Method and device for constructing aviation system research and development demand data management structure tree

Abstract

The invention belongs to the field of aircraft demand management, and particularly relates to an aviation system research and development demand data management structure tree construction method and device.The method comprises the steps that S1, an initial structure tree is constructed, the initial structure tree comprises multiple layers, and each layer is provided with multiple nodes; s2, extracting an input demand of each node; s3, extracting a keyword in the input demand, and converting the keyword into a first input vector; step S4, determining n node names and functions of the next layer included in the current node, and converting the n node names and functions into n second input vectors; and step S5, based on a pre-trained related node demand classification model, outputting n allocation demands as node contents of the current node. According to the method, the demand information is stored and managed in the form of the structure tree, the association relationship between the data is established, and the identifiability of the technical demand key information in the aviation complex product development process is improved.

Description

Technical Field

[0001] The present application belongs to the field of aircraft demand management, and in particular, relates to a method and device for constructing a structure tree for aviation system R&D demand data management. Background Art

[0002] The requirements of complex aviation products are divided into external stakeholder requirements and system technology requirements. External stakeholder requirements are customer development requirements, use requirements for pilots and ground crew, airworthiness requirements, and other legal and regulatory requirements. System technology requirements are the technical status baseline of the product, which is gradually formed and frozen during the product development process as the research and development level deepens.

[0003] Converting external stakeholder requirements into system technical requirements and decomposing them layer by layer to each level and device of the system is a technical problem that needs to be solved urgently. In addition, in the past, product R&D technical requirements were all carried out in various specification documents, such as system specification documents, development specification documents, and product specification documents. The documents record information in large paragraphs, making it difficult for designers or review experts to locate key information, resulting in ambiguity and inaccurate information transmission. Each document is compiled and maintained independently, resulting in unclear data evolution and decomposition processes and data traceability relationships. Summary of the invention

[0004] In order to solve the above problems, the present application provides a method and device for constructing an aviation system R&D demand data management structure tree.

[0005] The first aspect of the present application provides a method for constructing a structure tree for aviation system R&D demand data management, mainly comprising:

[0006] Step S1, constructing an initial structure tree, wherein the initial structure tree includes multiple layers, and each layer has multiple nodes;

[0007] Step S2: For each node, extract its input requirements, where the input requirements of the top-level nodes come from external inputs, and the input requirements of nodes in other layers are allocated by the upper layer;

[0008] Step S3: extract keywords from the input requirements and convert them into a first input vector;

[0009] Step S4, determining the names and functions of n nodes in the next layer included in the node, and converting them into n second input vectors;

[0010] Step S5: input the first input vector and multiple second input vectors into a pre-trained related node demand classification model, output n allocation requirements associated with the second input vectors as the node content of the current node, add them to the initial structure tree, and form an aviation system R&D demand data management structure tree.

[0011] Preferably, in step S1, the initial structure tree includes an equipment layer, a system layer and a device layer from top to bottom, the equipment layer covers equipment nodes to be developed for characterizing the aviation system, the equipment nodes to be developed are associated with multiple system nodes located in the system layer, and each system node is associated with multiple device nodes located in the device layer.

[0012] Preferably, step S5 further comprises:

[0013] Step S51: Assemble a requirement description including at least a subject and an action verb for the output allocation requirement, wherein the subject is a node name corresponding to a unit in a second input vector associated with the output unit, and the action verb is formed by an output vector of the output unit.

[0014] Preferably, step S5 further includes:

[0015] Step S6: present the allocation requirements to the designer, who will make corrections.

[0016] The second aspect of the present application provides a structure tree construction device for aviation system R&D demand data management, which mainly includes:

[0017] An initial structure tree building module is used to build an initial structure tree, wherein the initial structure tree includes multiple layers, each layer having multiple nodes;

[0018] The input requirement acquisition module is used to extract the input requirements of each node. The input requirements of the top-level nodes come from external inputs, and the input requirements of nodes in other layers are allocated by the upper layer.

