Industrial Simulation Model Generation Using LLM Ontology Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The integration of data and simulation models between different software components in industrial plants is hindered by the complexity of proprietary formats and the limitations of existing standards, leading to high manual effort and inefficiencies in model conversion and interoperability.
Innovation Solution
A method utilizing Large Language Models (LLMs) combined with ontologies and a refinement module to automatically refine and supplement textual information to generate simulation models that comply with industry standards, such as IFC and URDF, by comparing and aligning the generated text with standard ontologies to ensure completeness, correct relationships, and normalization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If data and simulation models are converted between different proprietary software formats, then interoperability between software components is achieved, but high manual effort and complexity are required
Solution Approach 1:
The patent introduces an intermediary conversion system that uses Large Language Models to automatically transform data and simulation models between different proprietary software formats. This intermediary layer handles the complex conversion processes, eliminating the need for manual format transformations and reducing both effort and complexity while maintaining interoperability.
Solution Approach 2:
The patent replaces manual mechanical conversion processes with an automated AI-based system. Large Language Models substitute the manual effort previously required for format conversion, automatically understanding and transforming data structures between different software components without human intervention.
2Manufacturing precision
If models are recreated from scratch based on original requirements, then model accuracy is improved, but development time and effort increase significantly
Solution Approach 1:
The patent applies preliminary action by using Large Language Models to automatically generate accurate simulation models from existing data models and requirements specifications. The AI system performs the model creation work in advance, maintaining high accuracy by understanding the semantic relationships in the requirements, while dramatically reducing the time and effort compared to manual model recreation.
Solution Approach 2:
The system creates accurate copies of the intended model behavior by using AI to interpret requirements and generate corresponding simulation models. The Large Language Models capture the essential characteristics and relationships from the source requirements and replicate them accurately in the target simulation model format without requiring manual recreation.
Data Source
Figure 1~3
Figure 4~5
Figure 6
AI summary
Industrial processes surrounding this industrial plant typically comprise numerous logically interconnected process steps and are characterized by a high degree of complexity with regard to the assets involved and the processes applied, both for plant operation and for construction and maintenance. New data and simulation models have generally been created based on existing solutions and the addition of new requirements. Therefore, the object of the invention is to propose an automated system that progressively refines and supplements the textual information generated by the LLM (Large Load Management) to automatically generate a simulation model. A suitable means of achieving this is through standards, or more precisely, the ontologies contained in many standards such as IFC and DEXPI.