Aero-engine systematic multi-dimensional digital characterization method and system

Abstract

The invention relates to the field of aero-engine digitization, and discloses an aero-engine systematic multi-dimensional digital characterization method and system, and the method comprises the steps: carrying out the step-by-step decomposition of a part structure of an aero-engine, forming an engine product decomposition structure, and according to the product level and inclusion relation of the engine product decomposition structure, carrying out the multi-dimensional digital characterization of the aero-engine. Constructing a multi-dimensional representation framework of the aero-engine; performing associated feature extraction on the model or data derived from each dimension to obtain a feature factor of each dimension; taking the feature factors extracted from each dimension as a variable ontology, and forming a skeleton variable graph taking the feature factors as a skeleton by adopting a semantic network technology; dynamically inputting data into the skeleton variable atlas to form an aero-engine characterization space based on the skeleton variable atlas; and carrying out characterization traversal calculation analysis on the characterization space of the aero-engine by adopting a machine learning algorithm to obtain a characterization traversal calculation analysis result. The characterization problem of an existing characterization technology in complex systems such as aero-engines is solved.

Description

Technical Field

[0001] The present invention relates to the field of aero-engine digitization and discloses a systematic multi-dimensional digital characterization method and system for aero-engines. Background Art

[0002] Representation in engineering refers to the process and method of transforming the key attributes, behaviors, or relationships of engineering physical systems, devices, or technical problems into structured, quantifiable forms for the purpose of describing, analyzing, or manipulating them. Its core lies in formal abstraction, which serves engineering design, simulation, optimization, or decision-making. The core elements of engineering representation include object simplification, mathematical / symbolic mapping, visualization / dataization, and computability.

[0003] Existing characterization techniques require engineers to possess advanced technical expertise. Geometric and graphical representations require extensive knowledge in structural and finite element methods, while mathematical and symbolic representations require a deep foundation in mathematics. Data-driven representations require expertise in data acquisition, data cleaning, and big data. Multiscale and multiphysics representations require the integration of cross-disciplinary knowledge. This specialized knowledge requires the efforts of multiple teams. Furthermore, existing characterization techniques are tailored to the needs of their respective domains, rather than focusing on a top-down approach to characterizing the entire system. Consequently, they lack systematicity and are incomplete. A systematic framework for representing system objects is needed. Furthermore, because these characterization techniques employ object simplification, they have specific adaptability and limitations when representing the system. These simplifications become ineffective when the conditions and environment change. For example, in the famous Leaning Tower of Pisa experiment, the effect of air resistance on the falling object's velocity could be ignored using two iron balls. However, when the test objects were replaced with feathers and iron balls, this simplification yielded completely different experimental results. The main problem in engineering is that before we have a clear understanding of the mechanism, we may not know whether a certain factor will have a significant impact on the result. If we simplify it at this time, it may lead to wrong results. Summary of the Invention

[0004] The purpose of the present invention is to provide a systematic multi-dimensional digital characterization method and system for aircraft engines, which can systematically establish a complete top-down engine characterization system through a hierarchical structured characterization system, and solve the characterization problems of existing characterization technologies in complex systems such as aircraft engines.

[0005] In order to achieve the above technical effects, the technical solution adopted by the present invention is: A systematic multi-dimensional digital characterization method for aircraft engines, comprising: Using the aircraft engine product decomposition structure as the backbone and the node objects at each layer of the structure as the carrier for the hierarchical organization of the representation system, multiple representation dimensions are recursively derived from the node objects at each layer of the structure. Based on the product hierarchy and inclusion relationships of the structure, a multi-dimensional representation framework for aircraft engines is constructed. A model analysis method is used to extract associated features from the models or data derived from each dimension to obtain characteristic factors for each dimension. The characteristic factors include design parameters, calculation parameters, and boundary conditions of corresponding aircraft engine components extracted from CAD models or CAE models, as well as variables, initial conditions, state vectors, input and output quantities, nodes, and logical relationship features extracted from differential equations, partial differential equations, state equations, and logical models. The calculation parameters include load parameters and material performance parameters in CAE models. The logical models include hierarchical models with tree structures expressing parent and child nodes, as well as many-to-many network models, data flow diagrams, and entity relationship diagrams. Using the characteristic factors extracted from each dimension as the variable ontology, semantic web technology is used to convert all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data into graph training data. Through machine learning training, a skeleton variable graph with the characteristic factors as the skeleton is formed. All or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data are input into the skeleton variable map as spatial training data, and an aircraft engine representation space based on the skeleton variable map is constructed using a machine learning algorithm; A machine learning algorithm is used to perform characterization traversal calculation and analysis on the aero-engine characterization space to obtain the characterization traversal calculation and analysis results.

