A process information mapping matching method and system based on a three-dimensional model of a ship

Abstract

The application discloses a kind of process information mapping matching method and system based on ship three-dimensional model, the present application is extracted to topological and geometric composite process feature by semantic association rule base to three-dimensional design model, and two-stage strategy of topological priority, geometric fine screen is used to form structured process attribute table;Through cosine similarity, process template library is matched, and process model instance is reconstructed in combination with constraint before and after process;Through bidirectional association index, traceable mapping between process and geometric entity is established;Through difference operator, change node is identified, and local incremental update is carried out to process model, not only improve the accuracy and stability of ship component process feature identification, also improve process template matching precision and reduce mismatching rate, simultaneously, bidirectional association between design model and process information is established, and accurate incremental update can be realized when design changes.

Description

Technical Field

[0001] This invention relates to the field of digital shipbuilding technology, and in particular to a method and system for mapping and matching process information based on a three-dimensional ship model. Background Technology

[0002] In existing ship 3D process template matching technology, process feature extraction is usually based on full-dimensional matching of geometric parameters without layering the topological connection relationship and geometric scale information of components. This can easily lead to mismatching of process templates or loss of semantics when there are large differences in component scale or similar geometric shapes.

[0003] Meanwhile, existing process template matching methods mostly use general distance measurement methods, which are not adaptable to the characteristics of ship processes, resulting in low process synchronization efficiency and poor data consistency.

[0004] Therefore, there is an urgent need for a technical solution that can improve the matching accuracy of ship process templates and support local incremental updates after design changes. Summary of the Invention

[0005] In view of this, the present invention provides a method and system for mapping and matching process information based on a 3D ship model. It extracts composite topological and geometric process features from the 3D design model using a semantic association rule base, and forms a structured process attribute table using a two-level strategy of topology priority and geometric refinement. It matches the process template library using cosine similarity and reconstructs process model instances by combining pre- and post-process constraints. It establishes a traceable mapping between processes and geometric entities through a bidirectional association index. It identifies change nodes using a difference operator and performs local incremental updates to the process model. This not only improves the accuracy and stability of process feature identification for ship components but also enhances the accuracy of process template matching and reduces the false matching rate. Simultaneously, it establishes a bidirectional association between the design model and process information, enabling precise incremental updates during design changes.

[0006] A method for mapping and matching process information based on a three-dimensional ship model, characterized by the following steps: S1. Based on the semantic association rule base, the process features of each component in the three-dimensional design model of the ship product are matched to obtain the structured process attribute table of each component. S2, using cosine similarity to match the structured process attribute table of the component with the standard process template in the process template library to obtain the optimal process template for each component; S3, based on the pre- and post-process constraints of each component, sorts the standard process chain of its optimal process template and binds the sorted process sequence to the component to establish a two-way association.

[0007] In a preferred embodiment, the specific steps in step S1 for matching the process features of each component in the 3D design model of the ship product based on the semantic association rule base are as follows: S11. Obtain the three-dimensional design model of the ship product and perform semantic analysis on it to identify and extract the composite process features of each component. The composite process features include topological connection features and geometric scale features. S12, Match the topological connection features of a component with the semantic association rules in the semantic association rule base to initially screen out all the process categories to which it belongs; S13, match the geometric scale features of the current component with the semantic association rules in the semantic association rule base, and refine the construction process attributes of each process category to which it belongs. S14. Generate a structured process attribute table for the component based on all process categories to which the current component belongs and the construction process attributes under each process category. S15, iterate through all components of the three-dimensional design model of the ship product, repeat steps S12-S14, and obtain the structured process attribute table of each component.

[0008] In a preferred embodiment, the semantic association rule base is used to store the association rules between the topological connectivity features and geometric scale features of components and their corresponding process categories. The topological connection characteristics of a component include the surface connection relationship, edge connection relationship, node connection relationship and assembly constraint relationship of the component; the geometric dimensional characteristics of a component include the length, width, height, wall thickness, aperture, angle, curvature, spacing and offset of the component. The process categories include cutting, assembly, welding, outfitting, and painting.

