Integrating Ontologies with Large Language Models for Enhanced Control Systems in Chemical Engineering

2025-10-30

AI Summary

To address the insufficient interpretability and auditability of large language models (LLMs) in chemical process control systems, this paper proposes an ontology-driven LLM enhancement framework. Methodologically, it leverages the COPE ontology as a knowledge backbone to orchestrate a synergistic pipeline comprising semantic preprocessing, ontology mapping, template-based fine-tuning, semantic retrieval, and iterative validation. Innovatively, it introduces a control decoding mechanism and a citation gating strategy to enforce strict adherence of model outputs to ontology-defined terms and logical constraints. The key contribution is the first deep integration of a domain-specific ontology with LLMs for process control and safety analysis tasks—achieving high linguistic fluency while significantly improving ontology accuracy (+23.6%) and inference traceability. This work establishes a verifiable, auditable neuro-symbolic paradigm tailored for high-reliability engineering AI applications.

Technology Category

• Natural Language Processing: Safety and Robustness
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Ontologies

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

This work presents an ontology-integrated large language model (LLM) framework for chemical engineering that unites structured domain knowledge with generative reasoning. The proposed pipeline aligns model training and inference with the COPE ontology through a sequence of data acquisition, semantic preprocessing, information extraction, and ontology mapping steps, producing templated question-answer pairs that guide fine-tuning. A control-focused decoding stage and citation gate enforce syntactic and factual grounding by constraining outputs to ontology-linked terms, while evaluation metrics quantify both linguistic quality and ontological accuracy. Feedback and future extensions, including semantic retrieval and iterative validation, further enhance the system's interpretability and reliability. This integration of symbolic structure and neural generation provides a transparent, auditable approach for applying LLMs to process control, safety analysis, and other critical engineering contexts.

Problem

• Integrating ontologies with LLMs for chemical engineering control systems
• Combining symbolic knowledge with generative reasoning for process control
• Developing transparent LLM applications for safety-critical engineering contexts

Innovation

• Integrating ontologies with large language models
• Fine-tuning with templated question-answer pairs
• Constraining outputs to ontology-linked terms