[0006] Furthermore, it also includes: According to the characterization traversal calculation analysis results, combined with a deterministic rule model, the skeleton variable map is optimized to obtain an optimized skeleton variable map; the deterministic rules include design rules, process rules and usage rules in engineering practice; Inputting the spatial training data into the optimized skeleton variable map to form an aero-engine representation space based on the optimized skeleton variable map; A machine learning algorithm is used to perform representation traversal calculation and analysis on the aero-engine representation space based on the optimized skeleton variable map to obtain the representation traversal calculation and analysis results.

[0007] Furthermore, the method of constructing an aero-engine representation space based on a skeleton variable graph using a machine learning algorithm includes: Based on the characteristic factors in the spatial training data, a machine learning algorithm is used to establish associations between the corresponding characteristic factors and the nodes of the skeleton variable map, and to establish associations between the characteristic factors; Taking the characteristic factors with correlation as the representation factors of the representation space, the representation factors associated with each node in the skeleton variable map and the corresponding correlation description are combined to form a representation model. All representation models are combined to form an aircraft engine representation space based on the skeleton variable map.

[0008] Furthermore, it also includes: After establishing the representation space, all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data accumulated over a certain period of time are input into the representation space as incremental data to extract new characteristic factors from the incremental data. Taking the representation space as the traversal space, relying on the correlation threshold between each representation factor in the representation space as the evolution direction, adding new feature factors to the current representation model for dynamic combination to form different new representation models; Conduct sensitivity analysis on each characterization factor in the newly added characterization model, and use the sensitivity analysis results as the basis for evaluating the effectiveness of the combination. The newly added characterization model with a sensitivity analysis result greater than the preset limit value will be considered a valid newly added characterization model; Add effective new representation models to the representation space to participate in representation traversal calculation and evolution.

[0009] Furthermore, the characterization dimensions in the multi-dimensional characterization framework of aero-engines include engine product decomposition structure dimension, external characteristic parameter dimension, function / performance parameter dimension, physical characteristic dimension, process manufacturing dimension, environmental factor dimension, task sequence dimension, and fault failure dimension.

[0010] To achieve the above technical effects, the present invention further provides a systematic multi-dimensional digital characterization system for aircraft engines, which is used to implement the above-mentioned systematic multi-dimensional digital characterization method for aircraft engines, comprising: A representation framework construction module is used to build a multi-dimensional representation framework for aircraft engines, using the aircraft engine product decomposition structure as the backbone and the node objects at each layer of the aircraft engine product decomposition structure as the carrier for the hierarchical structured organization of the representation system. Multiple representation dimensions are recursively derived from the node objects at each layer of the aircraft engine product decomposition structure. This module is then used to build a multi-dimensional representation framework for aircraft engines, based on the product hierarchy and inclusion relationships of the aircraft engine product decomposition structure. A characteristic factor extraction module is used to extract associated features from models or data derived from each dimension using a model analysis method to obtain characteristic factors for each dimension. The characteristic factors include design parameters, calculation parameters, and boundary conditions of corresponding aircraft engine components extracted from CAD models or CAE models, as well as variables, initial conditions, state vectors, input and output quantities, nodes, and logical relationship features extracted from differential equations, partial differential equations, state equations, and logical models. The calculation parameters include load parameters and material performance parameters in CAE models. The logical models include hierarchical models with tree structures expressing parent and child nodes, as well as many-to-many network models, data flow diagrams, and entity relationship diagrams. The variable graph generation module is used to use the characteristic factors extracted from each dimension as the variable body. It uses semantic web technology to convert all or part of the aircraft engine design data, production data, simulation history data, test history data, and operation history data into graph training data. Through machine learning training, it forms a skeleton variable graph with the characteristic factors as the skeleton. a representation space generation module for inputting all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data as spatial training data into the skeleton variable map, and constructing an aircraft engine representation space based on the skeleton variable map using a machine learning algorithm; The representation analysis module is used to perform representation traversal calculation analysis on the aircraft engine representation space using a machine learning algorithm to obtain the representation traversal calculation analysis results.