[0009] In a preferred embodiment, the specific steps in step S2 of matching the structured process attribute table of the component with the standard process template in the process template library using cosine similarity are as follows: S21, map the structured process attribute tables of all components to process attribute feature vectors respectively, and map each standard process template in the process template library to a template feature vector respectively; S22, calculate the cosine similarity between the process attribute feature vector of a component and the template feature vector of each standard process template, and sort the cosine similarity scores in descending order, and select the top N standard process templates as candidate process templates. S23, Select the best process template for the current component from the candidate process templates; S24. Traverse all components of the three-dimensional design model of the ship product, repeat steps S22 and S23, and obtain the optimal process template for each component.

[0010] In a preferred embodiment, in step S23, the standard process template with the highest cosine similarity score among the candidate process templates is taken as the best process template for the current component. Alternatively, based on the ship process verification rules, the pre- and post-process constraint verification rules, and the template validity rules, calculate the comprehensive scoring function for each candidate process template, and select the candidate process template with the highest comprehensive score as the best process template.

[0011] In a preferred embodiment, the formula for calculating the comprehensive scoring function is as follows:

[0012]

[0013] in, For the first Cosine similarity of candidate process templates; For the first The process constraint satisfaction score of each candidate process template; and These are the weighting coefficients. No. The formula for calculating the process constraint satisfaction score of each candidate process template is as follows:

[0014] in, For the first Satisfaction with ship process verification rules for each candidate process template For the first Satisfaction of pre- and post-process constraint rules for each candidate process template For the first The degree of template validity rule satisfaction for each candidate process template; The formula for calculating cosine similarity is:

[0015] in, This represents the feature vector of the component's technological attributes. For the first Template feature vectors of candidate process templates.

[0016] In a preferred embodiment, step S3, which involves sequencing the standard process chain of the optimal process template based on the pre- and post-process constraints of each component, includes the following steps: S31, Based on the pre- and post-process constraints of a certain component, its directed edge set is represented as follows:

[0017] in, Indicates the process of the current component. It must precede its process. implement; The standard process set for the optimal process template corresponding to the current component is as follows:

[0018] in, This refers to the m-th standard operation in the optimal process template, where m is the number of standard operations in the optimal process template. S32, based on the current component's directed edge set and its optimal process template's standard process set, the process sorting is modeled as a topological sorting problem on a directed acyclic graph. The sorted process sequence is represented as:

[0019] in, This is the sorted sequence of processes; S33, sort the process set The set of geometric entities of the current component Establish a bidirectional association index; S34. Traverse all components and repeat steps S31-S33. Sort the standard process chain of the optimal process template corresponding to each component and bind the sorted process sequence to the component to establish a two-way association.

[0020] In a preferred embodiment, step S4 is also included: when a local node change occurs in the three-dimensional design model of the ship product, incremental updates are performed on the changed node and the nodes within its influence range.

[0021] In a preferred embodiment, the specific steps for performing incremental updates on the changed node and the nodes within its influence range are as follows: Let the original three-dimensional design model of the ship product be... The revised 3D design model of the ship product is Then the difference operator can be expressed as:

[0022] in: Indicates the addition of new process elements; This indicates the deletion of a process element; This indicates a modification to the process elements; The set of affected nodes can be defined as:

[0023] Repeat steps S1-S3, recalculating the process only for the set of affected nodes:

[0024] in, Represents the original process model. This represents the updated process model.

[0025] A process information mapping and matching system based on a ship 3D model includes a process feature extraction module, a process template matching module, a process model reconstruction module, and a two-way indexing and version management module. The process feature extraction module is used to match the process features of each component in the three-dimensional design model of the ship product based on the semantic association rule base, and obtain the structured process attribute table of each component. The process template matching module is used to perform similarity matching between the structured process attribute table of the component and the standard process template in the process template library using cosine similarity, so as to obtain the optimal process template for each component. The process model reconstruction module is used to sort the standard process chain of the optimal process template according to the constraints between the processes of each component; The bidirectional index and version management module is used to bind the geometric entity of each component to the process sequence after sorting its best process template to establish a bidirectional association index, and to perform incremental updates and version tracking when there are local node changes in the three-dimensional design model of the ship product.

