Modular design method and system for gearbox based on knowledge graph and virtual device

Abstract

This invention relates to the field of gearbox design technology, and more particularly to a knowledge graph-based modular design method, system, and virtual device for gearboxes. The method includes: constructing a gearbox design knowledge graph; parsing design requirements and extracting key parameters as search conditions; performing module retrieval and compatibility screening within the knowledge graph to form a candidate module set; combining and optimizing the candidate modules based on relationships and / or historical success cases in the graph to determine the optimal module combination scheme; and verifying the scheme using engineering standards and verification rules from the graph and outputting the results. This invention achieves structured association and intelligent application of design knowledge through a knowledge graph, solving the problems of knowledge dispersion, inefficient retrieval, and compatibility conflicts in traditional modular design. It effectively improves the efficiency, accuracy, and reliability of gearbox modular design and reduces design costs.

Description

Technical Field

[0001] This invention relates to the field of gearbox design technology, and in particular to a knowledge graph-based modular design method, system, and virtual device for gearboxes. Background Technology

[0002] As a core component of mechanical transmission systems, gearboxes are widely used in industrial machinery, transportation, new energy, and other fields. Different application scenarios have significantly different requirements for gearbox power, speed, transmission ratio, and installation space. Traditional design methods often adopt a customized development model, which suffers from problems such as fragmented design knowledge, low module reuse rate, long design cycle, and high cost.

[0003] Modular design, by breaking down a product into reusable standard modules, enables flexible combination of these modules, effectively improving design efficiency and reducing costs. However, in existing gearbox modular designs, design knowledge is mostly stored in the form of documents and tables, lacking structured associations. This leads to module retrieval, matching, and combination processes relying on the designer's experience, which can easily result in module compatibility conflicts and substandard performance, thus limiting the application effectiveness of modular design. Summary of the Invention

[0004] In view of this, the purpose of this invention is to propose a knowledge graph-based modular design method, system, and virtual device for gearboxes, in order to solve the problems of module compatibility conflicts and substandard performance that are common in existing technologies, which limit the application effect of modular design.

[0005] To achieve the above objectives, this invention provides a knowledge graph-based modular design method for gearboxes, comprising the following steps:

[0006] Step S1: Extract knowledge from the structured, semi-structured, and unstructured data related to gearbox design, and construct a gearbox design knowledge graph that includes design entities, entity attributes, and relationships between entities. The design entities at least cover gearbox module types, components, design constraints, performance indicators, and engineering standards.

[0007] Step S2: Obtain and parse the gearbox design requirements, extract the key parameters in the requirements, and map the key parameters to the retrieval conditions that the knowledge graph can recognize. The key parameters include at least power, speed, transmission ratio, installation space and reliability requirements.

[0008] Step S3: Query the knowledge graph according to the search conditions, locate the initial module that meets the requirements, and filter the initial module based on the compatibility relationship between entities stored in the knowledge graph to remove modules with interface mismatch or performance conflict, forming a candidate module set.

[0009] Step S4: Based on the inter-module relationships and / or successful combination patterns of historical design cases recorded in the knowledge graph, combine the modules in the candidate module set to generate one or more feasible module combination schemes, and determine the optimal module combination scheme according to the preset optimization objective.

[0010] Step S5: Call the engineering standards and / or verification rules associated in the knowledge graph to verify the design of the optimal module combination scheme. If the verification passes, output the scheme as the final modular design result. If the verification fails, return to step S3 or step S4 for adjustment based on the feedback information from the knowledge graph.

[0011] Preferably, in step S1, the step of constructing the gearbox design knowledge graph further includes:

[0012] The system collects structured, semi-structured, and unstructured data in the field of gearbox design. The structured data includes component models, parameter tables, and engineering standard documents. The semi-structured data includes technical manuals and design specification documents. The unstructured data includes design experience summaries, historical design cases, and fault analysis reports.

[0013] The collected data is cleaned and preprocessed. The design entities and their attributes are extracted using entity recognition technology, and the relationships between entities are extracted. These relationships include the compositional relationships between modules and components, the compatibility relationships between modules, and the fulfillment relationships between components and design constraints.

[0014] The extracted entities, attributes, and relationships are stored in a graph database to construct the topology of the knowledge graph, and a knowledge update mechanism is established to support the dynamic access of new design knowledge, component information, and constraint rules.