[0011] Furthermore, the variable map generation module includes a variable map optimization unit for optimizing the skeleton variable map based on the characterization traversal calculation analysis results and in combination with a deterministic rule model to obtain an optimized skeleton variable map; the deterministic rules include design rules, process rules, and usage rules in engineering practice; The representation space generation module includes a representation space optimization unit, which is used to input spatial training data into the optimized skeleton variable map to form an aircraft engine representation space based on the optimized skeleton variable map; The representation analysis module is also used to use a machine learning algorithm to perform representation traversal calculation analysis on the aircraft engine representation space based on the optimized skeleton variable map to obtain the representation traversal calculation analysis results.

[0012] Furthermore, the representation space generation module further includes: A correlation establishing unit is used to establish correlations between the corresponding characteristic factors and the nodes of the skeleton variable map based on the characteristic factors in the spatial training data, and to establish correlations between the characteristic factors; The characterization model combination unit is used to use characteristic factors with correlation as characterization factors of the characterization space, and to form a characterization model together with the characterization factors associated with each node in the skeleton variable map and the corresponding correlation description. All characterization models are combined to form an aircraft engine characterization space based on the skeleton variable map.

[0013] Furthermore, in the characterization framework construction module, the characterization dimensions in the multi-dimensional characterization framework of the aero-engine include engine product decomposition structure dimension, external characteristic parameter dimension, function / performance parameter dimension, physical characteristic dimension, process manufacturing dimension, environmental factor dimension, task sequence dimension, and fault failure dimension.

[0014] Compared with the prior art, the present invention has the following beneficial effects: By establishing a correlation and quantitative relationship model between the characterization factors and specific targets, a strong correlation is established between the characterization factors and the targets. Then, through a hierarchical structured characterization system, the characterization complexity problem of complex systems such as aircraft engines is solved. Compared with traditional characterization and modeling technologies, the aircraft engine characterization space constructed by the present invention can comprehensively reflect information on various aspects of engine design, production, simulation, testing, and operation, making the characterization more systematic and complete. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of the systematic multi-dimensional digital characterization method for aircraft engines in Example 1; Figure 2 This is a structural block diagram of the systematic multi-dimensional digital characterization system for aircraft engines in Example 1; Figure 3 This is a flow chart of the systematic multi-dimensional digital characterization method for aircraft engines in Example 2; Among them, 1. Characterization framework construction module; 2. Feature factor extraction module; 3. Variable map generation module; 301. Variable map optimization unit; 4. Characterization space generation module; 401. Characterization space optimization unit; 402. Correlation establishment unit; 403. Characterization model combination unit; 5. Characterization analysis module. DETAILED DESCRIPTION

[0016] The present invention will be described in further detail below with reference to the embodiments and accompanying drawings. However, this should not be construed as limiting the scope of the present invention to the following embodiments, as all technologies implemented based on the present invention fall within the scope of the present invention.

[0017] Example 1 See also Figures 1 to 2 , a systematic multi-dimensional digital characterization method for aircraft engines, including: Using the aircraft engine product decomposition structure as the backbone and the node objects at each layer of the structure as the carrier for the hierarchical organization of the representation system, multiple representation dimensions are recursively derived from the node objects at each layer of the structure. Based on the product hierarchy and inclusion relationships of the structure, a multi-dimensional representation framework for aircraft engines is constructed. A model analysis method is used to extract associated features from the models or data derived from each dimension to obtain characteristic factors for each dimension. The characteristic factors include design parameters, calculation parameters, and boundary conditions of corresponding aircraft engine components extracted from CAD models or CAE models, as well as variables, initial conditions, state vectors, input and output quantities, nodes, and logical relationship features extracted from differential equations, partial differential equations, state equations, and logical models. The calculation parameters include load parameters and material performance parameters in CAE models. The logical models include hierarchical models with tree structures expressing parent and child nodes, as well as many-to-many network models, data flow diagrams, and entity relationship diagrams. Using the characteristic factors extracted from each dimension as the variable ontology, semantic web technology is used to convert all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data into graph training data. Through machine learning training, a skeleton variable graph with the characteristic factors as the skeleton is formed. All or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data are input into the skeleton variable map as spatial training data, and an aircraft engine representation space based on the skeleton variable map is constructed using a machine learning algorithm; A machine learning algorithm is used to perform characterization traversal calculation and analysis on the aero-engine characterization space to obtain the characterization traversal calculation and analysis results.