[0026] The beneficial effects of this invention are: 1. This invention extracts topological and geometric composite process features from a 3D design model using a semantic association rule base. A two-level strategy of topology priority and geometric refinement is employed to form a structured process attribute table. Cosine similarity is used to match process templates, and process model instances are reconstructed by incorporating pre- and post-process constraints. A traceable mapping between processes and geometric entities is established through a bidirectional association index. Change nodes are identified using a difference operator, enabling local incremental updates to the process model. This not only improves the accuracy and stability of process feature identification for ship components but also enhances the accuracy of process template matching and reduces the false matching rate. Furthermore, a bidirectional association is established between the design model and process information, enabling precise incremental updates during design changes.

[0027] 2. The present invention adopts a two-level process feature extraction mechanism that prioritizes topological semantics, which can effectively reduce the mismatch problem caused by size similarity, thereby improving the accuracy and stability of process feature identification of ship components.

[0028] 3. The present invention can significantly improve the accuracy of process template recognition through a template matching mechanism driven by cosine similarity, thereby improving the accuracy of process template matching and reducing the false matching rate.

[0029] 4. This invention improves the traceability between processes and geometric entities through a bidirectional association indexing mechanism, and establishes a bidirectional association between design models and process information.

[0030] 5. This invention identifies affected nodes by using a difference operator and performs recalculation and rematching only on local nodes, which can reduce the risk of redundant calculations and data inconsistencies caused by full reconstruction. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the overall process of the present invention.

[0033] Figure 2 This is a schematic diagram illustrating the generation of a structured process attribute table.

[0034] Figure 3 This is a diagram illustrating the similarity matching of standard process templates.

[0035] Figure 4 This is a schematic diagram of process model reconstruction and bidirectional association index.

[0036] Figure 5 This is a schematic diagram of incremental update driven by the difference operator. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention is described below with reference to specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0038] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0039] In the following description, suffixes such as "module," "part," or "unit" used to denote elements are used only for the convenience of the description of the invention and have no specific meaning in themselves. Therefore, "module" and "part" can be used interchangeably.

[0040] To better understand the technical solution of the present invention, the present invention will be described in detail below with reference to the accompanying drawings.

[0041] like Figures 1-5 As shown, this invention provides a method for mapping and matching process information based on a three-dimensional ship model, characterized by the following steps: S1. Based on the semantic association rule base, the process features of each component in the three-dimensional design model of the ship product are matched to obtain the structured process attribute table of each component.

[0042] like Figure 2 As shown, the specific steps for generating the structured process attribute table are as follows: S11. Obtain the three-dimensional design model of the ship product and perform semantic parsing on it to identify the geometric entities, assembly relationships and process semantic tags of the components. Extract the composite process features of each component, which include topological connection features and geometric scale features.

[0043] Topological connectivity features are used to characterize the connection relationships between components and within components, including but not limited to surface connections, edge connections, node connections, and assembly constraints.

[0044] Geometric scale features are used to characterize parameters such as component size, angle, thickness, and aperture, including but not limited to component length, width, height, wall thickness, aperture, angle, curvature, spacing, and offset.

[0045] member The characteristics of composite processes can be expressed as follows:

[0046]

[0047]

[0048] in, Represents the topological connectivity feature vector. This represents the geometric scale eigenvector.

[0049] S12, match the topological connection features of a component with the semantic association rules in the semantic association rule base to initially screen out all the process categories to which it belongs.

[0050] The semantic association rule base is used to establish the mapping relationship between the features of 3D model components and manufacturing process categories. The rules include association rules between component topological connection relationships, assembly constraint relationships and corresponding process categories. For example, based on the surface connection, edge connection, node connection relationships and assembly constraint relationships between components, the components are initially associated with major process categories such as cutting, assembly, welding, outfitting or painting.

[0051] The initial screening and extraction process can be represented as:

[0052] in, This represents the set of candidate process categories obtained after coarse screening using topological connectivity features.

[0053] Initial screening is used to classify components into one or more major process categories such as cutting, assembly, welding, outfitting, and painting.

