10 results about "Design knowledge" patented technology

This invention discloses a temporally enhanced knowledge graph-driven dialogue agent and its implementation method, belonging to the fields of artificial intelligence and natural language processing technology. The dialogue agent includes a dialogue interface, natural language understanding, a temporal knowledge graph, a temporal reasoning engine, dynamic update and failure management, and a natural language generation module. It includes designing the basic representation unit of the knowledge graph as a temporal quintuple; designing a temporal reasoning engine and a dynamic update and failure management method; designing a method for matching, verifying, and performing multi-hop reasoning of temporal information in natural language problems with temporal facts in the knowledge graph; achieving accurate parsing of temporal constraints in dialogue, real-time adaptation of dynamic knowledge, rigorous temporal logic reasoning, and continuous maintenance of contextual temporal consistency; and outputting a temporally accurate, knowledge-reliable, and logically coherent intelligent response, capable of meeting the needs of complex scenarios such as vertical industry decision support and high-reliability intelligent interaction.

An automatic ship design system and method using a conversational artificial intelligence model are disclosed. According to an embodiment of the present invention, the automatic ship design system using a conversational artificial intelligence model includes a memory in which one or more instructions, ship design knowledge information, and the conversational artificial intelligence model are stored; a communication unit for communication with a user terminal; and a processor configured to communicate with the user terminal via the communication unit and execute the one or more instructions stored in the memory, wherein the processor is configured to perform a conversation with the user through the user terminal by using the conversational artificial intelligence model, receive ship design variables through the conversation, and derive ship design information in real time using the ship design knowledge information according to the received ship design variables.

The application discloses a user demand recognition method and device based on a knowledge graph constraint and a large model, the method comprising: performing preset processing on multi-modal user metadata corresponding to a product to be analyzed to obtain a unified intermediate representation; inputting the unified intermediate representation into a multi-modal large model to generate candidate user demand feature items and candidate design element items and corresponding confidence degrees; mapping the candidate user demand feature items and the candidate design element items to corresponding nodes in a preset industrial design knowledge graph for constraint consistency processing to obtain a candidate result set that has passed constraint verification; if the candidate result set meets a retrieval trigger condition, retrieving technical document segments corresponding to the candidate result set from an external evidence library, and binding the retrieval results with the candidate result set; and outputting a structured result record according to a preset structured template, wherein the structured result record at least includes a demand result field, an evidence index field, a knowledge graph node reference field and a constraint verification marker field.

The application discloses a kind of engineering design knowledge retrieval method and system based on knowledge graph, comprising the following steps: engineering design data preprocessing is collected;Entity recognition and relationship extraction operation are executed, and engineering entity node set and relationship edge set are generated;Constraint node set and constraint dependency relationship set are constructed according to constraint rule information to generate constraint dependency graph;Receiving engineering design knowledge retrieval request and executing semantic analysis operation;In the improved NBFNet model, candidate domain expansion propagation, constraint coupling propagation and target convergence propagation calculation are executed to generate target matching candidate node set;Path consistency calculation is carried out on the reasoning path corresponding to the target matching candidate node set to generate path score set;Engineering design knowledge retrieval result is generated according to path score set ordering and output reasoning path.The application realizes efficient retrieval of engineering design knowledge, improves retrieval accuracy and reasoning path explainability.

This invention relates to a knowledge graph-based method for rapid reconfiguration design of personalized production systems. The method includes: acquiring historical engineering data related to products, processes, and equipment; performing knowledge fusion to establish a global-domain hierarchical ontology, forming a production system design knowledge graph; user-defined new product personalization requirements as task requirements for production system design reconfiguration; retrieving the design knowledge graph through processes such as requirement decomposition, equipment selection, and integration verification to obtain equipment integration scheme results; and completing the mechanical, electrical, and automation design of the production system based on the equipment integration scheme results. This invention automatically builds a production system design knowledge graph using engineering data. When product personalization requirements change, a design scheme meeting the new requirements is obtained through knowledge graph retrieval, enabling multidisciplinary collaborative design of the production system's mechanical, electrical, and automation components, and automatically verifying whether the scheme meets the requirements.