[0018] In this embodiment, a structured hierarchical representation organization approach is constructed based on the product breakdown structure (PBS). A universal aircraft engine representation framework is constructed from multiple dimensions. Model parsing methods are used to extract characteristic factors from models or data in each dimension. Semantic web technology is then used to convert the data into graph training data, using these characteristic factors as the core. A skeleton variable graph is then formed through machine learning training. By inputting aircraft engine-related data into the skeleton variable graph, a representation space is constructed using machine learning algorithms, forming a universal aircraft engine representation system. Representation traversal computational analysis is performed on the aircraft engine representation space to obtain representation traversal computational analysis results. This invention establishes a strong correlation between representation factors and specific targets by building a correlation and quantitative relationship model between them. This hierarchical structured representation system addresses the representation complexity of complex systems such as aircraft engines. Compared to traditional representation and modeling techniques, the aircraft engine representation space constructed in this invention can comprehensively reflect information from all aspects of engine design, production, simulation, testing, and operational use, resulting in a more systematic and complete representation.

[0019] Based on the same inventive concept, this embodiment further provides a systematic multi-dimensional digital characterization system for an aircraft engine, which is used to implement the systematic multi-dimensional digital characterization method for an aircraft engine, comprising: Representation framework construction module 1 is used to use the aircraft engine product decomposition structure as the mainstay and the node objects at each layer of the aircraft engine product decomposition structure as the carrier for the hierarchical structured organization of the representation system. Multiple representation dimensions are recursively derived from the node objects at each layer of the aircraft engine product decomposition structure. According to the product hierarchy and inclusion relationship of the aircraft engine product decomposition structure, a multi-dimensional representation framework for aircraft engines is constructed. Characteristic factor extraction module 2 is used to extract associated features from the models or data derived from each dimension using a model analysis method to obtain characteristic factors for each dimension; the characteristic factors include design parameters, calculation parameters, and boundary conditions of corresponding aircraft engine components extracted from CAD models or CAE models, as well as variables, initial conditions, state vectors, input and output quantities, nodes, and logical relationship features extracted from differential equations, partial differential equations, state equations, and logical models; the calculation parameters include load parameters and material performance parameters in CAE models; and the logical models include hierarchical models with tree structures expressing parent-child nodes, as well as many-to-many network models, data flow diagrams, and entity relationship diagrams. Variable graph generation module 3 is used to use the characteristic factors extracted from each dimension as the variable body, and uses semantic web technology to convert all or part of the aircraft engine design data, production data, simulation history data, test history data, and operation history data into graph training data. Through machine learning training, a skeleton variable graph is formed with the characteristic factors as the skeleton; a representation space generation module 4 for inputting all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data as spatial training data into the skeleton variable map, and constructing an aircraft engine representation space based on the skeleton variable map using a machine learning algorithm; The representation analysis module 5 is used to perform representation traversal calculation analysis on the aircraft engine representation space using a machine learning algorithm to obtain the representation traversal calculation analysis results.

[0020] In this embodiment, the variable map generation module 3 includes a variable map optimization unit 301, which is used to optimize the skeleton variable map based on the characterization traversal calculation analysis results and in combination with a deterministic rule model to obtain an optimized skeleton variable map; the deterministic rules include design rules, process rules, and usage rules in engineering practice; The representation space generation module 4 includes a representation space optimization unit 401, which is used to input the spatial training data into the optimized skeleton variable map to form an aircraft engine representation space based on the optimized skeleton variable map; On this basis, the representation analysis module 5 is further used to use a machine learning algorithm to perform a representation traversal calculation analysis on the aircraft engine representation space based on the optimized skeleton variable map to obtain a representation traversal calculation analysis result.

