RefiningGPT: Specialized language Models for Automated Refinery Unit-level Process Diagram Synthesis

TL;DR

RefineGPT is a specialized AI model that automates the synthesis of refinery process diagrams by combining hierarchical LLM architectures and a new training pipeline, improving topological consistency and feasibility.

Contribution

The paper introduces RefineGPT, a novel domain-specific language model architecture and training pipeline for automated refinery process diagram synthesis.

Findings

RefineGPT improves topological consistency in process diagrams.

The training pipeline extracts process motifs from legacy data.

Empirical results show enhanced chemical engineering feasibility.

Abstract

Applying LLMs to complex industrial processes remains challenging due to the semantic gap between natural language design intents and the rigorous physical logic of engineering. In the field of petroleum refining engineering, a critical bottleneck is the automated synthesis of Unit-level Process Diagrams (UPDs), which serve as the topological bridge connecting abstract requirements to concrete unit operations. In this paper, we propose RefineGPT, a domain-specialized agent for autonomous refinery design.RefineGPT adopts a hierarchical architecture in which a supervised fine-tuned small language model is responsible for selecting units that satisfy design requirements, while a large language model is used to connect these units to generate the final topology. To enable supervised training, we develop a pipeline that extracts latent process motifs from noisy, unstructured legacy topologies and synthesizes high-quality rationale-based Chain-of-Thought (CoT) training data. Empirical validation demonstrates that RefineGPT achieves substantial improvements in topological consistency and chemical engineering feasibility, establishing a high-fidelity pathway for AI-augmented industrial process synthesis.