[0054] S13, match the geometric scale features of the current component with the semantic association rules in the semantic association rule base, and refine the construction process attributes of each process category to obtain the construction process attributes of the component.

[0055] The fine screening and extraction process can be represented as:

[0056] in, This represents the structured process attribute table obtained from the final matching.

[0057] Assuming that component A is initially screened in step S12 and its process category is found to be cutting, assembly, and welding, further fine screening is conducted based on the geometric dimensional characteristics of the component. This involves matching the geometric dimensional characteristics of the component with semantic association rules in the semantic association rule base to obtain the process attributes of its cutting process (such as cutting method, plate thickness, cutting accuracy, etc.), assembly process (such as assembly method, positioning method, assembly tolerance, etc.), and welding process (such as bevel type, bevel angle, root gap, etc.).

[0058] S14. Generate a structured process attribute table for the component based on all process categories to which the current component belongs and the construction process attributes under each process category.

[0059] S15, iterate through all components of the three-dimensional design model of the ship product, repeat steps S12-S14, and obtain the structured process attribute table of each component.

[0060] S2, using cosine similarity, the structured process attribute table of the component is matched with the standard process template in the process template library to obtain the optimal process template for each component.

[0061] like Figure 3 As shown, the specific steps for matching the structured process attribute table of the component with the standard process template in the process template library include the following: S21, map the structured process attribute tables of all components to process attribute feature vectors respectively, and map each standard process template in the process template library to a template feature vector respectively.

[0062] The process template library stores various types of pre-established standard process chains. Each standard process template corresponds to a set of standard process attribute vectors and standard process chains.

[0063] The template library can be pre-built based on historical projects, process specifications, and industry experience, and can also be continuously expanded during use.

[0064] For the structured process attribute table of the component to be matched, map it into a process attribute feature vector. The first one in the process template library The template feature vector corresponding to each standard process template is denoted as . .

[0065] If the template library has a total If there are 10 candidate templates, then the candidate set can be defined as follows:

[0066] S22, Calculate the feature vector of the process attributes of a certain component. The cosine similarity between the template feature vector and each standard process template is calculated, and the cosine similarity scores are sorted in descending order. The top N standard process templates are selected as candidate process templates.

[0067] Cosine similarity is defined as follows:

[0068] Calculate the similarity for all standard process templates:

[0069] Sort by similarity in descending order and take the top... One standard process template is used as a candidate template set:

[0070] In this embodiment, let That is, take the first part of the output. A set of standard process templates were used as candidate templates.

[0071] S23, Select the best process template for the current component from the candidate process templates.

[0072] In one embodiment, the standard process template with the highest cosine similarity score among the candidate process templates can be directly used as the best process template for the current component. ,Right now

[0073] In another embodiment, the comprehensive scoring function of each candidate process template can be calculated according to the ship process verification rules, the pre- and post-process constraint verification rules, and the template validity rules, and the candidate process template with the highest comprehensive score can be taken as the best process template.

[0074] The formula for calculating the comprehensive scoring function is as follows:

[0075]

[0076] in, For the first Cosine similarity of candidate process templates; For the first The process constraint satisfaction score of each candidate process template; and These are the weighting coefficients. No. The formula for calculating the process constraint satisfaction score of each candidate process template is as follows:

[0077] in, For the first The ship process verification rule satisfaction of each candidate process template is in the range of [0, 1] and is calculated as "whether the process inspection items are satisfied / total number of inspection items". The ship process verification rules include the applicability of the process route, whether the processing / assembly is matched, whether the classification society's specifications are met, and the compatibility of materials and processes. For the first The satisfaction degree of the pre- and post-process constraint rules of each candidate process template, with a value range of [0, 1], is calculated as "the number of actually satisfied pre- and post-process constraint edges / the total number of pre- and post-process constraint edges defined in the template". The satisfaction degree of the pre- and post-process constraint rules includes whether the preceding process is completed, the reachability of the subsequent process, the process dependency relationship, and the availability of process resources. For the first The template validity rule satisfaction of each candidate process template is calculated, with a value range of [0, 1]. It is based on "number of valid processes in the template / total number of processes defined in the template". The template validity rule satisfaction includes template version validity, scope of application verification, expiration time verification, template usage frequency and reliability, etc. The best process template for the current component for:

[0078] S24. Traverse all components of the three-dimensional design model of the ship product, repeat steps S22 and S23, and obtain the optimal process template for each component.

