A method, system, device and medium for classifying installation types of finished products based on typical structural features

Abstract

The application relates to the technical field of aircraft manufacturing, in particular to a finished product installation type classification method, system, equipment and medium based on typical structure characteristics; the method first analyzes and extracts structured description content according to an acquired STG model; secondly, a finished product installation classification rule library is constructed according to the structured description content, and the finished product is preliminarily classified according to acquired finished product shape characteristics to obtain a first classification result; then, a second classification result is obtained according to acquired finished product basic characteristics; finally, a third classification result is obtained according to acquired finished product connection characteristics, and a final classification result is obtained according to the third classification result, so that the finished product installation type classification based on the typical structure characteristics is realized, and the accurate and clear judgment on the finished product category is completed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aircraft manufacturing, in particular to a finished product installation type classification method, system, device and medium based on typical structural features. BACKGROUND

[0002] The process procedure is a production process document prepared by the process department according to design requirements, process technical requirements and quality requirements. The process procedure not only includes process information, but also production information and quality information, and guides workers to operate the specific work instructions of the specified assembly process flow, including operation instructions, processes, assembly timing, change records and other information.

[0003] At present, the process procedure is still mainly prepared manually, and the cycle of process procedure preparation work is long, and the standardization of process procedure preparation result is poor, and even some quality errors may occur. Therefore, the automatic preparation of the process procedure becomes an urgent problem to be solved in aircraft development and production, and the automatic identification of aircraft parts and features is an important factor restricting the automatic preparation and planning of the process procedure.

[0004] The manufacturing and installation of an aircraft is a complex and precise process involving multiple stages from design to final delivery, mainly including conceptual design, detailed design, simulation and testing, part manufacturing, assembly and integration, internal installation, testing and verification, delivery and acceptance, etc.

[0005] Due to the complexity of the aircraft, the number of parts is large, and each part is provided by different manufacturers, so the types of finished products are various and the installation positions are different. Therefore, accurate identification and classification of the types of finished products play an important role in the automatic preparation / planning of the process procedure and the accuracy of the process procedure preparation. SUMMARY

[0006] The present application is directed to the problem that the existing classification method is prone to cause the misinstallation and missing installation of finished products, and proposes a finished product installation type classification method, system, device and medium based on typical structural features. The method first analyzes and extracts structured description content according to the obtained STG model; secondly, constructs a finished product installation classification rule library according to the structured description content, and preliminarily classifies the finished products according to the obtained finished product shape features to obtain a first classification result; then obtains a second classification result according to the obtained finished product basic features; finally, obtains a third classification result according to the obtained finished product connection features, and judges the final classification result according to the third classification result, realizes the finished product installation type classification based on typical structural features, and completes the accurate and clear judgment of the types of finished products.

[0007] The present application specifically realizes the following contents:

[0008] The method for classifying the installation type of a finished product based on typical structural features comprises the following steps: first, obtaining structured description content by analyzing the STG model; second, constructing a finished product installation classification rule library based on the structured description content, and preliminarily classifying the finished product based on the obtained shape features of the finished product to obtain a first classification result; third, obtaining a second classification result based on the obtained basic features of the finished product; and fourth, obtaining a third classification result based on the obtained connection features of the finished product, and determining a final classification result based on the third classification result.

[0009] To better realize the present application, the method for classifying the installation type of a finished product based on typical structural features further comprises the following steps:

[0010] Step S1: obtaining structured description content by analyzing an STG model, wherein the STG model is a model with local features of a three-dimensional model as vertices and an adjacency relationship between the local features as edges;

[0011] Step S2: constructing a finished product installation classification rule library based on the structured description content, and preliminarily classifying the finished product based on an obtained finished product model image to obtain a first classification result;

[0012] Step S3: classifying the preliminarily classified finished product based on the obtained basic attribute features of the finished product to obtain a second classification result;

[0013] Step S4: classifying to obtain a third classification result based on the obtained connection relationship features of the finished product and the second classification result, and determining a final classification result based on the third classification result.

[0014] To better realize the present application, the step S1 further comprises the following steps:

[0015] Step S11: obtaining three-dimensional coordinate information of the finished product by calling a model extraction technique based on the obtained STG model, and performing normalization display;

[0016] Step S12: obtaining structured description content by analyzing basic attribute information and shape data of the STG model based on the obtained STG model, wherein the basic attribute information comprises finished product coding, finished product name, and finished product connection relationship, and the shape data is stored in the form of an image.

