自动化学报2025，Vol.51Issue(11)：2387-2411,25.DOI:10.16383/j.aas.c250203
面向源网荷的智能化数据协同推断技术研究综述
A Review of Intelligent Data Collaborative Inference Techniques for Source-grid-load Systems
摘要
Abstract
With the continuous increase in the proportion of renewable energy integration,new energy generation forms such as wind power and photovoltaic systems have imposed higher requirements on the stability and intelli-gent dispatch of power systems.Under the integrated framework of source-grid-load-storage,how to efficiently util-ize multi-source heterogeneous power data for accurate forecasting and collaborative analysis has become a critical issue.In recent years,technologies such as deep learning,big data,and large-scale models have driven rapid ad-vancements in intelligent inference techniques.This paper first elaborates on the current research status of common technologies for collaborative inference of time series data,combined with deep learning techniques.Key ap-proaches including trend-seasonality decomposition,frequency-domain modeling,and exogenous variable fusion are emphasized,alongside an analysis of time series models based on different architectures.Secondly,the paper dis-cusses the key intelligent technologies for source-grid-load integration,further outlining critical technological path-ways in typical scenarios such as intelligent forecasting,state assessment,and load scheduling within the source-grid-load-storage system,with detailed analyses of their specific application contexts.Finally,in light of the increasingly complex power system environment,prospects for the future development of data collaborative inference technolo-gies are presented.关键词
源网荷智能化/数据协同推断/时间序列分析/深度学习Key words
Intelligentization of source-grid-load/data collaborative inference/time series analysis/deep learning引用本文复制引用
张辉,颜星雨,毛建旭,别克扎提·巴合提,杜瑞,王耀南..面向源网荷的智能化数据协同推断技术研究综述[J].自动化学报,2025,51(11):2387-2411,25.基金项目
国家自然科学基金重点项目(62433010),科技创新2030-"新一代人工智能"重大项目(2021ZD0114503),国家电网有限公司科技项目(5700-202423229A-1-1-ZN)资助Supported by the Key Program of National Natural Science Foundation of China(62433010),the National Key Research and Development Program of China(2021ZD0114503),and the Sci-ence and Technology Project of State Grid Corporation of China(5700-202423229A-1-1-ZN) （62433010）
