in the tens of thousands. Therefore, how to identify target parts from them is a relatively difficult problem, and the identification of duct parts is a typical application scenario of such problems. The current duct identification methods mainly search through keywords in the name. The naming method of parts is highly subjective and cannot be unified, making it impossible to obtain relatively comprehensive results all the time.

[0003] In the patent "A method for calculating the similarity of 3D CAD models based on the discrete bat algorithm (CN110334108A, publication date: 2019-10-15)", the shape similarity of faces is calculated according to the difference in the number of edges of each face matching pair between the source model and the target model, and the structural similarity of face matching pairs is calculated through the adjacency relationship between face matching pairs, so as to find similar 3D models. However, duct parts belong to flexible parts, and their appearance shapes vary greatly with different airframe structures. Therefore, the similarity method usually cannot achieve good results.

[0004] In the patent "A method for identifying manufacturing features of 3D CAD solid models based on PCA and CNN (CN107463533A, publication date: 2017-12-12)", the five-dimensional point cloud data group of the 3D CAD solid model to be identified is reduced to two dimensions by the principal component analysis method, so as to obtain the two-dimensional point cloud data group of the 3D CAD solid model to be identified; the data of the solid model to be identified is input into the trained CNN manufacturing feature recognizer, and the CNN manufacturing feature recognizer outputs all manufacturing features of the 3D CAD solid model to be identified according to the data of the solid model to be identified. However, this method uses a supervised learning method, requires a large number of 3D models as training samples, and requires multiple iterations of tuning to achieve accurate identification of 3D models, which brings difficulties to the implementation of enterprises. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for identifying aircraft duct parts, and solve the problem that it is difficult to effectively obtain duct parts in complex aircraft structures.

[0006] The present invention is realized through the following technical solutions: A method for identifying aircraft duct parts is carried out by means of multi-level screening and comparison identification, and specifically includes the following steps: Step S1: Traverse all part digital models in the aircraft, and add the part digital models whose names contain the keyword "pipe" to the first-level part set ; Step S2. Further filter the primary part set according to the geometric characteristics of the parts , and add the digital models of the parts that meet the requirements in to the secondary set ; Step S3. Obtain the shape vectors of the digital models of each part and their assembled part digital models in ; Step S4. Construct a data set of the digital models of bushing parts and extract their shape vectors; Step S5. Determine whether the assembled part digital models that have a connection relationship with the digital models of each part in are bushing parts, so as to realize the identification of conduit parts.

[0007] To better implement the present invention, further, the specific steps of further filtering the primary part set in step S2 according to the geometric characteristics of the parts include: Step S21. Obtain all the surface elements of the digital models of each part in , and classify them into four types: plane, cylindrical surface, free-form surface, and other surfaces in sequence; Step S22. Statistically calculate the sum of the surface areas of the surfaces of each type in the digital model of the part, and record them as , , and , where , is recorded as the specific surface area, , is recorded as the area of other surfaces; Step S23. Calculate the ratio of the specific surface area to the area of other surfaces, and the calculation method is: ; Step S24. Set a threshold for the surface ratio. If the digital model of the part satisfies , then add it to the secondary set ; Step S25. Traverse all the digital models of the parts in , and repeat steps S21 - S24 to obtain the final .

[0008] To better implement the present invention, further, the specific steps of obtaining the shape vectors of the digital models of each part and their assembled part digital models in in step S3 are as follows: Step S31. For the digital model of the part in p, obtain the assembly part set p having a connection relationship with the part digital model , where represents the i th assembly part digital model connected to the part digital model p , , n represents the number of assembly part digital models having a connection relationship with the part digital model p ; Step S32, obtain the shape information of the part digital model p and each assembly part digital model in the set , and describe it as a k-dimensional shape vector, denoted as: , where ; Step S33, traverse all the part digital models, and repeat Step S31 - Step S32 until the traversal is complete.

[0009] To better implement the present invention, further, in Step S31, through computer-aided interference checking, and then obtain the assembly part digital model set p having a connection relationship with the part digital model .