[0017] To better realize the present application, the step S2 further comprises the following steps: calling a recognition algorithm to learn the shape data, and preliminarily classifying the finished product to obtain a first classification result.

[0018] To better realize the present application, the step S4 further comprises the following steps:

[0019] Step S41: Classify the finished product connection relationship characteristics and the second classification result to obtain a third classification result; the connection relationship characteristics include connection mode and connection key materials;

[0020] Step S42: Obtain the final classification result based on the third classification result. If the number of repeated results is greater than 2, retain the third classification result as the final classification result of the finished product. Otherwise, use the second classification result as the final classification result of the finished product.

[0021] In order to better implement the present invention, further, step S41 specifically includes the following steps:

[0022] Step S411: describing the name and shape information of the obtained finished product;

[0023] Step S412: using a double interference inspection method to obtain the connection relationship between parts and establish an adjacency model;

[0024] Step S413: Calculating a connection relationship feature set that matches the finished product installation classification rule base based on the name and shape information of the finished product;

[0025] Step S414: Using the matching relationship as the search space, obtain the connection mode and key connection materials;

[0026] Step S415: Classify the connection method, key connection materials, and the second classification result to obtain a third classification result;

[0027] In order to better implement the present invention, further, step S411 specifically includes the following steps:

[0028] Step S4111: Initialize the part category set;

[0029] Step S4112: Get the name id of the model file corresponding to part p p ;

[0030] Step S4113: Determine the part category c of part p p Does it belong to the category set? If not, call the shape distribution algorithm to describe the shape information of part p as a k-dimensional shape vector s p , and create a part descriptor <id p , s p >, take part p as the new part category c p Add to category collection;

[0031] Step S4114: If the name id of the part model file p =id q , id q ∈c q, then create a part descriptor <id q , s q >, id q is the name of the model file corresponding to part q, c q is the part category of part q;

[0032] Step S4115: Repeat steps S4112 to S4114 until all parts are traversed.

[0033] In order to better implement the present invention, further, step S412 specifically includes the following steps:

[0034] Step S4121: Obtain assembly I from the finished product assembly classification rule library, initialize the m×m dimensional adjacency matrix G, and calculate the AABB bounding box R of the parts; where m is the number of parts in the complex product model A to be identified;

[0035] Step S4122: Select part p' from assembly I and initialize set S;

[0036] Step S4123: Randomly obtain part q from assembly I. If the bounding box R p’ ∩R q ≠ , then add part q to the set S, R p’ is the AABB bounding box of part p', R q is the AABB bounding box of part q;

[0037] Step S4124: Based on the part q in the set S, call the octree interference detection algorithm to detect the spatial interference between the part p' and the part q. If the detection result is contact or interference, then the adjacency matrix G p’q =1;

[0038] Step S4125: Repeat steps S4122 to S4124 until all parts of assembly I are traversed.

[0039] In order to better implement the present invention, further, step S413 specifically includes the following steps:

[0040] Step S4131: Initialize the matching parts set Com p ;

[0041] Step S4132: The name of the finished product model A to be identified is id p Add the parts to the matching parts collection Com p ;

[0042] Step S4133: Calculate the shape vector s of part p p The shape vector s of the part in the i-th category in the part category set Ci similarity;

[0043] Step S4134: Based on the set shape similarity threshold, determine whether the similarity is greater than or equal to the shape similarity threshold. If so, add all parts in the i-th category to the matching parts set Com p .

[0044] In order to better implement the present invention, further, the shape vector s of the part p is calculated in step S4133 p The shape vector s of the part in the i-th category in the part category set C i The formula for the similarity is:

[0045] ;

[0046] Among them, s p Represents the shape vector of part p, s q is the shape vector of part q in the i-th category, k represents the dimension of the k-dimensional shape vector of the part, l j p Represents the j-th shape vector of part p, l j q Represents the j-th shape vector of part q.

[0047] In order to better implement the present invention, the specific operation of step S4134 is as follows: according to the set shape similarity threshold , if the similarity sim(s p , s i )≥ , then add all parts in the i-th category to the matching parts set Com p , s i is the shape vector of the parts in the i-th category.