[0019] A first input vector generating module, used to extract keywords from the input requirements and convert them into first input vectors;

[0020] A second input vector generating module is used to determine the names and functions of n nodes in the next layer included in the node, and convert them into n second input vectors;

[0021] The demand allocation module is used to input the first input vector and multiple second input vectors into a pre-trained related node demand classification model, output n allocation requirements associated with the second input vectors as the node content of the current node, add them to the initial structure tree, and form an aviation system R&D demand data management structure tree.

[0022] Preferably, in the initial structure tree building module, the initial structure tree includes an equipment layer, a system layer and a device layer from top to bottom, the equipment layer covers equipment nodes to be developed for characterizing the aviation system, the equipment nodes to be developed are associated with multiple system nodes located in the system layer, and each system node is associated with multiple device nodes located in the device layer.

[0023] Preferably, the demand allocation module includes:

[0024] The requirement description generating unit is used to assemble a requirement description containing at least a subject and an action verb for the output allocation requirement, wherein the subject is a node name corresponding to a unit in a second input vector associated with the output unit, and the action verb is formed by an output vector of the output unit.

[0025] Preferably, the device further comprises:

[0026] The correction module is used to present the allocation requirements to the designer, who will then make corrections.

[0027] This application stores and manages demand information in the form of a structure tree, establishes associations between data, improves the identifiability of key information on technical requirements in the development process of complex aviation products, and makes its traceability and associations intuitively visualized; facilitates the allocation of product technical requirements by comparing the organizational chart to organize development activities and reasonably arrange the storage location of outputs; facilitates the clarification of the organizational ownership of data, and at the same time can clarify the data transmission relationship between organizations. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flowchart of a preferred embodiment of the method for constructing a structure tree for aviation system R&D demand data management of the present application.

[0029] Figure 2 This application Figure 1 A schematic diagram of a structure tree for aviation system R&D requirement data management in the illustrated embodiment. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the implementation of this application will be described in more detail in combination with the drawings in the implementation of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described implementation is a part of the implementation of this application, not all of the implementations. The implementation described below with reference to the drawings is exemplary and is intended to be used to explain this application, and cannot be understood as a limitation on this application. Based on the implementation in this application, all other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The implementation of this application is described in detail below in combination with the drawings.

[0031] The first aspect of the present application provides a method for constructing a structure tree for aviation system R&D demand data management, such as Figure 1 As shown, it mainly includes:

[0032] Step S1, constructing an initial structure tree, wherein the initial structure tree includes multiple layers, and each layer has multiple nodes;

[0033] Step S2: For each node, extract its input requirements, where the input requirements of the top-level nodes come from external inputs, and the input requirements of nodes in other layers are allocated by the upper layer;

[0034] Step S3: extract keywords from the input requirements and convert them into a first input vector;

[0035] Step S4, determining the names and functions of n nodes in the next layer included in the node, and converting them into n second input vectors;

[0036] Step S5: input the first input vector and multiple second input vectors into a pre-trained related node demand classification model, output n allocation requirements associated with the second input vectors as the node content of the current node, add them to the initial structure tree, and form an aviation system R&D demand data management structure tree.

[0037] This application constructs an initial structure tree for managing requirements through step S1, and then sends the top-level requirements input from the outside to each node of the initial structure tree layer by layer, thereby completing the construction of the aviation system R&D requirement data management structure tree.

[0038] In step S1, in order to match the existing aircraft product breakdown structure, the constructed initial structure tree is formed completely in accordance with the aircraft product breakdown structure (PBS). For example, in some optional embodiments, in step S1, the initial structure tree includes an equipment layer, a system layer, and a device layer from top to bottom. The equipment layer covers equipment nodes to be developed for characterizing the aviation system. The equipment nodes to be developed are associated with multiple system nodes located in the system layer, and each system node is associated with multiple device nodes located in the device layer.

[0039] like Figure 2 As shown, the equipment layer forms the equipment-level technical requirements, the system layer forms the system-level technical requirements, and the equipment layer forms the equipment-level technical requirements. Furthermore, according to some aircraft decomposition structures, it can be divided into five more detailed levels, namely the top product layer, aircraft layer, system layer, subsystem layer, and subsystem layer / equipment layer. Then the system technical requirements of the project are divided into five levels, namely the top product layer requirements, aircraft layer requirements, system layer requirements, subsystem layer requirements, and subsystem layer / equipment layer requirements. Each layer consists of input technical requirements, technical requirements of this layer, and technical requirements output to the lower level. The number of technical requirements output to the lower level is less than or equal to the number of lower-level systems. In the technical requirements of the lowest-level equipment / component, there are no technical requirements output to the lower level.