[0079] S3, based on the pre- and post-process constraints of each component, sorts the standard process chain of its optimal process template and binds the sorted process sequence to the component to establish a two-way association.

[0080] like Figure 4 As shown, rearranging the standard process chain of the optimal process template and bidirectionally associating the rearranged process sequence with the part specifically includes the following steps: S31, based on the pre- and post-process constraints of a certain component, its directed edge set is represented as:

[0081] in, Indicates the process of the current component. It must precede its process. implement; The constraints between processes are determined by a combination of three types of constraints: structural feature constraints, assembly relationship constraints, and manufacturing constraints.

[0082] Structural feature constraints: Based on the geometric dimensions of a component, such as length, wall thickness, hole diameter, etc., the structural process rule library is matched to generate the basic process sequence (e.g., cutting first → machining then surface treatment). Assembly relationship constraints: Based on the face, edge, and node connection relationships in the topological connection features of a component, extract the sub-process chain and its sequence from the corresponding assembly process template; Manufacturing constraints: Additional constraints such as prohibiting parallel processing and forcing serial processing are added based on the material, processing accuracy, welding / painting and other special process requirements of the components.

[0083] By combining the above three types of constraints, a set of directed edges for the component is generated.

[0084] The standard process set for the optimal process template corresponding to the current component is as follows:

[0085] in, This refers to the m-th standard operation in the optimal process template, where m is the number of standard operations in the optimal process template. S32, based on the current component's directed edge set and its optimal process template's standard process set, the process sorting is modeled as a topological sorting problem on a directed acyclic graph. The sorted process sequence is represented as:

[0086] in, This is the sorted sequence of processes.

[0087] This means adjusting the standard process sequence of the optimal process template based on the set of directed edges of the current component.

[0088] Assuming the standard process chain of the optimal process template corresponding to component A is layout and blanking → component processing → assembly and positioning → welding (segmentation) → flipping → welding (back side) → straightening → inspection, and the two processes of "assembly and positioning" and "welding (segmentation)" in its directed edge set need to be performed simultaneously, then the process is adjusted to layout and blanking → component processing → assembly and positioning, welding (segmentation) → flipping → welding (back side) → straightening → inspection.

[0089] S33, sort the process set The set of geometric entities of the current component Create a bidirectional association index.

[0090] Sorted process set The set of geometric entities of the current component The positive mapping between them can be represented as:

[0091] Sorted process set The set of geometric entities of the current component The inverse mapping between them can be represented as:

[0092] in, The power set of a set of geometric entities. This represents the power set of the sorted process set.

[0093] Sorted process set The set of geometric entities of the current component The bidirectional association between them can be represented as:

[0094] That is, any process It can locate the corresponding set of geometric entities, any geometric entity It is also possible to perform a reverse query on its associated process set. Φ(OP) j ) indicates based on the process object OP j The set of associated process nodes obtained by performing a forward query; Ψ(g) represents the set of associated process nodes based on the geometric entity. The set of related process objects obtained by performing a reverse query.

[0095] When the process object Associated with geometric entities (process nodes) through positive mapping. At that time, the corresponding operation object (OP) should be able to be retrieved through reverse mapping. j That is, satisfying: g∈Φ(OP) j )⇒ OP j ∈Ψ(g); Meanwhile: OP j ∈Ψ(g)⇒g∈Φ(OP j Therefore, the above bidirectional association constraints are used to ensure the consistency of forward and reverse mapping results, and to realize the bidirectional association index between process objects and process nodes.

[0096] In practical engineering, bidirectional indexes can include the following types of sub-indexes:

[0097] in: Indicates the drawing area index; Indicates the index of the bill of materials item; Indicates the index of the test point; Indicates the index of geometric feature elements.