[0015] Preferably, step S3 further includes:

[0016] Based on the search criteria, the entity association query function of the knowledge graph is used to retrieve the initial modules that match the performance indicators and / or constraint requirements.

[0017] Based on the pre-stored interface size matching relationship and / or rotation speed range adaptation relationship between modules in the knowledge graph, the initial modules are subjected to compatibility verification to eliminate incompatible modules and form the candidate module set.

[0018] Preferably, in step S4, combining the modules in the candidate module set further includes:

[0019] Based on the candidate module set, multiple module combination schemes are generated by combining the inter-module association weights stored in the knowledge graph; the association weights are determined based on historical reuse rate and / or reliability score.

[0020] Based on the preset parameter priorities in the design requirements, multi-objective optimization is performed on the generated multiple module combination schemes. The multi-objective optimization algorithm is a linear weighted algorithm, which multiplies multiple conflicting indicators by custom weights, adds them together, and merges them into a single objective function, which is then solved directly according to the single-objective optimization method. This can be expressed by formula F. total The expression is represented as =w1f1+w2f2+w3f3+…, where w is the weight and f is the index. The parameter priority is directly controlled by the weight. The indices for the multi-objective optimization include at least transmission performance, manufacturing cost, and installation volume.

[0021] Preferably, step S4 further includes:

[0022] Query the historical design cases stored in the knowledge graph to obtain successful module combination patterns similar to the current design requirements;

[0023] Refer to the successful module combination pattern to assist in generating the module combination scheme and / or to select the optimal module combination scheme.

[0024] Preferably, in step S5, the design verification of the optimal module combination scheme further includes:

[0025] Based on the engineering standard entities associated in the knowledge graph, verify whether the optimal module combination scheme meets the corresponding industry standards or design specifications.

[0026] By querying the knowledge graph, historical verification data or failure case records of modules in the optimal module combination scheme are obtained, and their operational reliability in similar application scenarios is evaluated.

[0027] This invention also provides a knowledge graph-based modular design system for gearboxes, comprising:

[0028] The knowledge graph construction and maintenance module is used to extract and fuse knowledge from multi-source data related to gearbox design, and to build and maintain a knowledge graph containing design entities, entity attributes and relationships between entities.

[0029] The requirement parsing module is used to acquire and parse the gearbox design requirements, extract key parameters and map them to the retrieval conditions of the knowledge graph;

[0030] The intelligent retrieval and matching module is used to retrieve and filter a set of compatible candidate modules in the knowledge graph based on the retrieval conditions.

[0031] The combinatorial optimization module is used to generate and perform multi-objective optimization on the candidate module set based on the associations and / or historical design cases in the knowledge graph, to determine the optimal module combination scheme. The multi-objective optimization algorithm is a linear weighted algorithm, which multiplies multiple conflict indicators by custom weights and adds them together to form a single objective function, which is then solved directly according to the single-objective optimization method. This can be expressed by formula F. total =w1f1+w2f2+w3f3+…, where w is the weight and f is the index, and the parameter priority is directly controlled by the weight.

[0032] The verification and output module is used to call the relevant rules and standards in the knowledge graph to verify the optimal module combination scheme and output the final design scheme.

[0033] The present invention also provides a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the above-described method.

[0034] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0035] The beneficial effects of this invention are:

[0036] 1. This invention constructs a gearbox design knowledge graph, which uniformly extracts and represents structured, semi-structured, and unstructured design knowledge scattered in documents, tables, and experience cases, forming a semantic network centered on entity-relationship. This approach overcomes the limitations of isolated knowledge in traditional modular design, achieving systematic accumulation and intelligent retrieval of design knowledge, and significantly improving knowledge reuse rate.

[0037] 2. This invention parses design requirements into identifiable retrieval conditions based on a knowledge graph. Through entity association queries and compatibility filtering mechanisms within the knowledge graph, it can automatically and accurately locate a set of candidate modules that meet performance standards and have compatible interfaces. Compared to existing methods that rely on designers' experience to manually review documents and compare parameters, this invention effectively avoids module selection errors caused by information omissions or insufficient experience, significantly improving the recall and precision of module retrieval and shortening the design cycle in the initial solution selection phase.