The application discloses a bulk material transportation method and system based on a knowledge graph and multiple agents, and a storage medium, and the method comprises the following steps: constructing a standard correlation graph and a corresponding conversion path by using multiple source data; obtaining sample data to fine-tune a large model and construct a system intelligent design knowledge base model. In the AI application stage, the multiple agent cooperation mechanism is used to respond to user demand: firstly, the natural language design demand input by the user is received, and semantic analysis is performed on the large model; then, the product definition content, the key parameters and the detailed parameters are generated in sequence by combining the historical data and the knowledge base, and the user confirmation mechanism is introduced in each parameter generation link to ensure accuracy; finally, the design drawing is automatically generated by integrating and generating data, the business service is called to store the design results and generate relevant business objects, the processing result is fed back to the client, and the whole-process intelligent auxiliary design from the demand input to the drawing generation is realized.

This invention relates to a method and system for generating site-specific sustainable architectural design. The method first acquires architectural design text and case images to construct expected and unexpected sample sets. Then, a large language model is used to process the expected sample set to construct a knowledge graph for site-specific sustainable architectural design. Simultaneously, expert and public perceptions are extracted from the sample set to generate feature heatmaps. These heatmaps are then analyzed by the large language model with the knowledge graph to extract design elements and design language conducive to site-specific sustainable architectural design, constructing a knowledge base for site-specific sustainable architectural design. Subsequently, a LoRA model is trained based on this knowledge base, a supporting prompt word system is constructed, and an evaluation model is attached to create a knowledge-guided AIGC workflow for site-specific sustainable architectural design. Finally, the current status image of the target building is input into the workflow, and the output results are filtered to obtain the site-specific sustainable design scheme for the target building. Compared with existing technologies, this invention has advantages such as improved design generation efficiency and quality.

The application provides a fracturing design report generation method and device, comprising: performing feature extraction and labeling on obtained historical fracturing engineering design and construction data and fracturing engineering design well related basic data to obtain a fracturing design knowledge base; using the fracturing design knowledge base to construct a fracturing design report generation agent in a multi-agent cooperation manner; inputting a fracturing engineering design well prompt word into the fracturing design report generation agent to obtain a fracturing design report. The application can improve the accuracy and efficiency of generating a fracturing design report based on knowledge graph, large model intelligent fine-tuning and model intelligent optimization technology.

The present application belongs to the technical field of digital design of nuclear power equipment, and particularly relates to a steam generator throttling component parameterization design platform and method. The platform comprises: a three-dimensional parameterization modeling module, containing parameterization templates of pre-constructed multi-stage eccentric orifice plates, gear flow disturbers and T-shaped orifice plates, main parameters and model features being associated through rules; a resistance coefficient associated calculation module, used for iterative correction according to input target resistance coefficients, combining a fitting model and an optimized Realizable kε turbulent flow model, and outputting high-precision structural parameters; an automatic drawing module, generating two-dimensional engineering drawings conforming to ASME / GB / T standards based on three-dimensional models; a design knowledge base and data management module, providing case matching and parameter reuse based on resistance coefficient similarity. The method comprises parameter input verification, structural parameter calculation, three-dimensional model generation, automatic engineering drawing generation and data archiving. The present application shortens the design cycle from 2 weeks to 3 days, controls the resistance coefficient calculation error within 3%, and realizes full-flow automatic and intelligent throttling component design.

This invention discloses an automatic generation method for power design drawings based on knowledge graphs, comprising: Step 1: Constructing a power design knowledge graph; Step 2: Obtaining power design input parameters and generating drawing assembly task seeds; Step 3: Extracting conditional subgraphs and performing breadth-first traversal to generate node insertion sequences and port connection sequences, inputting an improved GraphRNN model with a port-driven mapping mechanism to generate an atomic assembly chain graph; Step 4: Constructing a drawing skeleton mesh and forming a slot mapping structure; Step 5: Constructing a slot conflict matrix and performing slot adjustment to obtain the target assembly sequence; Step 6: Generating a primitive connection script and performing primitive connection to obtain the power design drawings; Step 7: Performing verification and backtracking updates to output the final power design drawings. This invention achieves automatic generation and structural consistency optimization of power design drawings.