[0040] It can be seen from steps S2 to S5 that the present application constructs a relevant node demand classification model for each node. The model uses the BP neural network algorithm. The input parameters are the input requirements of the node in the current layer and the node name and function of the next layer it contains, so as to allocate the input requirements to the nodes in the next layer, that is, the output parameters are multiple allocated requirements. The input parameters can be directly formed into vectors by extracting keywords and converting them from the language description, such as converting the demand description into the first input vector in step S3, and converting the system or device name and its functions into the second input vector in step S4. The output parameter design is relatively complex. In this example, for each output node, a set of multiple probabilities is constructed. Each probability in the set corresponds to the probability of using the function of the node. For example, for device A, it has three functions A1, A2, and A3, and its output value is [P A1 , P A2 , P A3 ], where P A1 is the probability of using the A1 function, P A2 is the probability of using the A2 function, P A3 is the probability of using function A3, and finally, according to the comparison relationship between the probability value and the threshold, it is determined which functions are allocated to device A.

[0041] refer to Figure 2, for the demand allocation model at system A, after receiving demand 201, the model internally believes that it can be converted into demands 301 and 302. After calculation, demand 301 should be allocated to device A.1, and demand 302 should be allocated to devices A.1 and A.2. That is, demand 401 is converted and allocated to device A.1, and stored in the technical requirements output to the lower-level device A.1. Demands 402 and 403 are converted and allocated to device A.2, and stored in the technical requirements output to the lower-level device A.2.

[0042] In some optional implementations, step S5 further includes:

[0043] Step S51: Assemble a requirement description including at least a subject and an action verb for the output allocation requirement, wherein the subject is a node name corresponding to a unit in a second input vector associated with the output unit, and the action verb is formed by an output vector of the output unit.

[0044] This embodiment is used to assemble functions. As mentioned above, although it is possible to determine which functions of device A are used by looking up a table, these functions are usually disorganized. For this reason, this example constructs a descriptive sentence for functional requirements, such as "When <status condition>, <subject> should <action verb><object><optional description item>." For example, the technical requirements of the top-level system: the product should be able to power the system independently according to maintenance and usage requirements. Technical requirements assigned to the A power supply system 1: System A should support the function of powering on the entire system; 2: System A should support the function of powering on system B independently. Technical requirements assigned to system B 1: System B should be able to control the power-on of other devices except those powered on with the product; 2: System B should be able to control the power-off of other devices except those powered on with the product.

[0045] In some optional implementations, step S5 further includes:

[0046] Step S6: present the allocation requirements to the designer, who will make corrections.

[0047] It is understandable that the present application uses the BP neural network algorithm for demand allocation, but the number of training samples is limited, so the output result may deviate greatly from the actual situation. Therefore, this embodiment ensures the reliability of demand allocation by adding manual correction.

[0048] The second aspect of the present application provides a structure tree construction device for aviation system R&D demand data management, which mainly includes:

[0049] An initial structure tree building module is used to build an initial structure tree, wherein the initial structure tree includes multiple layers, each layer having multiple nodes;

[0050] The input requirement acquisition module is used to extract the input requirements of each node. The input requirements of the top-level nodes come from external inputs, and the input requirements of nodes in other layers are allocated by the upper layer.

[0051] A first input vector generating module, used to extract keywords from the input requirements and convert them into first input vectors;

[0052] A second input vector generating module is used to determine the names and functions of n nodes in the next layer included in the node, and convert them into n second input vectors;

[0053] The demand allocation module is used to input the first input vector and multiple second input vectors into a pre-trained related node demand classification model, output n allocation requirements associated with the second input vectors as the node content of the current node, add them to the initial structure tree, and form an aviation system R&D demand data management structure tree.

