🤖 AI Summary
This work proposes RefineGPT, a domain-specific intelligent agent for the automated synthesis of unit-level process diagrams (UPDs) in petroleum refining, addressing the semantic gap between natural language design intents and the rigorous physical logic of refining engineering. The approach employs a hierarchical architecture wherein a small language model selects appropriate process units based on design requirements, while a large language model constructs the complete process topology. Innovatively, the method introduces a pipeline that extracts implicit process patterns from unstructured historical diagrams to synthesize high-quality, chain-of-reasoning–based training data, thereby deeply integrating domain knowledge with large-model capabilities. Experimental results demonstrate that RefineGPT significantly outperforms existing methods in both topological consistency and chemical feasibility, offering a high-fidelity pathway toward AI-driven industrial process synthesis.
📝 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.