[0048] In order to better implement the present invention, the specific operation of step S414 is as follows: using the part matching relationship as the search space, calling the Ullmann subgraph matching algorithm in the adjacency matrix G of the finished product model A to be identified A Search the adjacency matrix G of assembly I I The subgraphs with the same structure are matched to identify the connection mode and key connection materials.

[0049] Further, in order to better realize the present application, the process of establishing the STG model in step S1 is: first, a vertex adjacency graph is established according to the topological relationship between the obtained B-rep data geometric elements; second, a local feature of the entity model is obtained by searching a maximum clique in the adjacency graph and taking the geometric region corresponding to the maximum clique as the local feature; then, a statistical method is called to convert the shape information of the local feature into a feature vector, form a feature space, and suppress the local features by using an unsupervised learning algorithm; finally, the STG model is established, taking the local features as vertices and the adjacency relationship between the local features as edges.

[0050] Based on the above-mentioned product installation type classification method based on typical structural features, in order to better realize the present application, further, a product installation type classification system based on typical structural features is proposed, which is used to execute the above-mentioned product installation type classification method based on typical structural features; comprising an analysis extraction unit, a preliminary classification unit, an attribute classification unit, and a classification output unit.

[0051] The analysis extraction unit is used to analyze and extract structured description content according to the obtained STG model.

[0052] The preliminary classification unit is used to construct a product installation classification rule library according to the structured description content, and preliminarily classify the product according to the obtained product model image to obtain a first classification result.

[0053] The attribute classification unit is used to classify the preliminarily classified product according to the obtained product basic attribute information to obtain a second classification result.

[0054] The classification output unit is used to classify to obtain a third classification result according to the obtained product connection relationship feature and the second classification result, and determine a final classification result according to the third classification result.

[0055] Based on the above-mentioned product installation type classification method based on typical structural features, in order to better realize the present application, further, an electronic device is proposed, comprising a memory and a processor; the memory stores a computer program; when the computer program is executed on the processor, the above-mentioned product installation type classification method based on typical structural features is realized.

[0056] Based on the above-mentioned product installation type classification method based on typical structural features, in order to better realize the present application, further, a computer readable storage medium is proposed, the computer storage medium stores computer instructions; when the computer instructions are executed on the above-mentioned electronic device, the above-mentioned product installation type classification method based on typical structural features is realized.

[0057] The present application has the following beneficial effects:

[0058] (1) The application constructs a finished product installation type atlas network, realizes finished product installation type classification based on typical structure characteristics, efficiently identifies the category of the finished product, and solves the problem of easy missed or wrong installation of the finished product.

[0059] (2) The application realizes finished product installation type classification based on typical structure characteristics based on clustering and other reasoning logic technologies, and completes accurate and clear judgment on the category of the finished product. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 A flowchart diagram of the finished product installation type classification method based on typical structure characteristics provided by the application is shown. DETAILED DESCRIPTION

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the application, the technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. It should be understood that the described embodiments are only a part of the embodiments of the application, but not all the embodiments, and therefore should not be regarded as limiting the protection scope. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the protection scope of the application.

[0062] In the description of the application, it should be noted that unless otherwise explicitly specified and limited, the terms "set", "connected", "connected" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be connected inside two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the application can be understood according to the specific circumstances.

[0063] Embodiment 1:

[0064] In this embodiment, first, the structured description content is analyzed and extracted according to the obtained STG model; second, the finished product installation classification rule library is constructed according to the structured description content, and the finished product is preliminarily classified according to the obtained finished product shape characteristics to obtain a first classification result; then, a second classification result is obtained according to the obtained finished product basic characteristics; finally, a third classification result is obtained according to the obtained finished product connection characteristics, and a final classification result is obtained according to the third classification result.

[0065] Working principle: The embodiment realizes the finished product installation type classification method based on typical structure characteristics by constructing the finished product installation type atlas network, efficiently identifies the category of the finished product, and solves the problem of easy missed or wrong installation of the finished product.

[0066] Embodiment 2:

[0067] The embodiment is described in the form of steps based on the above embodiment 1, as shown in the figure, the product installation type classification method based on typical structural features specifically includes the following steps: Figure 1

[0068] Step S1: According to the obtained STG model, analyze and extract the structured description content.

