CN-121979494-A - IVCPS demand generation method and device, electronic equipment and storage medium

CN121979494ACN 121979494 ACN121979494 ACN 121979494ACN-121979494-A

Abstract

The embodiment of the application provides a method, a device, electronic equipment and a storage medium for generating requirements of IVCPS, wherein the method comprises the steps of establishing ontology models of different elements, generating corresponding knowledge maps according to the ontology models, correlating the different knowledge maps to obtain IVCPS requirement element knowledge base, further after obtaining target element requirements, matching the IVCPS requirement element knowledge base to obtain corresponding entity nodes, generating target requirement links according to the entity nodes, analyzing the target requirement links according to a trained large model to obtain target description language corresponding to the target requirement links, and thus obtaining target requirement data corresponding to target element requirements, generating corresponding target description language according to target element requirements of one party in stakeholders, integrating constraint conditions of multiple parties, realizing full coverage of regional elements, reducing and eliminating requirement omission, and improving requirement analysis efficiency.

Inventors

LI CHUNYANG
ZHANG XIAODAN
HOU DEZAO
SUN MING

Assignees

交通运输部公路科学研究所

Dates

Publication Date: 20260505
Application Date: 20260407

Claims (12)

1. A method of demand generation of IVCPS, the method comprising: acquiring a target element requirement; Determining entity nodes corresponding to the target element requirements according to a pre-established IVCPS requirement element knowledge base, wherein the IVCPS requirement element knowledge base is obtained by establishing ontology models of different elements, generating corresponding knowledge maps according to the ontology models, and correlating the different knowledge maps; Generating a target demand link corresponding to the entity node according to the entity node; Analyzing the target demand link by adopting a pre-trained large model to obtain a target description language corresponding to the target demand link, and determining the target description language as target demand data corresponding to the target element demand.
2. The demand generation method of claim 1, wherein the obtaining the target element demand includes: acquiring a target element requirement; Carrying out structural processing on the text data of the target element requirement to obtain a processed target element requirement; and carrying out semantic annotation on the processed target requirement to obtain retrievable requirement text data.
3. The method for generating a demand according to claim 2, wherein the determining the entity node corresponding to the target element demand according to the pre-established IVCPS demand element knowledge base includes: Determining a first feature vector of the retrievable required text data; respectively calculating the similarity of the first feature vector and a second feature vector of an entity node in the IVCPS required element knowledge base; If the similarity is greater than a similarity threshold, determining that the retrievable required text data matches the entity node; And determining attribute information corresponding to the retrievable demand text data according to the pre-established IVCPS demand element knowledge base, wherein the pre-established IVCPS demand element knowledge base comprises an entity and attribute information corresponding to the entity.
4. A method for generating a demand according to claim 3, wherein generating a target demand link corresponding to the entity node according to the entity node comprises: According to a pre-established demand association mode, a general rule reasoning engine under a semantic web framework is adopted, entity types and semantic relations defined by a demand ontology are used as basic constraints, and an RDF triplet form is adopted to match the entity nodes with the pre-established IVCPS demand element knowledge base, so that a first demand link corresponding to the entity nodes is obtained; Constructing an implicit demand relation of description logic according to class levels, attribute constraints and inheritance relations which are defined in a demand body, and generating a second demand link corresponding to the entity node according to the implicit demand relation of the description logic; Determining a probability corresponding to the entity node by adopting a graph rolling network model according to the knowledge graph, the embedded vector of the entity node and the semantic relation type based on a demand relation completion algorithm of the graph rolling network model GCN, and determining a third demand link corresponding to the entity node according to the probability; And determining a target demand link corresponding to the entity node according to the first demand link, the second demand link and the third demand link.
5. The method of claim IVCPS of claim 4, wherein the determining a target demand link corresponding to the entity node according to the first demand link, the second demand link, and the third demand link comprises: and determining a target demand link from a plurality of demand links corresponding to the entity nodes by adopting a node importance calculation method, a subgraph connectivity analysis method and a link importance aggregation method.
6. The method of claim IVCPS, wherein the parsing the target demand link with the pre-trained large model to obtain a target description language corresponding to the target demand link and determining the target description language as target demand data corresponding to the target element demand comprises: Converting a demand link corresponding to the entity node according to a pre-established domain semantic constraint library to obtain standardized expression data corresponding to the demand link, wherein the pre-established domain semantic constraint library comprises term mapping rules, logic association rules and description specification rules; Inputting a preset prompt word, a demand link sequence in the standardized expression data and associated node attributes into a large model to obtain a target demand corresponding to the demand link, wherein the preset prompt word at least comprises a top-level target prompt word, a scene prompt word, a stakeholder prompt word, an entity prompt word, a function service prompt word, a relationship prompt word and a performance prompt word.
7. The method of claim IVCPS, wherein the IVCPS requirement element knowledge base is obtained by: According to standard specifications, policy documents, technical white books, scientific achievements and typical application cases, a Stanford seven-step method and a cyclic iteration method are adopted for processing, and an ontology model is generated; extracting entities and relations from the ontology model by adopting an entity and relation extraction algorithm to obtain a knowledge graph corresponding to the ontology model; adopting a graph embedding method to convert entity nodes in the knowledge graph corresponding to the ontology model into embedded vectors corresponding to the entity nodes; calculating the association relationship between knowledge maps corresponding to two different ontology models by adopting a hierarchical clustering algorithm, a point similarity measurement method and community discovery; and obtaining the IVCPS required element knowledge base according to the plurality of knowledge maps and the association relation among the plurality of knowledge maps.
8. The method of claim IVCPS, wherein the transforming the entity node in the knowledge-graph corresponding to the ontology model into the embedded vector corresponding to the entity node by using a graph embedding method comprises: And generating an embedded vector of an entity node in the knowledge graph corresponding to the ontology model by adopting a BGE text vector model, wherein the BGE text vector model adopts a Transformer Encoder-based double-tower structure, and the context semantic relationship is modeled through a multi-layer self-attention mechanism after text input is segmented and embedded.
9. A demand generation apparatus IVCPS, the apparatus comprising: the acquisition module is used for acquiring the target element requirement; The determining module is used for determining entity nodes corresponding to the target element requirements according to a pre-established IVCPS requirement element knowledge base, wherein the IVCPS requirement element knowledge base is obtained by establishing ontology models of different elements, generating corresponding knowledge maps according to the ontology models and correlating the different knowledge maps; The generation module is used for generating a target demand link corresponding to the entity node according to the entity node; the analysis module is used for analyzing the target demand link by adopting a pre-trained large model to obtain a target description language corresponding to the target demand link, and determining the target description language as target demand data corresponding to the target element demand.
10. The demand generation apparatus of claim IVCPS, wherein the acquisition module is configured to: acquiring a target element requirement; Carrying out structural processing on the text data of the target element requirement to obtain a processed target element requirement; and carrying out semantic annotation on the processed target requirement to obtain retrievable requirement text data.
11. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the program, implements the demand generation method of IVCPS of any one of claims 1-8.
12. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and wherein the program is executed by a processor to implement the demand generation method of IVCPS according to any one of claims 1 to 8.

