ProOPF: Benchmarking and Improving LLMs for Professional-Grade Power Systems Optimization Modeling

2026-02-03

AI Summary

This work addresses the operational uncertainties introduced by high-penetration renewable energy integration and the lack of high-quality evaluation benchmarks for large language models (LLMs) in professional-grade optimal power flow (OPF) modeling. To bridge this gap, the authors propose the NL-to-OPF framework, which integrates natural language processing, LLMs, and domain expertise to automatically generate and verify executable OPF code from natural-language descriptions of dispatch requirements. The core contribution is the first-ever OPF modeling dataset and benchmark tailored to power system scenarios—ProOPF-D and ProOPF-B—comprising 12K training samples and 121 expert-annotated test cases. This benchmark enables end-to-end evaluation of both concrete and abstract OPF modeling tasks, establishing a rigorous and reproducible standard for assessing LLM capabilities in power system optimization.

Technology Category

• Natural Language Processing: Generation
• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Growing renewable penetration introduces substantial uncertainty into power system operations, necessitating frequent adaptation of dispatch objectives and constraints and challenging expertise-intensive, near-real-time modeling workflows. Large Language Models (LLMs) provide a promising avenue for automating this process by translating natural-language (NL) operational requirements into executable optimization models via semantic reasoning and code synthesis. Yet existing LLM datasets and benchmarks for optimization modeling primarily target coarse-grained cross-domain generalization, offering limited, rigorous evaluation in power-system settings, particularly for Optimal Power Flow (OPF). We therefore introduce \textbf{ProOPF-D} and \textbf{ProOPF-B}, a dataset and benchmark for professional-grade OPF modeling: ProOPF-D contains 12K instances pairing NL requests with parameter adjustments and structural extensions to a canonical OPF, together with executable implementations; ProOPF-B provides 121 expert-annotated test cases with ground-truth code, enabling end-to-end evaluation under both concrete and abstract OPF modeling regimes.

Problem

• Optimal Power Flow
• Large Language Models
• Power Systems Optimization
• Natural Language to Code
• Benchmarking

Innovation

• Large Language Models
• Optimal Power Flow
• Power Systems Optimization
• Natural Language to Code
• Benchmark Dataset