[0021] In this embodiment, the representation space generation module 4 further includes: A correlation establishing unit 402 is used to establish correlations between the corresponding characteristic factors and the nodes of the skeleton variable map based on the characteristic factors in the spatial training data, and to establish correlations between the characteristic factors; The characterization model combination unit 403 is used to use characteristic factors with correlation as characterization factors of the characterization space, and to form a characterization model together with the characterization factors and corresponding correlation descriptions that are correlated with each node in the skeleton variable map. All characterization models are combined to form an aircraft engine characterization space based on the skeleton variable map.

[0022] Example 2 See also Figure 3 , a systematic multi-dimensional digital characterization method for aircraft engines, including: Step 1: Using the aircraft engine product decomposition structure as the backbone and the node objects at each level of the aircraft engine product decomposition structure as the carrier for the hierarchical structured organization of the representation system, multiple representation dimensions are recursively derived from the node objects at each level of the aircraft engine product decomposition structure. Based on the product hierarchy and inclusion relationships of the aircraft engine product decomposition structure, a multi-dimensional representation framework for aircraft engines is constructed. In this example, an 8-Dimensional Representation Framework (8DRF) for aircraft engines is constructed based on the aircraft engine product breakdown structure (PBS). This framework includes: Dimension 1: Engine product breakdown structure dimension; Dimension 2: External characteristic parameter dimension, including: aircraft type adaptation, flight characteristic parameters, environmental characteristic requirements, user characteristic requirements, etc., which can be decomposed based on the requirements of relevant parties for engine operation, use and maintenance; Dimension 3: Function / performance parameter dimension, including indicators such as engine thrust, EGT, vibration spectrum, as well as the relationship between external characteristics and function / performance, and the relationship between function / performance and the physical environment. It can be decomposed into tensors based on the engine performance indicator system and expressed in horizontal and vertical relationships; Dimension 4: Physical properties, including: dimensions, shape, weight, center of gravity, moment of inertia, material properties (material creep coefficient, fatigue crack growth rate, etc.), internal and external physical environment, etc.; Dimension 5: Process manufacturing dimension, including manufacturing tolerance, assembly clearance, etc. Dimension 6: Environmental factors, including altitude, temperature, humidity, Mach number, etc. Dimension 7: Mission sequence dimension, including flight cycle, mission profile, etc. Dimension 8: Failure and failure dimension, including FMEA, FTA, etc.

[0023] Step 2: Use a model analysis method to extract associated features from the models or data derived from each dimension to obtain characteristic factors for each dimension; the characteristic factors include design parameters, calculation parameters, and boundary conditions of corresponding aircraft engine components extracted from the CAD model or CAE model, as well as variables, initial conditions, state vectors, input and output quantities, nodes, and logical relationship features extracted from differential equations, partial differential equations, state equations, and logical models; the calculation parameters include load parameters and material performance parameters in the CAE model; the logical model includes a hierarchical model with a tree structure expressing parent and child nodes, as well as a many-to-many network model, a data flow diagram, and an entity relationship diagram; In this embodiment, key variables include single independent variables on which single-variable functions and their derivatives depend, and multiple independent variables on which multivariate functions and their partial derivatives depend; the state vector includes each variable in the minimum independent variable group (such as capacitor voltage and inductor current in a circuit system, displacement and velocity in a mechanical system, etc.); input and output quantities include external action quantities such as applied force, voltage and other control inputs, and observable physical quantities such as sensor measurement values and other output quantities; nodes include parent-child nodes in a tree structure, many-to-many nodes in a graph structure, active nodes in an activity graph, etc.; logical relationship characteristics include hierarchical relationships, sequential relationships, and associative relationships, etc.

[0024] In this embodiment, the characteristic factor extraction of the engine product decomposition structure dimension is taken as an example to describe the characteristic factor extraction process of the corresponding dimension in this step in detail. The specific steps are as follows: 2.1 Based on the engine product breakdown structure (PBS), the engine technology breakdown structure (TBS) and work breakdown structure (WBS) are constructed. Based on the work breakdown structure (WBS), the engine schedule decomposition and resource breakdown structure (OBS) are completed. Based on the resource breakdown structure (OBS), the engine cost breakdown structure (CBS) is completed. 2.2 The model analysis method is used to extract the associated features of the models derived from the engine product decomposition structure PBS; the models derived from the engine product decomposition structure PBS include CAD models, CAE models, differential equations / partial differential equations / state equations / logic models.