[0038] 3. When generating module combination schemes, this invention not only relies on preset engineering constraints but also further incorporates the inter-module association weights recorded in the knowledge graph (such as historical reuse rate and reliability score) and successful combination patterns from historical design cases as optimization criteria. Through a multi-objective optimization mechanism, it can balance multiple design indicators such as transmission performance, manufacturing cost, and installation volume, assisting designers in quickly determining the optimal module combination under complex constraints and improving the rationality and reliability of the design scheme.

[0039] 4. This invention utilizes engineering standard entities and historical verification data linked by a knowledge graph to perform standard compliance verification and operational reliability assessment on the generated optimal module combination scheme, realizing knowledge-driven and automated design verification. Simultaneously, by establishing a closed-loop feedback mechanism of "returning for adjustment if verification fails," design conflicts can be accurately located and scheme corrections can be guided, reducing design iterations in later physical prototype manufacturing and testing stages, thereby effectively reducing overall R&D costs. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart of the knowledge graph-based modular design method for gearboxes according to an embodiment of the present invention;

[0042] Figure 2 This is a flowchart illustrating the knowledge graph construction process for gearbox design according to an embodiment of the present invention.

[0043] Figure 3 This is a schematic diagram of a knowledge graph of gearbox design for some embodiments of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0045] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0046] Example 1:

[0047] like Figure 1 As shown in the figure, this embodiment provides a modular design method for gearboxes based on knowledge graphs. This method is implemented by a computer program and can run on a server or local computing device. Figure 1 The overall flowchart of the method is shown, which specifically includes the following steps:

[0048] Step S1: Construct a knowledge graph for gearbox design

[0049] The goal of this step is to establish a structured, machine-readable knowledge base for gearbox design. For example... Figure 2 As shown, the construction process of a knowledge graph specifically includes three stages: data collection, data preprocessing and knowledge extraction, and graph storage and maintenance.

[0050] First, data collection was conducted. Multi-source heterogeneous data accumulated by an industrial gearbox company over its long-term design practice were collected. This data was categorized into three types based on its level of structure:

[0051] Structured data includes, but is not limited to, a model list of 10 core modules (such as input modules, transmission modules, speed change modules, etc.), detailed parameter tables of more than 500 parts, and more than 20 relevant national and industry engineering standard documents (such as the gear strength calculation standard GB / T 3480 series).

[0052] Semi-structured data includes various product technical manuals, internal design specification documents, design task books, etc.

[0053] Unstructured data includes designers' experience summary notes, over 300 historical design case records and their corresponding design scheme drawings, as well as product failure analysis reports.

[0054] Next, data preprocessing and knowledge extraction are performed. The collected raw data is cleaned to remove redundant, duplicate, and erroneous information. Subsequently, entity recognition technology in natural language processing, specifically named entity recognition (NER), is used to extract key design entities such as components, materials, structural parameters, performance indicators, and constraints from the natural language text of design requirements. Its rule scheme is based on domain-specific dictionaries, regular expression matching, and custom industry rules to complete entity extraction and identify the relationships between entities.

[0055] The design entity mainly includes gearbox module types (such as power input module, transmission module, speed change module, output module, auxiliary support module), specific components (such as gears, bearings, housing, shafts), design constraints (such as strength constraints, spatial dimension constraints, interface dimension constraints), performance indicators (such as transmission efficiency, noise, allowable power, speed range), and engineering standards (such as GB / T 3480.3).

[0056] Entity attributes: For example, the attributes of a gear entity include module, number of teeth, material, and precision grade; the attributes of a module entity include rated power range, allowable input speed, and interface flange size. For example, suitable input modules and transmission module candidates are retrieved based on "power ≥ 55kW". The candidate modules can achieve continuous parameter quantification matching using a piecewise linear matching degree function. Parameter names, constraint operators, and thresholds are extracted through named entity recognition. Combined with actual parameters from the module library, constraint types such as upper limits, lower limits, and intervals are distinguished to construct a 0~1 normalized matching degree model. Module solutions that do not meet the hard constraints are assigned a score of zero for screening. Solutions that meet the constraints are calculated with a refined matching score based on the degree of parameter redundancy. Finally, a multi-objective optimization solution is obtained through a weighted fusion algorithm.

