电力信息与通信技术2026，Vol.24Issue(8)：36-44,9.DOI:10.16543/j.2095-641x.electric.power.ict.2026.08.04
调度推理知识驱动的电力调度大模型训练微调框架
A Training and Fine-tuning Framework for Power Grid Dispatching Large Language Models Driven by Dispatching Reasoning Knowledge
摘要
Abstract
To address the problems of scarce domain reasoning data and insufficient model domain adaptability faced by power dispatching question answering in new-type power systems,this paper proposes a large model training and fine-tuning framework driven by power dispatching reasoning knowledge.First,a rule-generation combined data synthesis method for power dispatching question answering is proposed.By means of rule-guided and generative distillation strategies,a high-quality domain reasoning dataset covering key dispatching scenarios is constructed.Second,a multi-expert agent reward model oriented to multiple types of dispatching demands is designed to realize multi-dimensional comprehensive evaluation of model-generated content.Finally,a two-stage training strategy combining supervised fine-tuning and reinforcement learning is adopted to achieve in-depth domain adaptation for power dispatching based on mainstream foundation models.Experimental results show that the proposed method achieves excellent performance in core evaluation metrics on the dedicated test set,significantly outperforming current mainstream closed-source and open-source models.It effectively improves the logical accuracy and scenario practicality of the power dispatching question answering system,providing key technical support for the safe and stable operation of power grids.关键词
电力调度知识问答/推理驱动训练/两阶段微调Key words
power grid dispatching knowledge QA/reasoning-driven training/two-stage fine-tuning分类
信息技术与安全科学引用本文复制引用
姚铁锤,赵如意,张文思,常珂,柴博,王轶申..调度推理知识驱动的电力调度大模型训练微调框架[J].电力信息与通信技术,2026,24(8):36-44,9.基金项目
国家电网有限公司总部科技项目"面向电网调度智能辅助的大模型关键技术研究"(5108-202455044A-1-1-ZN). （5108-202455044A-1-1-ZN）