[0025] In addition, for the extraction of characteristic factors of the "function / performance parameter dimension", we can use the existing aviation engine index system as a basis to extract the L-layer index factors and L+1-layer index factors in the index tree (L takes values of 0, 1, 2, 3... The 0-layer index factors are the top-level target factors of the engine), as well as the correlation between the L-layer and L+1-layer index factors (including but not limited to inclusion relationships, correlations, etc.), to construct a characterization system target factor tree, thereby obtaining the characteristic factors of the function / performance parameter dimension.

[0026] The engine's 8-Dimensional Representation Framework (8DRF) primarily extracts dimensional features through semantic recognition, large language models, knowledge graphs, and model parsing (applicable to rule-based models, computational models, task models, and behavioral models). For example, engine model adaptation features, mission scenario features, flight characteristics, mission environment characteristics, and user characteristics can be extracted from text descriptions, graphics, and models in the external feature dimension.

[0027] In this embodiment, feature extraction is the basic work of the present invention. Through the APIs of different CAD / CAE and other modeling tools and through adapter development, fine-grained analysis of the model can be achieved, and various feature parameters such as design size parameters, material parameters, load and boundary conditions can be obtained from the model and data to form a set of feature parameters.

[0028] Step 3: Using the characteristic factors extracted from each dimension as the variable ontology, semantic web technology is used to convert all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data into graph training data. Through machine learning training, a skeleton variable graph is formed with the characteristic factors as the skeleton. The variable ontology is a semantic data model that defines the concepts of engine variables and the attributes that can be used to describe them. In this embodiment, the variable ontology has three main components: Concepts (classes): different concepts / types that exist in the data; Relationship: The property that connects two concepts; Attributes: Properties that describe a single concept.

[0029] Multiple variable ontologies can be designed based on the engine's multi-dimensional, multi-view scenarios and business needs. Variable ontologies can define variables, variable relationships, and variable attribute information. Variable ontologies can be designed by dragging and dropping on the canvas, achieving what you see is what you get. Variable ontologies support version control to support graph changes and flexibility.

[0030] A variable graph is an application based on the knowledge graph. It is a database that graphically presents the relationships between different variable entities. It connects variable information of different types, dimensions, and levels (such as targets, entities, events, concepts, parameters, and attributes) through relationships, forming an organic variable knowledge network. The variable graph is constructed based on the design of the engine variable ontology. In this embodiment, a general graph construction tool can be used to construct multiple graphs using the engine's collected feature set, indicator system, variable set, and other elements as input. The graph is then guided through the steps to construct the graph.

[0031] Step 4: using all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data as spatial training data and inputting it into the skeleton variable map, and constructing an aircraft engine representation space based on the skeleton variable map using a machine learning algorithm; In this embodiment, first, based on the characteristic factors in the spatial training data, a machine learning algorithm is used to establish associations between the corresponding characteristic factors and the nodes of the skeleton variable map, and to establish associations between the characteristic factors; Taking the characteristic factors with correlation as the representation factors of the representation space, the representation factors associated with each node in the skeleton variable map and the corresponding correlation description are combined to form a representation model. All representation models are combined to form an aircraft engine representation space based on the skeleton variable map.

[0032] In actual application of this embodiment, the skeleton variable map can also be optimized according to the characterization traversal calculation and analysis results in combination with a deterministic rule model to obtain an optimized skeleton variable map; the deterministic rules include design rules, process rules and usage rules in engineering practice; the spatial training data is input into the optimized skeleton variable map to form an aerospace engine characterization space based on the optimized skeleton variable map.

[0033] Step 5: Use machine learning algorithms to perform characterization traversal calculation analysis on the aircraft engine characterization space to obtain the characterization traversal calculation analysis results.