[0069] The step S1 specifically includes the following steps:

[0070] Step S11: According to the obtained STG model, call the model extraction technology to obtain the three-dimensional coordinate information of the product, and perform normalization display;

[0071] Step S12: According to the obtained STG model, extract the basic attribute information and shape data of the STG model, and analyze to obtain the structured description content; the basic attribute information includes product code, product name, and product connection relationship; the shape data is stored in the form of image.

[0072] Step S2: Construct a product installation classification rule library according to the structured description content, and preliminarily classify the product according to the obtained product model image to obtain a first classification result.

[0073] The step S2 specifically includes the following steps:

[0074] Step S21: Construct a product installation classification rule library according to the structured description content;

[0075] Step S22: Call the recognition algorithm to learn the shape data, and preliminarily classify the product to obtain a first classification result.

[0076] Step S3: According to the obtained product basic attribute features, classify the preliminarily classified product to obtain a second classification result.

[0077] The specific operation of the step S3 is: according to the obtained product basic attribute information, call the semantic analysis method to classify the preliminarily classified product to obtain a second classification result.

[0078] Step S4: According to the obtained product connection relationship features and the second classification result, classify to obtain a third classification result, and judge the final classification result according to the third classification result.

[0079] The step S4 specifically includes the following steps:

[0080] Step S41: According to the obtained product connection relationship features and the second classification result, classify to obtain a third classification result; the connection relationship features include connection mode and connection key material; ​

[0081] The connection mode and the acquisition mode of the connection key material in step S41 are as follows: firstly, the name and shape information of the acquired finished product are described; secondly, a double interference verification method is used to acquire the connection relationship between parts, and an adjacency model is established; then, the connection relationship feature set matched with the finished product installation classification rule library is calculated according to the name and shape information of the finished product; and finally, the matching relationship is taken as a search space to obtain the connection mode and the connection key material.

[0082] Step S42: The final classification result is obtained according to the third classification result. If the repeatedly appearing result is greater than 2, the third classification result is retained as the final classification result of the finished product; otherwise, the second classification result is taken as the final classification result of the finished product.

[0083] Working principle: In this embodiment, a finished product installation type graph network is constructed by means of semantic recognition, model extraction, graph construction and the like, and a finished product installation type classification method based on typical structural features is realized based on clustering and other reasoning logic technologies, so that an accurate and clear judgment on the finished product category is completed.

[0084] The other parts of this embodiment are the same as those of the above-described embodiment 1, and thus will not be described again.

[0085] Embodiment 3

[0086] This embodiment is based on any one of the above-described embodiments 1-2 and is described in detail by taking a specific embodiment.

[0087] Step S1: The existing STG three-dimensional model is processed by means of vocabulary analysis, grammar analysis, syntax analysis and model extraction technology, and structured description content is acquired.

[0088] The three-dimensional model is an important carrier of bearing aircraft assembly information, which includes important information such as the installation position of the finished product, the basic attribute information of the finished product, and the shape of the finished product; the basic attribute information of the finished product is the finished product code and the finished product name.

[0089] Through the model extraction technology, the three-dimensional coordinate information of the finished product is acquired, which is stored in the form of (X, Y, Z) and normalized to obtain the coordinate value (X1, Y1, Z1).

[0090] At the same time, the basic attribute information in the model is extracted, including the finished product code, the finished product name, the finished product connection relationship and other data. The shape data is stored in the form of a picture.

[0091] In the embodiment, the STG model obtained in step S1 is first established according to the topological relationship between vertices, faces and other geometric elements in the B-rep data; a vertex adjacency graph is established; a local feature is obtained by searching a maximum clique in the graph and taking the corresponding geometric region as the local feature; shape information of the local feature is converted into a feature vector based on a statistical method to form a feature space, and an unsupervised learning algorithm is used for local feature difference suppression, so that local features with similar shapes have the same code; and finally, an STG model is established with local features as vertices and adjacency relationships as edges.

[0092] Step S2: Constructing a finished product installation classification rule library, and completing the judgment of the finished product classification based on the rule library.

[0093] The finished product installation classification rule library defines the classification principle of the finished product.

[0094] Through machine learning, the model pictures of various finished products are learned, and the finished products are roughly classified according to the classification algorithm to obtain a classification result Rp {Rp1, Rp2, Rp3…Rpn}, where Rpi represents the classification result of the ith finished product.