[0098] S34. Traverse all components and repeat steps S31-S33. Sort the standard process chain of the optimal process template corresponding to each component and bind the sorted process sequence to the component to establish a two-way association.

[0099] In a preferred embodiment, the method further includes step S4, whereby when a local node change occurs in the three-dimensional design model of a ship product, incremental updates are performed on the changed node and the nodes within its influence range.

[0100] like Figure 5 As shown, the specific steps for performing incremental updates on the changed node and the nodes within its influence range are as follows: Let the original three-dimensional design model of the ship product be... The revised 3D design model of the ship product is Then the difference operator can be expressed as:

[0101] in: Indicates the addition of new process elements; This indicates the deletion of a process element; This indicates a modification to the process elements; The set of affected nodes can be defined as:

[0102] Repeat steps S1-S3, recalculating the process only for the set of affected nodes:

[0103] in, Represents the original process model. This represents the updated process model.

[0104] Version iteration can be represented as:

[0105] in, Indicates the first Version of the process model or work package, This indicates the incremental change (including newly added, modified, or deleted process information).

[0106] The present invention also provides a process information mapping and matching system based on a ship 3D model, including a process feature extraction module, a process template matching module, a process model reconstruction module, and a bidirectional index and version management module.

[0107] The process feature extraction module is used to match the process features of each component in the three-dimensional design model of the ship product based on the semantic association rule base, and obtain the structured process attribute table of each component. The process template matching module is used to perform similarity matching between the structured process attribute table of the component and the standard process template in the process template library using cosine similarity, so as to obtain the optimal process template for each component. The process model reconstruction module is used to sort the standard process chain of the optimal process template according to the constraints between the processes of each component; The bidirectional index and version management module is used to bind the geometric entity of each component to the process sequence after sorting its best process template to establish a bidirectional association index, and to perform incremental updates and version tracking when there are local node changes in the three-dimensional design model of the ship product.

[0108] It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A method for mapping and matching process information based on a three-dimensional ship model, characterized in that, Specifically, the following steps are included: S1. Based on the semantic association rule base, the process features of each component in the three-dimensional design model of the ship product are matched to obtain the structured process attribute table of each component. S2, using cosine similarity to match the structured process attribute table of the component with the standard process template in the process template library to obtain the optimal process template for each component; S3, based on the pre- and post-process constraints of each component, sorts the standard process chain of its optimal process template and binds the sorted process sequence to the component to establish a two-way association.

2. The method for mapping and matching process information based on a three-dimensional ship model according to claim 1, characterized in that, The specific steps in step S1 for matching the process features of each component in the 3D design model of the ship product based on the semantic association rule base are as follows: S11. Obtain the three-dimensional design model of the ship product and perform semantic analysis on it to identify and extract the composite process features of each component. The composite process features include topological connection features and geometric scale features. S12, Match the topological connection features of a component with the semantic association rules in the semantic association rule base to initially screen out all the process categories to which it belongs; S13, match the geometric scale features of the current component with the semantic association rules in the semantic association rule base, and refine the construction process attributes of each process category to which it belongs. S14. Generate a structured process attribute table for the component based on all process categories to which the current component belongs and the construction process attributes under each process category. S15, iterate through all components of the three-dimensional design model of the ship product, repeat steps S12-S14, and obtain the structured process attribute table of each component.

3. The method for mapping and matching process information based on a three-dimensional ship model according to claim 2, characterized in that, The semantic association rule base is used to store the association rules between the topological connection features and geometric scale features of components and their corresponding process categories. The topological connection characteristics of a component include the surface connection relationship, edge connection relationship, node connection relationship and assembly constraint relationship of the component; the geometric dimensional characteristics of a component include the length, width, height, wall thickness, aperture, angle, curvature, spacing and offset of the component. The process categories include cutting, assembly, welding, outfitting, and painting.

4. The method for mapping and matching process information based on a three-dimensional ship model according to claim 1, characterized in that, The specific steps in step S2, which use cosine similarity to match the structured process attribute table of the component with the standard process template in the process template library, are as follows: S21, map the structured process attribute tables of all components to process attribute feature vectors respectively, and map each standard process template in the process template library to a template feature vector respectively; S22, calculate the cosine similarity between the process attribute feature vector of a component and the template feature vector of each standard process template, and sort the cosine similarity scores in descending order, and select the top N standard process templates as candidate process templates. S23, Select the best process template for the current component from the candidate process templates; S24. Traverse all components of the three-dimensional design model of the ship product, repeat steps S22 and S23, and obtain the optimal process template for each component.