[0057] Relationships: Various semantic relationships are defined to form a knowledge network, such as the "inclusion" relationship between "transmission module" and "gear"; the "interface compatibility" relationship between "input module A" and "transmission module B"; and the "must satisfy" relationship between the "gear" entity and the "strength constraint" entity.

[0058] Finally, graph storage and maintenance are performed. The more than 1200 design entities, more than 3000 relationships and their attributes extracted above are stored in the form of a graph database. In this embodiment, Neo4j graph database is specifically selected. By constructing nodes (representing entities) and edges (representing relationships), a graph is formed as follows: Figure 3 The diagram shows a partial topology of the gearbox design knowledge graph. Simultaneously, a knowledge update mechanism is established to support the dynamic integration of newly developed modules and newly defined design rules into the graph, ensuring the timeliness of the knowledge.

[0059] Step S2: Analyze design requirements and map search criteria

[0060] In this embodiment, the specific design requirements proposed by the user are: "Gearbox for wind turbine, power 1.5MW, input speed 1500rpm, output speed 150rpm, installation space ≤1200×800×600mm, service life ≥20 years".

[0061] The requirements described in this natural language were analyzed, and the key quantifiable parameters were extracted: rated power P = 1.5MW; rated input speed n. in =1500rpm; Rated output speed n out =150rpm, transmission ratio i=10; installation space constraints are length ≤1200mm, width ≤800mm, height ≤600mm, and reliability requirements are indicated by service life ≥20 years.

[0062] Next, these key parameters are mapped to structured search conditions that the knowledge graph can recognize. For example, "power 1.5MW" is mapped to a query for "≥1.5MW" for the "rated power" attribute of the "input module" and "transmission module" entities; "transmission ratio 10" is mapped to a matching query for the "transmission ratio" attribute of the "gear shift module" entity.

[0063] Step S3: Knowledge Graph-Based Module Retrieval and Compatibility Filtering

[0064] Based on the search criteria generated in step S2, a query operation is performed on the constructed knowledge graph to locate the initial modules that meet the performance indicators and constraints. For example, the graph query reveals that there are 3 types of input modules, 4 types of transmission modules, 2 types of speed change modules, 3 types of output modules, and 2 types of auxiliary support modules that meet the power and speed requirements.

[0065] After obtaining the initial set of modules, further filtering is performed based on the pre-stored compatibility relationships between entities in the knowledge graph. The "compatibility" edges between "module-modules" in the graph are queried to verify whether the interface dimensions of any two connectable modules match and whether their operating speed ranges overlap. For example, if the flange interface size of a candidate transmission module does not match the interfaces of any candidate input modules, that transmission module is removed from the set. Through this compatibility filtering step, a final set of candidate modules is formed, in which each module meets the performance requirements and has feasible connection relationships with other modules.

[0066] Step S4: Generation and Optimization of Module Combination Scheme

[0067] Based on the candidate module set obtained in step S3, module combination and scheme optimization are performed.

[0068] First, a module combination scheme is generated. The inter-module association weights recorded in the knowledge graph are used to assist in the combination. These association weights are calculated using the co-occurrence frequency normalization method, with the specific formula as follows: , where N ab Let max( be the historical co-occurrence frequency of module a and module b) Ni,j The highest co-occurrence frequency of all module combinations is used as the formula to normalize the association weights to the [0,1] interval, resulting in higher weights for mature modules and zero weights for mutually exclusive modules. Integrating the module association weights into the combination evaluation system and weighting them with the continuous parameter matching score improves the engineering rationality and maturity of the combination scheme. This weight can be set based on historical reuse rate (i.e., the frequency with which two modules are used together in past designs) and / or reliability score (i.e., the performance of the combination in historical cases). By calculating the sum of the weights of different module combination paths, three relatively optimal feasible combination schemes can be initially generated.

[0069] Secondly, multi-objective optimization is performed. Based on the priorities implicit in user requirements (e.g., prioritizing cost control while meeting rigid requirements for power and lifespan), multiple generated solutions are evaluated against various objectives. Evaluation metrics include transmission performance (transmission efficiency), manufacturing cost (the sum of the costs of each module), and installation volume (the combined external dimensions).

[0070] Simultaneously, this step also incorporates a historical case-based decision-making mechanism. It queries over 300 historical design cases stored in the knowledge graph to determine if there are successful cases similar to the current requirement of "wind turbine gearbox, 1.5MW power" and their respective module combination patterns. In this embodiment, the knowledge graph provides design records for several similar wind turbine projects.