[0054] In some optional embodiments, in the initial structure tree building module, the initial structure tree includes an equipment layer, a system layer and a device layer from top to bottom, the equipment layer covers equipment nodes to be developed for characterizing the aviation system, the equipment nodes to be developed are associated with multiple system nodes located in the system layer, and each system node is associated with multiple device nodes located in the device layer.

[0055] In some optional implementations, the demand allocation module includes:

[0056] The requirement description generating unit is used to assemble a requirement description containing at least a subject and an action verb for the output allocation requirement, wherein the subject is a node name corresponding to a unit in a second input vector associated with the output unit, and the action verb is formed by an output vector of the output unit.

[0057] In some optional embodiments, the device further comprises:

[0058] The correction module is used to present the allocation requirements to the designer, who will then make corrections.

[0059] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for constructing a structure tree for aviation system R&D demand data management, characterized in that: include: Step S1, constructing an initial structure tree, wherein the initial structure tree includes multiple layers, and each layer has multiple nodes; Step S2: For each node, extract its input requirements, where the input requirements of the top-level nodes come from external inputs, and the input requirements of nodes in other layers are allocated by the upper layer; Step S3: extract keywords from the input requirements and convert them into a first input vector; Step S4, determining the names and functions of n nodes in the next layer included in the node, and converting them into n second input vectors; Step S5: input the first input vector and multiple second input vectors into a pre-trained related node demand classification model, output n allocation requirements associated with the second input vectors as the node content of the current node, add them to the initial structure tree, and form an aviation system R&D demand data management structure tree.

2. The method for constructing a structure tree for aviation system R&D demand data management as claimed in claim 1, characterized in that: In step S1, the initial structure tree includes an equipment layer, a system layer and a device layer from top to bottom. The equipment layer covers equipment nodes to be developed for characterizing the aviation system. The equipment nodes to be developed are associated with multiple system nodes located in the system layer, and each system node is associated with multiple device nodes located in the device layer.

3. The method for constructing a structure tree for aviation system R&D demand data management as claimed in claim 1, characterized in that: Step S5 further comprises: Step S51: Assemble a requirement description including at least a subject and an action verb for the output allocation requirement, wherein the subject is a node name corresponding to a unit in a second input vector associated with the output unit, and the action verb is formed by an output vector of the output unit.

4. The method for constructing a structure tree for aviation system R&D demand data management as claimed in claim 1, characterized in that: Step S5 further includes: Step S6: present the allocation requirements to the designer, who will make corrections.

5. A structure tree construction device for aviation system R&D demand data management, characterized in that: include: An initial structure tree building module is used to build an initial structure tree, wherein the initial structure tree includes multiple layers, each layer having multiple nodes; The input requirement acquisition module is used to extract the input requirements of each node. The input requirements of the top-level nodes come from external inputs, and the input requirements of nodes in other layers are allocated by the upper layer. A first input vector generating module, used to extract keywords from the input requirements and convert them into first input vectors; A second input vector generating module is used to determine the names and functions of n nodes in the next layer included in the node, and convert them into n second input vectors; The demand allocation module is used to input the first input vector and multiple second input vectors into a pre-trained related node demand classification model, output n allocation requirements associated with the second input vectors as the node content of the current node, add them to the initial structure tree, and form an aviation system R&D demand data management structure tree.

6. The device for constructing a structure tree for aviation system R&D demand data management as claimed in claim 5, characterized in that: In the initial structure tree building module, the initial structure tree includes an equipment layer, a system layer and a device layer from top to bottom. The equipment layer covers equipment nodes to be developed for characterizing the aviation system. The equipment nodes to be developed are associated with multiple system nodes located in the system layer, and each system node is associated with multiple device nodes located in the device layer.

7. The device for constructing a structure tree for aviation system R&D demand data management as claimed in claim 5, characterized in that: The demand allocation module comprises: The requirement description generating unit is used to assemble a requirement description containing at least a subject and an action verb for the output allocation requirement, wherein the subject is a node name corresponding to a unit in a second input vector associated with the output unit, and the action verb is formed by an output vector of the output unit.

8. The device for constructing a structure tree for aviation system R&D demand data management as claimed in claim 5, characterized in that: The device also includes: The correction module is used to present the allocation requirements to the designer, who will then make corrections.

Images