[0034] After establishing the representation space, this embodiment can periodically input all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational use history data accumulated over a certain period of time into the representation space as incremental data to extract new characteristic factors from the incremental data. The representation space is then used as a traversal space, and the correlation threshold between the various representation factors in the representation space is used as the evolution direction. The newly added characteristic factors are added to the current representation model for dynamic combination to form different newly added representation models. A sensitivity analysis of each representation factor in the newly added representation model is performed, and the sensitivity analysis results are used as the basis for evaluating the effectiveness of the combination. New representation models whose sensitivity analysis results are greater than the preset limit value become valid new representation models. The valid new representation models are added to the representation space to participate in the representation traversal calculation and evolution.

[0035] Furthermore, the multi-dimensional characterization framework of aero-engines includes the engine product decomposition structure dimension, external characteristic parameter dimension, function / performance parameter dimension, physical characteristic dimension, process manufacturing dimension, environmental factor dimension, task sequence dimension, and fault failure dimension.

[0036] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A systematic multi-dimensional digital characterization method for aircraft engines, characterized by: include: Using the aircraft engine product decomposition structure as the backbone and the node objects at each layer of the structure as the carrier for the hierarchical organization of the representation system, multiple representation dimensions are recursively derived from the node objects at each layer of the structure. Based on the product hierarchy and inclusion relationships of the structure, a multi-dimensional representation framework for aircraft engines is constructed. A model analysis method is used to extract associated features from the models or data derived from each dimension to obtain characteristic factors for each dimension. The characteristic factors include design parameters, calculation parameters, and boundary conditions of corresponding aircraft engine components extracted from CAD models or CAE models, as well as variables, initial conditions, state vectors, input and output quantities, nodes, and logical relationship features extracted from differential equations, partial differential equations, state equations, and logical models. The calculation parameters include load parameters and material performance parameters in CAE models. The logical models include hierarchical models with tree structures expressing parent and child nodes, as well as many-to-many network models, data flow diagrams, and entity relationship diagrams. Using the characteristic factors extracted from each dimension as the variable ontology, semantic web technology is used to convert all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data into graph training data. Through machine learning training, a skeleton variable graph with the characteristic factors as the skeleton is formed. All or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data are input into the skeleton variable map as spatial training data, and an aircraft engine representation space based on the skeleton variable map is constructed using a machine learning algorithm; A machine learning algorithm is used to perform characterization traversal calculation and analysis on the aero-engine characterization space to obtain the characterization traversal calculation and analysis results.

2. The method for systematic multi-dimensional digital characterization of an aircraft engine according to claim 1, characterized in that: Also includes: According to the characterization traversal calculation analysis results, combined with a deterministic rule model, the skeleton variable map is optimized to obtain an optimized skeleton variable map; the deterministic rules include design rules, process rules and usage rules in engineering practice; Inputting the spatial training data into the optimized skeleton variable map to form an aero-engine representation space based on the optimized skeleton variable map; A machine learning algorithm is used to perform representation traversal calculation and analysis on the aero-engine representation space based on the optimized skeleton variable map to obtain the representation traversal calculation and analysis results.

3. The method for systematic multi-dimensional digital characterization of an aircraft engine according to claim 1, characterized in that: Methods for constructing an aero-engine representation space based on a skeleton variable graph using machine learning algorithms include: Based on the characteristic factors in the spatial training data, a machine learning algorithm is used to establish associations between the corresponding characteristic factors and the nodes of the skeleton variable map, and to establish associations between the characteristic factors; Taking the characteristic factors with correlation as the representation factors of the representation space, the representation factors associated with each node in the skeleton variable map and the corresponding correlation description are combined to form a representation model. All representation models are combined to form an aircraft engine representation space based on the skeleton variable map.

4. The method for systematic multi-dimensional digital characterization of an aircraft engine according to claim 3, characterized in that: Also includes: After establishing the representation space, all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data accumulated over a certain period of time are input into the representation space as incremental data to extract new characteristic factors from the incremental data. Taking the representation space as the traversal space, relying on the correlation threshold between each representation factor in the representation space as the evolution direction, adding new feature factors to the current representation model for dynamic combination to form different new representation models; Conduct sensitivity analysis on each characterization factor in the newly added characterization model, and use the sensitivity analysis results as the basis for evaluating the effectiveness of the combination. The newly added characterization model with a sensitivity analysis result greater than the preset limit value will be considered a valid newly added characterization model; Add effective new representation models to the representation space to participate in representation traversal calculation and evolution.