[0071] Based on comprehensive weight calculation, multi-objective optimization evaluation, and reference to historical successful models, the optimal module combination scheme consisting of "input module M1, transmission module T3, speed change module S2, output module O2, and auxiliary module A1" was finally determined.

[0072] Step S5: Design Verification and Output

[0073] Before outputting the final solution, design verification must be performed to ensure the compliance and reliability of the solution.

[0074] First, standard compliance verification is performed. The engineering standard entities associated with the knowledge graph (such as design specifications related to wind turbine gearboxes) are called to automatically verify whether the strength verification results of each gear and the overall efficiency calculation value in the combined scheme of "input module M1 + transmission module T3 + speed change module S2 + output module O2 + auxiliary module A1" meet the requirements of the standard.

[0075] Secondly, reliability verification was conducted. A reverse query using a knowledge graph was performed to retrieve historical verification data and failure case records for key modules (or similar module combinations) in past projects. The query results showed that this specific module combination (or highly similar combinations) had been used in three similar wind turbine application projects with good operating records and no related failure reports.

[0076] The optimal module combination scheme was verified to meet all design requirements and verification rules, and the verification was successful. Finally, the system outputs a complete modular design scheme for the gearbox, which details the module composition list (M1, T3, S2, O2, A1), the interface size parameters of each module, and the assembly technical requirements.

[0077] If any non-compliance is found during the verification step (e.g., the calculated stress of a gear exceeds the allowable value), the system returns to step S3 to reselect candidate modules or to step S4 to adjust the combination optimization strategy, based on the specific conflict information fed back from the knowledge graph (e.g., "Gear X's strength does not meet GB / T 3480.3 constraint"). The specific adjustment method involves tiered control based on the extent of performance deviation: for minor strength exceedances, the system returns to step S4, correcting the multi-objective optimization bias and optimizing the module combination matching logic by increasing the weight of mechanical indicators, strengthening constraint penalty coefficients, and dynamically lowering the association weight of weak modules; for serious failures that do not meet mandatory standards, the system returns to step S3, updating the candidate module set by eliminating unqualified modules, replacing them with high-strength alternatives, and adding pre-screening rules for strength standards. Through a closed-loop mechanism of conflict feedback—tiered adjustment—iterative optimization, the module selection and combination strategy are continuously corrected until all structural, strength, and standard constraints are met, outputting a qualified and feasible module combination scheme.

[0078] Example 2:

[0079] This embodiment provides a knowledge graph-based modular design system for gearboxes, used to execute the design method provided in Embodiment 1. The system includes:

[0080] Knowledge Graph Construction and Maintenance Module: Responsible for executing step S1, extracting and fusing knowledge from multi-source heterogeneous data, and constructing and continuously maintaining the gearbox design knowledge graph.

[0081] Requirements Analysis Module: Responsible for executing step S2, obtaining the design requirements input by the user, extracting key parameters, and mapping them to graph retrieval conditions.

[0082] Intelligent retrieval and matching module: responsible for executing step S3, retrieving initial modules that meet the conditions in the knowledge graph, performing compatibility filtering, and outputting a set of candidate modules.

[0083] Combinatorial optimization module: responsible for executing step S4, performing scheme combination and multi-objective optimization based on the candidate module set, and determining the optimal combination.

[0084] Verification and Output Module: Responsible for executing step S5, calling on the graph knowledge to verify the scheme, and outputting the final design result.

[0085] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those 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 this application.

[0086] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0087] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0088] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0089] The implementation of all or part of the processes in the methods of the above embodiments can also be accomplished by a computer program product. When the computer program product is run on a terminal device, the terminal device can implement the steps in the various method embodiments described above.