5. The method for mapping and matching process information based on a three-dimensional ship model according to claim 4, characterized in that, In step S23, the standard process template with the highest cosine similarity score among the candidate process templates is taken as the best process template for the current component. Alternatively, based on the ship process verification rules, the pre- and post-process constraint verification rules, and the template validity rules, calculate the comprehensive scoring function for each candidate process template, and select the candidate process template with the highest comprehensive score as the best process template.

6. The method for mapping and matching process information based on a three-dimensional ship model according to claim 5, characterized in that, The formula for calculating the comprehensive scoring function is as follows: in, For the first Cosine similarity of candidate process templates; For the first The process constraint satisfaction score of each candidate process template; and These are the weighting coefficients. No. The formula for calculating the process constraint satisfaction score of each candidate process template is as follows: in, For the first Satisfaction with ship process verification rules for each candidate process template For the first Satisfaction of pre- and post-process constraint rules for each candidate process template For the first The degree of template validity rule satisfaction for each candidate process template; The formula for calculating cosine similarity is: in, This is the feature vector of the component's technological attributes. For the first Template feature vectors of candidate process templates.

7. The method for mapping and matching process information based on a three-dimensional ship model according to claim 1, characterized in that, In step S3, the specific steps for sequencing the standard process chain of the optimal process template for each component based on the constraints between its processes are as follows: S31, Based on the pre- and post-process constraints of a certain component, its directed edge set is represented as follows: in, Indicates the process of the current component. It must precede its process. implement; The standard process set for the optimal process template corresponding to the current component is as follows: in, This refers to the m-th standard operation in the optimal process template, where m is the number of standard operations in the optimal process template. S32, based on the current component's directed edge set and its optimal process template's standard process set, the process sorting is modeled as a topological sorting problem on a directed acyclic graph. The sorted process sequence is represented as: in, This is the sorted sequence of processes; S33, sort the process set The set of geometric entities of the current component Establish a bidirectional association index; S34. Traverse all components and repeat steps S31-S33. Sort the standard process chain of the optimal process template corresponding to each component and bind the sorted process sequence to the component to establish a two-way association.

8. The method for mapping and matching process information based on a three-dimensional ship model according to claim 1, characterized in that, It also includes step S4, which involves performing incremental updates on the changed nodes and the nodes within their influence range when a local node change occurs in the three-dimensional design model of the ship product.

9. The method for mapping and matching process information based on a three-dimensional ship model according to claim 8, characterized in that, The specific steps for performing incremental updates on the changed node and the nodes within its affected radiation range are as follows: Let the original three-dimensional design model of the ship product be... The revised 3D design model of the ship product is Then the difference operator can be expressed as: in: Indicates the addition of new process elements; This indicates the deletion of a process element; This indicates a modification to the process elements; The set of affected nodes can be defined as: Repeat steps S1-S3, recalculating the process only for the set of affected nodes: in, Represents the original process model. This represents the updated process model.

10. A process information mapping and matching system based on a three-dimensional ship model, characterized in that, It includes a process feature extraction module, a process template matching module, a process model reconstruction module, and a two-way indexing and version management module. The process feature extraction module is used to match the process features of each component in the three-dimensional design model of the ship product based on the semantic association rule base, and obtain a structured process attribute table for each component. The process template matching module is used to perform similarity matching between the structured process attribute table of the component and the standard process template in the process template library using cosine similarity, so as to obtain the optimal process template for each component. The process model reconstruction module is used to sort the standard process chain of the optimal process template according to the constraints between the processes of each component; The bidirectional index and version management module is used to bind the geometric entity of each component to the process sequence after sorting its best process template to establish a bidirectional association index, and to perform incremental updates and version tracking when there are local node changes in the three-dimensional design model of the ship product.

Images