[0095] Step S3: According to the basic attribute information, the finished products are classified according to the information. This is a classification according to the characteristics of the finished product, mainly based on the basic information, and classified by semantic recognition method such as code and name. A classification result RL {RL1, RL2, RL3…RLn} is obtained, where Rli represents the classification result of the ith finished product. For example, a certain finished product can be roughly classified as {mechanical finished product, electrical finished product…}, the attributes of the mechanical finished product include support, pipeline, rack, strut, brace, etc., and the electrical finished product includes plug, rear accessory and sensor electrical finished product.

[0096] Step S4: According to the connection relationship characteristics, the finished products are classified. The finished products are classified according to the connection method and the different connection key materials. A classification result RR {RR1, RR2, RR3…RRn} is obtained, where RRi represents the classification result of the ith finished product. {mechanical finished product, electrical finished product…}, and the connection object of the mechanical finished product is a part, and the connection object of the electrical finished product is a cable plug.

[0097] Through the above three dimensions, three classification results of a finished product are obtained, wherein the logic for obtaining the final result is that if the repeated result is greater than 2, the result is retained as the final classification result of the finished product. Otherwise, the classification result of the basic attribute information is used as the standard. Through the above classification, the finished product can be roughly classified as a mechanical finished product and an electrical finished product.

[0098] Among them, the method for obtaining the connection relationship features in step S4 of this embodiment is: constructing part categories based on the finished product names, and extracting the shape information of the parts in each category; using the predefined finished product installation classification rule library as input, comprehensively analyzing the category information and shape information to quickly judge the part matching relationship, and on this basis, using the graph matching algorithm to take the connection relationship and the connection key materials as typical structures to achieve effective identification of the connection relationship and the connection key materials.

[0099] The rest of this embodiment is the same as any of the above-mentioned embodiments 1 and 2, and thus will not be described in detail.

[0100] Example 4:

[0101] This embodiment, based on any one of the above embodiments 1 to 3, describes step S4 in detail using a specific embodiment.

[0102] The step S4 specifically includes the following steps:

[0103] Step S41: classifying the finished product connection relationship characteristics and the second classification result to obtain a third classification result; the connection relationship characteristics include connection mode and connection key materials.

[0104] The step S41 specifically includes the following steps:

[0105] Step S411: Describe the name and shape information of the obtained finished product;

[0106] The step S411 specifically includes the following steps:

[0107] Step S4111: Initialize the part category set;

[0108] Step S4112: Get the name id of the model file corresponding to part p p ;

[0109] Step S4113: Determine the part category c of part p p Does it belong to the category set? If not, call the shape distribution algorithm to describe the shape information of part p as a k-dimensional shape vector s p , and create a part descriptor <id p , s p >, take part p as the new part category c p Add to category collection;

[0110] Step S4114: If the name id of the part model file p =id q , id q ∈c q , then create a part descriptor <id qs q id q is the name of the part q corresponding model file, c q is the part category of the part q;

[0111] Step S4115: Repeat steps S4112-S4114 until all parts are traversed.

[0112] Step S412: Obtain the connection relationship between parts by using a double interference checking method, and establish an adjacency model.

[0113] The step S412 specifically comprises the following steps:

[0114] Step S4121: Obtain the assembly body I from the finished product installation classification rule library, initialize the m*m dimensional adjacency matrix G, and calculate the AABB bounding box R of the part; wherein m is the number of parts in the complex product model A to be identified;

[0115] Step S4122: Select a part p' from the assembly body I, and initialize a set S;

[0116] Step S4123: Randomly obtain a part q from the assembly body I, if the bounding box R p’ ∩R q ≠ , then add the part q to the set S, R p’ is the AABB bounding box of the part p', and R q is the AABB bounding box of the part q;

[0117] Step S4124: According to the part q in the set S, call the octree interference checking algorithm to check the spatial interference between the part p' and the part q, if the checking result is contact or interference, then the adjacency matrix G p’q =1;

[0118] Step S4125: Repeat steps S4122-S4124 until all parts of the assembly body I are traversed.

[0119] Step S413: Calculate the connection relationship feature set matched with the finished product installation classification rule library according to the name and shape information of the finished product.