[0010] To better implement the present invention, further, in Step S32, by randomly sampling points on the surfaces of the part digital model p and each assembly part digital model in the set , and using the shape distribution algorithm to obtain the shape information of the part digital model and its assembly part digital models.

[0011] To better implement the present invention, further, the specific steps of constructing the sleeve part digital model data set and extracting its shape vector in Step S4 are as follows: Step S41, sort out all types of sleeve parts in the aircraft; Step S42, obtain the shape information of the digital models of each type of sleeve part, and describe it as k -dimensional shape vector, denoted as , where , and add it to the set GT .

[0012] To better implement the present invention, further, in Step S42, by randomly sampling points on the surfaces of the digital models of each type of sleeve part, and using the shape distribution algorithm to obtain the shape information of each sleeve part digital model.

[0013] To better implement the present invention, further, in Step S5, judge with The specific steps for determining whether the assembly part digital model with connection relationships among the parts in it is a bushing part digital model are as follows: Step S51. For the part digital models p , assume that is the assembly part digital model that has a connection relationship with the part digital model p . Let be the shape vector of the assembly part digital model q . Calculate the similarity q between the assembly part digital model GT and the bushing part digital model in : ; where represents the shape vector of a type of bushing part digital model; is the value of the th dimension in the i vector, and is the value of the th dimension in the i vector; Step S52. Set the similarity threshold . Assume there is an assembly part digital model , and at the same time there is a such that . Then, consider the part digital model p as a conduit part; Step S53. Traverse , and repeat Steps S51 - S52 to complete the identification of all conduit parts of the aircraft.

[0014] Compared with the prior art, the present invention has the following advantages and beneficial effects: (1) According to work experience and business needs, the process personnel of the present invention construct a data set of bushing parts in the aircraft; then, all parts in the aircraft are screened from two aspects: the keyword of the part name and the proportion of the sum of the surface areas of the curved surfaces in the part in the surface area, filtering out irrelevant part digital models to reduce the computational amount of conduit part identification and obtaining a part candidate set; in the part candidate set, by using the shape similarity calculation method to determine whether there are bushing-like parts among the other parts assembled on each part, it is possible to avoid the influence of the shape uncertainty of conduit-like parts on the identification process, and thus achieve the successful identification of conduit-like parts; (2) The technical solution proposed by the present invention is simple and feasible, only requiring the statistics of simple information such as the surface area, patch area, and shape of each part, and the required information can be directly extracted from three-dimensional modeling software, and it can be quickly deployed and applied in aviation enterprises; (3) The catheter part recognition method proposed by the present invention is universal and has good applicability to different types of aircraft. Description of the Drawings

[0015] Figure 1 It is the overall flowchart of the present invention.

[0016] Figure 2 It is a digital mock-up schematic diagram of a pipeline assembly in an embodiment of the present invention.

[0017] Figure 3 It is a schematic diagram for calculating the proportion of the surface area of the curved surfaces of the digital mock-ups of three types of parts in the surface area in an embodiment of the present invention.

[0018] Figure 4 It is a digital mock-up schematic diagram of the first type of pipe sleeve part in an embodiment of the present invention.

[0019] Figure 5 It is a digital mock-up schematic diagram of the second type of pipe sleeve part in an embodiment of the present invention.

[0020] Figure 6 It is a digital mock-up schematic diagram of the third type of pipe sleeve part in an embodiment of the present invention.

[0021] Figure 7 It is a digital mock-up schematic diagram of the fourth type of pipe sleeve part in an embodiment of the present invention. Detailed Embodiments

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can also be directly connected, or indirectly connected through an intermediate medium, and can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0024] Embodiment 1: This embodiment provides a method for recognizing aircraft catheter parts, which is carried out by means of multi-level screening and comparison recognition, with reference to Figure 1 , and specifically includes the following steps: Step S1: Traverse all the part digital models in the aircraft, and add the part digital models with the keyword "pipe" in their names to the first-level part set ; Step S2: For each part digital model in , further filter the first-level part set according to the part geometric characteristics (the proportion of the sum of the surface areas of the curved surfaces in the part in the surface area), and add the part digital models that meet the requirements in to the second-level set ; Step S3: Obtain the shape vectors of each part digital model in and its assembled part digital models; Step S4: Construct a data set of pipe sleeve part digital models and extract their shape vectors; Step S5: Determine whether the assembled part digital models with connection relationships with each part digital model in are pipe sleeve parts, so as to realize the identification of conduit parts.