Description

IVCPS demand generation method and device, electronic equipment and storage medium Technical Field The application relates to the technical field of intelligent transportation, in particular to a IVCPS demand generation method, a IVCPS demand generation device, electronic equipment and a storage medium. Background IVCPS (intelligent automobile information physical system) is used as a complex system for deep fusion of vehicles, roads, clouds, networks and diagrams, and needs to meet the dynamic requirements of multi-field coordination (vehicle control, road infrastructure, cloud computing platform, communication network and high-precision map) and multi-benefit relatives (road manager, vehicle manufacturer, autopilot operator, driver, passengers, pedestrians and non-motor vehicles) at the same time, and in the development process of the intelligent automobile information physical system (IVCPS), the requirement analysis is a core link for defining system functions, performance indexes and safety constraints. However, in the prior art, the demand analysis method generally depends on a static flow which is manually dominant, and cannot dynamically adapt to real-time changes (such as road conditions, user behaviors and sensor data fluctuation) of the intelligent automobile environment. That is, if the demand of a certain executive is changed and needs to be limited by the capacity of other executive, that is, the demand analysis is only based on unilateral input (such as a demand document provided by a vehicle manufacturer), the demands of other stakeholders cannot be considered, that is, a cross-domain and cross-stakeholder dynamic coordination mechanism is not established, which results in one-sided demand design, unrecognized conflict and delayed decision, and finally, system design deviation is caused, for example, automatic driving decision is not included due to the safety demand of pedestrians, resulting in accident risk, prolonged system development period and increased cost. Disclosure of Invention An object of some embodiments of the present application is to provide IVCPS a method, an apparatus, an electronic device, and a storage medium for generating a demand link corresponding to a target element by acquiring a target element demand through a technical solution of an embodiment of the present application, determining entity nodes corresponding to the target element demand according to a pre-established IVCPS demand element knowledge base, wherein the IVCPS demand element knowledge base is obtained by establishing an ontology model of different elements, generating a corresponding knowledge map according to the ontology model, and correlating the different knowledge maps, generating a target demand link corresponding to the entity nodes according to the entity nodes, analyzing the target demand link by using a pre-trained large model to obtain a target description language corresponding to the target demand link, determining the target description language as target demand data corresponding to the target element demand, and automatically obtaining a corresponding demand link according to a target demand profile by establishing a ontology model of different elements, generating a corresponding knowledge map according to the ontology model, correlating the different element knowledge base IVCPS, further obtaining a physical demand link corresponding to the target element demand 34 after acquiring the target element demand, automatically obtaining a corresponding requirement link, and automatically obtaining a target demand link, and automatically obtaining a corresponding requirement information according to a target demand system, and a target element, and a target interest, automatically obtaining a target interest, and a target interest, according to a corresponding requirement system, and a target element, and a target interest, according to a target element demand information, and a target interest, according to a target element, and a target interest, and a target element, and a target system according to a requirement system, and a target element requirement system, and a corresponding system, the functional requirements of each level, the performance requirements of each level and the design constraint requirements are combined, constraint conditions of multiple parties are integrated, corresponding target description languages are generated, the whole coverage of area elements is realized, the requirement omission is reduced and eliminated, and the requirement analysis efficiency is improved. In a first aspect, some embodiments of the present application provide a method for generating a demand of IVCPS, including: acquiring a target element requirement; determining entity nodes corresponding to the target element requirements according to a pre-established IVCPS requirement element knowledge base, wherein the IVCPS requirement element knowledge base is obtained by establishing ontology mo