[0120] The step S413 specifically comprises the following steps:

[0121] Step S4131: Initialize the matching part set Com p ;

[0122] Step S4132: Add the part with the name id p in the to-be-identified finished product model A to the matching part set Com p ;

[0123] Step S4133: calculating the shape vector s of the part p p sim(s i , s p ) represents the similarity of the shape vector s of the part p and the shape vector s of the part q in the i-th category in the part category set C.

[0124] The specific operation of the step S4133 is as follows:

[0125] ;

[0126] wherein s p represents the shape vector of the part p, s q is the shape vector of the part q in the i-th category, k represents the dimension of the shape vector of the part k, and l j p represents the j-th dimension shape vector of the part p, and l j q represents the j-th dimension shape vector of the part q.

[0127] Step S4134: determining whether the similarity is greater than or equal to the shape similarity threshold value according to the set shape similarity threshold value, and if so, adding all parts in the i-th category to the matched part set Com p .

[0128] The specific operation of the step S4134 is as follows: determining whether the similarity sim(s p , s i ) is greater than or equal to the shape similarity threshold value according to the set shape similarity threshold value , if the similarity sim(s p , s i ) ≥ , then adding all parts in the i-th category to the matched part set Com p , s i is the shape vector of the part in the i-th category, i.e. Com p = Com p ∪{c q |sim(s p , s q ) ≥ }, wherein 1≤i≤n and q≠p.

[0129] Step S414: taking the matched relationship as the search space to obtain the connection mode and the connection key material.

[0130] The specific operation of the step S414 is as follows: taking the part matching relationship as the search space, and calling the Ullmann subgraph matching algorithm to search in the adjacency matrix G A of the to-be-identified finished product model A and the adjacency matrix G IThe subgraph matching result of the same structure identifies the connection mode and the connection key material.

[0131] Step S415: According to the connection mode and the connection key material, the third classification result is classified in combination with the second classification result.

[0132] Step S42: According to the third classification result, the final classification result is determined. If the repeated result is greater than 2, the third classification result is retained as the final classification result of the finished product, otherwise, the second classification result classified according to the basic attribute characteristics is taken as the final classification result of the finished product.

[0133] The other parts of this embodiment are the same as any one of the above-mentioned embodiments 1-3, and will not be repeated.

[0134] Embodiment 5:

[0135] This embodiment is based on any one of the above-mentioned embodiments 1-4, and proposes a finished product installation type classification system based on typical structure characteristics, which is used to execute the above-mentioned finished product installation type classification method based on typical structure characteristics; including an analysis extraction unit, a preliminary classification unit, an attribute classification unit, and a classification output unit.

[0136] The analysis extraction unit is configured to analyze and extract the structured description content according to the obtained STG model.

[0137] The preliminary classification unit constructs a finished product installation classification rule library according to the structured description content, and preliminarily classifies the finished product according to the obtained finished product model image to obtain a first classification result.

[0138] The attribute classification unit is configured to classify the preliminarily classified finished product according to the obtained basic attribute information of the finished product to obtain a second classification result.

[0139] The classification output unit is configured to classify to obtain a third classification result according to the obtained finished product connection relationship characteristics and the second classification result, and to determine a final classification result according to the third classification result.

[0140] This embodiment also proposes an electronic device including a memory and a processor; the memory stores a computer program; when the computer program is executed on the processor, the above-mentioned finished product installation type classification method based on typical structure characteristics is realized.

[0141] This embodiment also proposes a computer readable storage medium, and the computer storage medium stores computer instructions; when the computer instructions are executed on the above-mentioned electronic device, the above-mentioned finished product installation type classification method based on typical structure characteristics is realized.

[0142] Other parts of this embodiment are the same as any one of Embodiment 1 to Embodiment 4 described above, and thus will not be described again.

[0143] The processor involved in the embodiments of the present application can be a chip. For example, it can be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), a central processor unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD) or other integrated chip.

[0144] The memory, as to which embodiments of the present application are directed, can be volatile memory or nonvolatile memory, or can include both volatile and nonvolatile memory. Among them, the nonvolatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically EPROM (EEPROM), or flash memory. The volatile memory can be random access memory (RAM) used as an external cache. By way of example, and not limitation, many forms of RAM can be used, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct Rambus RAM (DR RAM). It should be noted that the memory of the system and method described herein is intended to include, but not be limited to, these and any other suitable types of memory.