[0025] In this embodiment, the pipe sleeve part refers to a standard part connected to both ends of the conduit part, which is usually used to connect the conduit part with other parts. The pipe sleeve part and the conduit part together form Figure 2 the pipeline shown.

[0026] Embodiment 2: This embodiment is further extended on the basis of Embodiment 1. The specific steps of further filtering the first-level part set according to the part geometric characteristics in step S2 include: Step S21: Obtain all the surface elements of each part digital model in , and divide them into four types: plane, cylindrical surface, free-form surface, and other curved surfaces in turn; Step S22: Statistically calculate the sum of the surface areas of each type of surface in the part digital model, and record them as , , and respectively, where , are recorded as the specific curved surface area, and , are recorded as the other surface area; Step S23: Calculate the ratio of the specific curved surface area to the other surface area. The calculation method is: ; Step S24: Set the curved surface proportion threshold = 0.95. If the part digital model satisfies , then add it to the second-level set ; In Figure 3 If the digital models of the last two parts involved both meet the conditions, add them to the secondary set ; Step S25, traverse each digital model of the parts in , and repeat steps S21 - S24 to obtain the final

[0027] The specific steps to obtain the shape vectors of the digital models of the parts and their assembled part digital models in Step S31, for a certain digital model of a part in p , obtain the set of assembled part models p that have a connection relationship with the digital model of the part , where represents the i th assembled part digital model that is connected to the digital model of the part p , , n represents the number of assembled part digital models that have a connection relationship with the digital model of the part p ; Step S32, by randomly sampling points on the surfaces of the digital model of the part p and each assembled part digital model in the set , and using the shape distribution algorithm to obtain the shape information of the digital model of the part p and each assembled part digital model in the set , and describe it as a k - dimensional shape vector, denoted as: , where ; Step S33, traverse all the digital models of the parts in

[0028] In step S31, through computer - aided interference checking, further obtain the set of assembled part digital models p that have a connection relationship with the digital model of the part .

[0029] The specific steps to construct the data set of the digital models of the bushing parts and extract their shape vectors in step S4 are as follows: Step S41, the process personnel sort out all types of bushing parts in the aircraft according to the actual situation, referring to Figure 4 , Figure 5 , Figure 6 and Figure 7 as four different types of bushing parts; Step S42: By randomly sampling points on the digital mock-up surfaces of the sleeve parts in each category and using the shape distribution algorithm to obtain the shape information of the digital mock-ups of the sleeve parts in each category, and describing it as k a dimensional shape vector, denoted as , where , and adding it to the set GT .

[0030] The specific steps for determining whether the assembly part digital mock-up that has a connection relationship with each part digital mock-up in is a sleeve part in step S5 are as follows: Step S51: For a certain part digital mock-up in p , assume is the assembly part digital mock-up that has a connection relationship with p , is the shape vector of the assembly part digital mock-up q , calculate the similarity q between the assembly part digital mock-up GT and the sleeve part digital mock-up in : ; where represents the shape vector of a certain type of sleeve part digital mock-up; is the value of the i th dimension in the vector, is i the value of the th dimension in the vector; Step S52: Set a similarity threshold , assume there is an assembly part digital mock-up , and at the same time there is a p such that Figure 2 , then it is considered that the part digital mock-up p is a conduit part; for example, in the

[0031] Step S53: Traverse , repeat steps S51 - S52 to complete the identification of all conduit parts of the aircraft.

[0032] This implementation example shows that the aircraft conduit part identification method proposed by the present invention can be used for the rapid identification of all conduit parts in aircraft products and can achieve good results.

[0033] The above are only the preferred embodiments of the present invention, and do 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 shall fall within the protection scope of the present invention.

