A large language model for advanced power dispatch

Abstract Power dispatch is essential for providing society with stable, cost-effective, and eco-friendly electricity. However, traditional methods falter as power systems grow in scale and complexity, struggling with multitasking, swift problem-solving, and human-machine collaboration. This paper in...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuheng Cheng, Huan Zhao, Xiyuan Zhou, Junhua Zhao, Yuji Cao, Chao Yang, Xinlei Cai
Format: Article
Language:English
Published: Nature Portfolio 2025-03-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-025-91940-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850039600431497216
author Yuheng Cheng
Huan Zhao
Xiyuan Zhou
Junhua Zhao
Yuji Cao
Chao Yang
Xinlei Cai
author_facet Yuheng Cheng
Huan Zhao
Xiyuan Zhou
Junhua Zhao
Yuji Cao
Chao Yang
Xinlei Cai
author_sort Yuheng Cheng
collection DOAJ
description Abstract Power dispatch is essential for providing society with stable, cost-effective, and eco-friendly electricity. However, traditional methods falter as power systems grow in scale and complexity, struggling with multitasking, swift problem-solving, and human-machine collaboration. This paper introduces Grid Artificial Intelligent Assistant (GAIA), a pioneering Large Language Model (LLM) designed to assist with a variety of power system operational tasks, including operation adjustment, operation monitoring, and black start scenarios. We have developed a novel dataset construction technique that harnesses various data sources to fine-tune GAIA for optimal performance in this domain. This approach streamlines LLM training, allowing for the seamless integration of multidimensional data in power system management. Additionally, we have crafted specialized prompt strategies to boost GAIA’s input-output efficiency in dispatch scenarios. When evaluated on the ElecBench benchmark, GAIA surpasses the baseline model Large Language Model Meta AI-2 (LLaMA2) on multiple metrics. In practical applications, GAIA has demonstrated its ability to enhance decision-making processes, improve operational efficiency, and facilitate better human-machine interactions in power dispatch operations. This paper expands the application of LLMs to power dispatch and validates their practical utility, paving the way for future innovations in this field.
format Article
id doaj-art-e85a01e034134dbd824be83c7e1ba277
institution DOAJ
issn 2045-2322
language English
publishDate 2025-03-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e85a01e034134dbd824be83c7e1ba2772025-08-20T02:56:16ZengNature PortfolioScientific Reports2045-23222025-03-0115111710.1038/s41598-025-91940-xA large language model for advanced power dispatchYuheng Cheng0Huan Zhao1Xiyuan Zhou2Junhua Zhao3Yuji Cao4Chao Yang5Xinlei Cai6Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS)Department of Building Environment and Energy Engineering, Hong Kong Polytechnic UniversitySchool of Electrical and Electronic Engineering, Nanyang Technological UniversityShenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS)Department of Mechanical and Automation Engineering, Chinese University of Hong KongSchool of Electrical and Electronic Engineering, North China Electric Power UniversityChina Southern Power Grid (China)Abstract Power dispatch is essential for providing society with stable, cost-effective, and eco-friendly electricity. However, traditional methods falter as power systems grow in scale and complexity, struggling with multitasking, swift problem-solving, and human-machine collaboration. This paper introduces Grid Artificial Intelligent Assistant (GAIA), a pioneering Large Language Model (LLM) designed to assist with a variety of power system operational tasks, including operation adjustment, operation monitoring, and black start scenarios. We have developed a novel dataset construction technique that harnesses various data sources to fine-tune GAIA for optimal performance in this domain. This approach streamlines LLM training, allowing for the seamless integration of multidimensional data in power system management. Additionally, we have crafted specialized prompt strategies to boost GAIA’s input-output efficiency in dispatch scenarios. When evaluated on the ElecBench benchmark, GAIA surpasses the baseline model Large Language Model Meta AI-2 (LLaMA2) on multiple metrics. In practical applications, GAIA has demonstrated its ability to enhance decision-making processes, improve operational efficiency, and facilitate better human-machine interactions in power dispatch operations. This paper expands the application of LLMs to power dispatch and validates their practical utility, paving the way for future innovations in this field.https://doi.org/10.1038/s41598-025-91940-x
spellingShingle Yuheng Cheng
Huan Zhao
Xiyuan Zhou
Junhua Zhao
Yuji Cao
Chao Yang
Xinlei Cai
A large language model for advanced power dispatch
Scientific Reports
title A large language model for advanced power dispatch
title_full A large language model for advanced power dispatch
title_fullStr A large language model for advanced power dispatch
title_full_unstemmed A large language model for advanced power dispatch
title_short A large language model for advanced power dispatch
title_sort large language model for advanced power dispatch
url https://doi.org/10.1038/s41598-025-91940-x
work_keys_str_mv AT yuhengcheng alargelanguagemodelforadvancedpowerdispatch
AT huanzhao alargelanguagemodelforadvancedpowerdispatch
AT xiyuanzhou alargelanguagemodelforadvancedpowerdispatch
AT junhuazhao alargelanguagemodelforadvancedpowerdispatch
AT yujicao alargelanguagemodelforadvancedpowerdispatch
AT chaoyang alargelanguagemodelforadvancedpowerdispatch
AT xinleicai alargelanguagemodelforadvancedpowerdispatch
AT yuhengcheng largelanguagemodelforadvancedpowerdispatch
AT huanzhao largelanguagemodelforadvancedpowerdispatch
AT xiyuanzhou largelanguagemodelforadvancedpowerdispatch
AT junhuazhao largelanguagemodelforadvancedpowerdispatch
AT yujicao largelanguagemodelforadvancedpowerdispatch
AT chaoyang largelanguagemodelforadvancedpowerdispatch
AT xinleicai largelanguagemodelforadvancedpowerdispatch