5. The method for systematic multi-dimensional digital characterization of an aircraft engine according to any one of claims 1 to 4, characterized in that: The characterization dimensions in the multi-dimensional characterization framework of aero-engines include engine product decomposition structure dimension, external characteristic parameter dimension, function / performance parameter dimension, physical characteristic dimension, process manufacturing dimension, environmental factor dimension, task sequence dimension, and fault failure dimension.

6. A systematic multi-dimensional digital characterization system for an aircraft engine, used to implement the systematic multi-dimensional digital characterization method for an aircraft engine according to claim 1, characterized in that: include: A representation framework construction module is used to build a multi-dimensional representation framework for aircraft engines, using the aircraft engine product decomposition structure as the backbone and the node objects at each layer of the aircraft engine product decomposition structure as the carrier for the hierarchical structured organization of the representation system. Multiple representation dimensions are recursively derived from the node objects at each layer of the aircraft engine product decomposition structure. This module is then used to build a multi-dimensional representation framework for aircraft engines, based on the product hierarchy and inclusion relationships of the aircraft engine product decomposition structure. A characteristic factor extraction module is used to extract associated features from models or data derived from each dimension using a model analysis method to obtain characteristic factors for each dimension. The characteristic factors include design parameters, calculation parameters, and boundary conditions of corresponding aircraft engine components extracted from CAD models or CAE models, as well as variables, initial conditions, state vectors, input and output quantities, nodes, and logical relationship features extracted from differential equations, partial differential equations, state equations, and logical models. The calculation parameters include load parameters and material performance parameters in CAE models. The logical models include hierarchical models with tree structures expressing parent and child nodes, as well as many-to-many network models, data flow diagrams, and entity relationship diagrams. The variable graph generation module is used to use the characteristic factors extracted from each dimension as the variable body. It uses semantic web technology to convert all or part of the aircraft engine design data, production data, simulation history data, test history data, and operation history data into graph training data. Through machine learning training, it forms a skeleton variable graph with the characteristic factors as the skeleton. a representation space generation module for inputting all or part of the aircraft engine design data, production data, simulation history data, test history data, and operational history data as spatial training data into the skeleton variable map, and constructing an aircraft engine representation space based on the skeleton variable map using a machine learning algorithm; The representation analysis module is used to perform representation traversal calculation analysis on the aircraft engine representation space using a machine learning algorithm to obtain the representation traversal calculation analysis results.

7. The systematic multi-dimensional digital characterization system for aircraft engines according to claim 6, characterized in that: The variable map generation module includes a variable map optimization unit for optimizing the skeleton variable map based on the characterization traversal calculation analysis results and in combination with a deterministic rule model to obtain an optimized skeleton variable map; the deterministic rules include design rules, process rules, and usage rules in engineering practice; The representation space generation module includes a representation space optimization unit, which is used to input spatial training data into the optimized skeleton variable map to form an aircraft engine representation space based on the optimized skeleton variable map; The representation analysis module is also used to use a machine learning algorithm to perform representation traversal calculation analysis on the aircraft engine representation space based on the optimized skeleton variable map to obtain the representation traversal calculation analysis results.

8. The systematic multi-dimensional digital characterization system for aircraft engines according to claim 6, characterized in that: The representation space generation module further includes: A correlation establishing unit is used to establish correlations between the corresponding characteristic factors and the nodes of the skeleton variable map based on the characteristic factors in the spatial training data, and to establish correlations between the characteristic factors; The characterization model combination unit is used to use characteristic factors with correlation as characterization factors of the characterization space, and to form a characterization model together with the characterization factors associated with each node in the skeleton variable map and the corresponding correlation description. All characterization models are combined to form an aircraft engine characterization space based on the skeleton variable map.

9. The systematic multi-dimensional digital characterization system for aircraft engines according to any one of claims 6 to 8, characterized in that: In the characterization framework construction module, the characterization dimensions in the aircraft engine multi-dimensional characterization framework include engine product decomposition structure dimension, external characteristic parameter dimension, function / performance parameter dimension, physical characteristic dimension, process manufacturing dimension, environmental factor dimension, task sequence dimension, and fault failure dimension.

Images