[0090] The embodiments described above are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A knowledge graph based modular design method for gearbox, characterized in that, The method includes the following steps: Step S1: Extract knowledge from the structured, semi-structured, and unstructured data related to gearbox design, and construct a gearbox design knowledge graph that includes design entities, entity attributes, and relationships between entities. The design entities at least cover gearbox module types, components, design constraints, performance indicators, and engineering standards. Step S2: Obtain and parse the gearbox design requirements, extract the key parameters in the requirements, and map the key parameters to the retrieval conditions that the knowledge graph can recognize. The key parameters include at least power, speed, transmission ratio, installation space and reliability requirements. Step S3: Query the knowledge graph according to the search conditions, locate the initial module that meets the requirements, and filter the initial module based on the compatibility relationship between entities stored in the knowledge graph to remove modules with interface mismatch or performance conflict, forming a candidate module set. Step S4: Based on the inter-module relationships and / or successful combination patterns of historical design cases recorded in the knowledge graph, combine the modules in the candidate module set to generate one or more feasible module combination schemes, and determine the optimal module combination scheme according to the preset optimization objective. Step S5: Call the engineering standards and / or verification rules associated in the knowledge graph to verify the design of the optimal module combination scheme. If the verification passes, output the scheme as the final modular design result. If the verification fails, return to step S3 or step S4 for adjustment based on the feedback information from the knowledge graph.

2. The knowledge graph based gearbox modular design method of claim 1, wherein, In step S1, the step of constructing the gearbox design knowledge graph further includes: The system collects structured, semi-structured, and unstructured data in the field of gearbox design. The structured data includes component models, parameter tables, and engineering standard documents. The semi-structured data includes technical manuals and design specification documents. The unstructured data includes design experience summaries, historical design cases, and fault analysis reports. The collected data is cleaned and preprocessed. The design entities and their attributes are extracted using entity recognition technology, and the relationships between entities are extracted. These relationships include the compositional relationships between modules and components, the compatibility relationships between modules, and the fulfillment relationships between components and design constraints. The extracted entities, attributes, and relationships are stored in a graph database to construct the topology of the knowledge graph, and a knowledge update mechanism is established to support the dynamic access of new design knowledge, component information, and constraint rules.

3. The knowledge graph based gearbox modular design method of claim 1, wherein, Step S3 further includes: Based on the search criteria, the entity association query function of the knowledge graph is used to retrieve the initial modules that match the performance indicators and / or constraint requirements. Based on the pre-stored interface size matching relationship and / or rotation speed range adaptation relationship between modules in the knowledge graph, the initial modules are subjected to compatibility verification to eliminate incompatible modules and form the candidate module set.

4. The knowledge graph-based modular design method for gearboxes according to claim 1, characterized in that, In step S4, combining the modules in the candidate module set further includes: Based on the candidate module set, multiple module combination schemes are generated by combining the inter-module association weights stored in the knowledge graph; the association weights are determined based on historical reuse rate and / or reliability score. Based on the preset parameter priorities in the design requirements, the generated multiple module combination schemes are optimized in multiple objectives. The indicators of the multi-objective optimization include at least transmission performance, manufacturing cost and installation volume.

5. The knowledge graph-based modular design method for gearboxes according to claim 4, characterized in that, Step S4 further includes: Query the historical design cases stored in the knowledge graph to obtain successful module combination patterns similar to the current design requirements; Refer to the successful module combination pattern to assist in generating the module combination scheme and / or to select the optimal module combination scheme.

6. The knowledge graph-based modular design method for gearboxes according to claim 1, characterized in that, In step S5, the design verification of the optimal module combination scheme further includes: Based on the engineering standard entities associated in the knowledge graph, verify whether the optimal module combination scheme meets the corresponding industry standards or design specifications. By querying the knowledge graph, historical verification data or failure case records of modules in the optimal module combination scheme are obtained, and their operational reliability in similar application scenarios is evaluated.

7. A modular design system for gearboxes based on knowledge graphs, characterized in that, include: The knowledge graph construction and maintenance module is used to extract and fuse knowledge from multi-source data related to gearbox design, and to build and maintain a knowledge graph containing design entities, entity attributes and relationships between entities. The requirement parsing module is used to acquire and parse the gearbox design requirements, extract key parameters and map them to the retrieval conditions of the knowledge graph; The intelligent retrieval and matching module is used to retrieve and filter a set of compatible candidate modules in the knowledge graph based on the retrieval conditions. The combination optimization module is used to combine and optimize the candidate module set based on the associations and / or historical design cases in the knowledge graph, and to determine the optimal module combination scheme. The verification and output module is used to call the relevant rules and standards in the knowledge graph to verify the optimal module combination scheme and output the final design scheme.

8. A computing device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 6.

Images