中国电机工程学报2026，Vol.46Issue(10)：3967-3980,中插4,15.DOI:10.13334/j.0258-8013.pcsee.252646
面向电力任务的预训练LLM适配方法:多源异构数据表征学习与有监督微调
Adaptation Approach for Pre-trained LLMs towards Power System Tasks:Representation Learning and Supervised Fine-tuning With Multi-source Heterogeneous Data
摘要
Abstract
Large language model(LLM)has demonstrated substantial application potential in power systems.However,conventional text-based data input struggles to accurately capture the inherent structural characteristics and numerical precision of power system data,thereby limiting their effectiveness in power-related tasks.To address this issue,this paper proposes an adaptation method for pre-trained LLMs tailored to power system tasks,integrating multi-source heterogeneous data representation learning with supervised fine-tuning.First,power system data are categorized into four typical forms,snapshot measurements,time series,graph structures,and textual descriptions,and the characteristic attributes of each type are analyzed.Then,a unified representation framework for multi-source heterogeneous data is constructed,in which feature extraction and semantic alignment techniques are employed to achieve efficient and lossless mapping of heterogeneous data into the semantic space of the LLM.Finally,experimental results from load forecasting and distribution network state estimation demonstrate that effectively leveraging pre-trained LLMs significantly improves task accuracy.In the load forecasting task,the proposed method reduces prediction error by 8.0％on average compared to traditional methods.In the distribution network state estimation task,the proposed method with only 8％measurement ratio achieves the same estimation accuracy as baseline methods with 16％measurement ratio,while the training process can be completed on consumer-grade graphics processing units.关键词
电力大模型/多源异构数据/特征提取/表征学习Key words
power system large language models/multi-source heterogeneous data/feature extraction/representation learning分类
信息技术与安全科学引用本文复制引用
高明阳,周苏洋,顾伟,吴志,樊继利,周爱华,彭林,刘梅招..面向电力任务的预训练LLM适配方法:多源异构数据表征学习与有监督微调[J].中国电机工程学报,2026,46(10):3967-3980,中插4,15.基金项目
智能电网国家科技重大专项(2024ZD0802200) （2024ZD0802200）
国家电网有限公司科技项目(5700-202458232A-1-1-ZN).Smart Grid National Science and Technology Major Project(2024ZD0802200) （5700-202458232A-1-1-ZN）
Science and Technology Project of State Grid Corporation of China(5700-202458232A-1-1-ZN). （5700-202458232A-1-1-ZN）