[0145] It should be understood that the size of the sequence number of each process described above does not mean the order of execution in various embodiments of the present application, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0146] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be realized in electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0147] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and module described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0148] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the embodiments of the device described above are merely schematic. For example, the division of the modules is merely logical function division. There can be another division manner for the actual implementation. For example, a plurality of modules or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or modules, and can be in electrical, mechanical or other forms.

[0149] The modules illustrated as separate components can or can not be physically separate, and the components illustrated as modules can or can not be physical modules, i.e., can be located in one device or distributed on a plurality of devices. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0150] In addition, each functional module in the embodiments of the present application can be integrated in one device, or each module can be physically present alone, or two or more modules can be integrated in one device.

[0151] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium that can be accessed by a computer or include one or more data storage devices such as servers, data centers, etc. integrated with the medium. The available medium can be magnetic medium (such as floppy disk, hard disk, magnetic tape), optical medium (such as DVD), or semiconductor medium (such as solid state disk (SSD)) and the like.

[0152] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for classifying finished product installation types based on typical structural features, characterized in that: The following steps are involved: Step S1: Analyze and extract the structured description content based on the acquired STG model, wherein the STG model is a model established with local features of the three-dimensional model as vertices and adjacency relationships between local features as edges; Step S2: constructing a finished product installation classification rule base based on the structured description content, and preliminarily classifying the finished products based on the acquired finished product model images to obtain a first classification result; Step S3: Based on the acquired basic attribute characteristics of the finished product, a semantic analysis method is used to classify the finished product after the preliminary classification to obtain a second classification result; Step S4: classifying the finished product connection relationship features and the second classification result to obtain a third classification result, and determining the final classification result based on the third classification result; The establishment process of the STG model in step S1 is as follows: first, a vertex adjacency graph is established based on the topological relationship between the geometric elements of the acquired B-rep data; second, the maximum clique in the adjacency graph is searched and the geometric area corresponding to the maximum clique is used as the local feature of the solid model; then, the shape information of the local feature is converted into a feature vector by using a statistical method to form a feature space, and an unsupervised learning algorithm is used to differentially suppress the local features; finally, an STG model is established with local features as vertices and the adjacency relationship between local features as edges; Step S1 specifically includes the following steps: Step S11: Based on the obtained STG model, the model extraction technology is called to obtain the three-dimensional coordinate information of the finished product and perform normalized display; Step S12: Based on the acquired STG model, extract the basic attribute information and appearance data of the STG model, and analyze to obtain structured description content; the basic attribute information includes finished product code, finished product name, and finished product connection relationship; the appearance data is stored in the form of an image.

2. A method for classifying finished product installation types based on typical structural features according to claim 1, characterized in that: In step S2, a recognition algorithm is called to learn the appearance data, and the finished products are preliminarily classified to obtain a first classification result.

3. The method for classifying the installation type of finished products based on typical structural features according to claim 1, characterized in that: The step S4 specifically includes the following steps: Step S41: Classify the finished product connection relationship characteristics and the second classification result to obtain a third classification result; the connection relationship characteristics include connection mode and connection key materials; Step S42: Obtain the final classification result based on the third classification result. If the number of repeated results is greater than 2, retain the third classification result as the final classification result of the finished product. Otherwise, use the second classification result as the final classification result of the finished product.

4. A method for classifying finished product installation types based on typical structural features according to claim 3, characterized in that: The step S41 specifically includes the following steps: Step S411: describing the name and shape information of the obtained finished product; Step S412: using a double interference inspection method to obtain the connection relationship between parts and establish an adjacency model; Step S413: Calculating a connection relationship feature set that matches the finished product installation classification rule base based on the name and shape information of the finished product; Step S414: Using the matching relationship as the search space, obtain the connection mode and key connection materials; Step S415: Classify the connection method, key connection materials, and the second classification result to obtain a third classification result.