Claims

1. A method for identifying aircraft duct parts, characterized in that, It is carried out by means of multi-level screening and comparison recognition, specifically including the following steps: Step S1: Traverse all the part digital models in the aircraft, and add the part digital models whose names contain the keyword "pipe" to the first-level part set ; Step S2: Further filter the first-level part set according to the geometric characteristics of the parts and add the digital models of the parts that meet the requirements in to the second-level set ; Step S3, obtain the shape vectors of the digital models of each part and the digital models of its assembled parts in Step S4: Construct a digital model data set of bushing parts and extract their shape vectors; Step S5: Determine whether the assembly part digital model that has a connection relationship with each part digital model in is a bushing part, so as to realize the identification of conduit parts.

2. The method for identifying aircraft duct parts according to claim 1, characterized in that, In the step S2, the specific steps for further filtering the first-level part set according to the part geometric characteristics include: Step S21, obtain all the surface elements of the digital models of each part in, and classify them into four types: plane, cylindrical surface, free-form surface, and other surfaces in sequence; Step S22: Calculate the sum of the surface areas of each type of surface in the part digital model, and denote them as , , and , where , is denoted as the specific curved surface area, , is denoted as the area of other surfaces; Step S23, calculate the ratio of the area of the specific surface to the areas of other surfaces , and the calculation method is as follows: ； Step S24, set the threshold of the surface proportion , if the part digital model meets , then add it to the secondary set ; Step S25, traverse all the part digital models in, and repeat Step S21 - Step S24 to obtain the final .

3. The aircraft duct part recognition method according to claim 2, wherein The specific steps of obtaining the shape vectors of the digital models of each part and the digital models of its assembled parts in Step S31. For the part digital model in p , obtain the set of assembly parts p that have a connection relationship with the part digital model . Among them, represents the i th assembly part digital model connected to the part digital model p , , n represents the number of assembly part digital models that have a connection relationship with the part digital model p . Step S32: Obtain the part digital model p and the set of the shape information of each assembled part digital model in, and describe it as a k-dimensional shape vector, denoted as: , where ; Step S33, traverse all the part digital models in, and repeat Step S31 - Step S32 until the traversal is completed.

4. The identification method of an aircraft duct part according to claim 3, characterized in that In the step S31, through computer-aided interference inspection, the digital model of the parts p and the set of digital models of the assembled parts having a connection relationship are obtained.

5. The method for identifying aircraft duct parts according to claim 3, characterized in that, In the step S32, by randomly sampling points on the surfaces of the part digital models p and the set of the assembly part digital models, and using the shape distribution algorithm to obtain the shape information of the part digital model and its assembly part digital models.

6. The identification method of an aircraft duct part according to claim 3, wherein The specific steps of constructing a digital model data set of bushing parts and extracting their shape vectors in step S4 are as follows: Step S41: Sort out all categories of bushing parts in the aircraft; Step S42: Obtain the shape information of the digital mock-ups of the bushing parts for each category and describe it as k a dimensional shape vector, denoted as , where , and add it to the set GT .

7. A method for identifying aircraft duct parts according to claim 6, characterized in that, In step S42, random point sampling is carried out on the surface of the digital models of bushing parts in each category, and the shape distribution algorithm is used to obtain the shape information of the digital models of each bushing part.

8. The identification method of an aircraft duct part according to claim 6, characterized in that, In the step S5, the specific steps for determining whether the digital model of the assembled part that has a connection relationship with each part digital model in is the digital model of the bushing part are as follows: Step S51. For the part digital model p , assume is the assembly part digital model having a connection relationship with the part digital model p . Let be the shape vector of the assembly part digital model q . Calculate the similarity between the assembly part digital model q and the sleeve part digital model in GT : ​​ ； Among them, represents the shape vector of a class of sleeve part digital models; is the value of the i -th dimension in the vector, is the value of the i -th dimension in the vector; Step S52: Set the similarity threshold Suppose there is a digital model of an assembled part and there is also a such that then the digital model of the part p is considered to be the catheter part; Step S53, traverse , repeat Step S51 - Step S52 to complete the recognition of all duct parts of the aircraft.

Images