5. The method for classifying the installation type of finished products based on typical structural features according to claim 4, characterized in that: The step S411 specifically includes the following steps: Step S4111: Initialize the part category set; Step S4112: Get the name id of the model file corresponding to part p p ; Step S4113: Determine the part category c of part p p Does it belong to the category set? If not, call the shape distribution algorithm to describe the shape information of part p as a k-dimensional shape vector s p , and create a part descriptor <id p , s p >, take part p as the new part category c p Add to category collection; Step S4114: If the name id of the part model file p =id q , id q c q , then create a part descriptor <id q , s q >, id q is the name of the model file corresponding to part q, c q is the part category of part q; Step S4115: Repeat steps S4112 to S4114 until all parts are traversed.

6. A method for classifying finished product installation types based on typical structural features according to claim 5, characterized in that: The step S412 specifically includes the following steps: Step S4121: Obtain assembly I from the finished product assembly classification rule library, initialize the m×m dimensional adjacency matrix G, and calculate the AABB bounding box R of the parts; where m is the number of parts in the complex product model A to be identified; Step S4122: Select part p' from assembly I and initialize set S; Step S4123: Randomly obtain part q from assembly I. If the bounding box R p’ ∩R q ≠ , then add part q to the set S, R p’ is the AABB bounding box of part p', R q is the AABB bounding box of part q; Step S4124: For part q in set S, call the octree interference check algorithm to check the spatial interference between part p' and part q. If the test result is contact or interference, then the adjacency matrix G p’q =1; Step S4125: Repeat steps S4122 to S4124 until all parts of assembly I are traversed.

7. A method for classifying finished product installation types based on typical structural features according to claim 6, characterized in that: The step S413 specifically includes the following steps: Step S4131: Initialize the matching parts set Com p ; Step S4132: The name of the finished product model A to be identified is id p Add the parts to the matching parts collection Com p ; Step S4133: Calculate the shape vector s of part p p The shape vector s of the part in the i-th category in the part category set C i similarity; Step S4134: Based on the set shape similarity threshold, determine whether the similarity is greater than or equal to the shape similarity threshold. If so, add all parts in the i-th category to the matching parts set Com p .

8. The method for classifying the installation type of finished products based on typical structural features according to claim 7, characterized in that: The shape vector s of the part p is calculated in step S4133 p The shape vector s of the part in the i-th category in the part category set C i The formula for the similarity is: ； Among them, s p Represents the shape vector of part p, s q is the shape vector of part q in the i-th category, k represents the dimension of the k-dimensional shape vector of the part, l j p Represents the j-th shape vector of part p, l j q Represents the j-th shape vector of part q.

9. The method for classifying the installation type of finished products based on typical structural features according to claim 8, characterized in that: The specific operation of step S4134 is: according to the set shape similarity threshold , if the similarity sim(s p , s i )≥ , then add all parts in the i-th category to the matching parts set Com p , s i is the shape vector of the parts in the i-th category.

10. A method for classifying finished product installation types based on typical structural features according to claim 9, characterized in that: The specific operation of step S414 is: using the part matching relationship as the search space, calling the Ullmann subgraph matching algorithm on the adjacency matrix G of the finished product model A to be identified A Search the adjacency matrix G of assembly I I The subgraphs with the same structure are matched to identify the connection mode and key connection materials.

11. A system for classifying finished product installation types based on typical structural features, for executing the method for classifying finished product installation types based on typical structural features as claimed in claim 1; characterized in that: It includes analysis and extraction unit, preliminary classification unit, attribute classification unit and classification output unit; The analysis and extraction unit is used to analyze and extract the structured description content according to the acquired STG model; The preliminary classification unit constructs a finished product installation classification rule base according to the structured description content, and preliminarily classifies the finished products according to the acquired finished product model images to obtain a first classification result; The attribute classification unit is used to classify the finished products after preliminary classification based on the acquired basic attribute information of the finished products to obtain a second classification result; The classification output unit is used to classify and obtain a third classification result based on the acquired finished product connection relationship characteristics and the second classification result, and to determine and obtain a final classification result based on the third classification result.

12. An electronic device, characterized in that: The invention comprises a memory and a processor; a computer program is stored in the memory; when the computer program is executed on the processor, the method for classifying the installation type of finished products based on typical structural features as described in any one of claims 1 to 10 is implemented.

13. A computer-readable storage medium, characterized in that The computer storage medium stores computer instructions; when the computer instructions are executed on the electronic device according to claim 12, the method for classifying finished product installation types based on typical structural features according to any one of claims 1 to 10 is implemented.

Images