中国水利Issue(6)：14-30,17.DOI:10.3969/j.issn.1000-1123.2025.06.002
数字孪生水利建设中的人工智能大模型应用探索
Exploration of artificial intelligence large model applications in digital twin water conservancy construction
摘要
Abstract
Artificial intelligence(AI)large models can provide new momentum for enhancing the quality and efficiency of digital twin water conservancy construction.Following the approach of"positioning analysis,route exploration,needs assessment,implementation,and model promotion",this study analyzes the challenges in digital twin water conservancy,the development and industry applications of AI large models,and the necessity and feasibility of their application in this field.From the perspectives of technology,business,and management,the study explores the application of AI large models in digital twin water conservancy scenarios.Using the technical path of"scenario digitization,intelligent simulation,and precise decision-making",the study outlines key technologies such as dynamic digital scenario construction,intelligent simulation of complex systems,and precise human-machine collaborative decision-making.Based on a"2+N"business demand framework,the application routes and steps of AI large models are illustrated through specific scenarios,such as the"four pre"(forecasting,warning,rehearsal,and planning)flood control applications and network security protection.The study proposes a co-construction and sharing model for the application of AI large models in digital twin water conservancy,offering a"co-construction and sharing,unified and distributed,collaborative promotion"development approach.The findings provide references for the functional positioning and business scenario implementation of AI large models in digital twin water conservancy construction.关键词
数字孪生水利/人工智能/大模型/智能模拟/人机协同Key words
digital twin water conservancy/artificial intelligence/large models/intelligent simulation/human-machine collaboration分类
水利科学引用本文复制引用
舒全英,马媛,陈亮,李磊,郭磊,吴健柏..数字孪生水利建设中的人工智能大模型应用探索[J].中国水利,2025,(6):14-30,17.基金项目
国家重点研发计划"流域多源异构信息智能融合与数据底板构建"(2023YFC3209201) （2023YFC3209201）
国家自然科学基金长江水科学研究联合基金计划"基于数字孪生技术的长江下游感潮河网地区多目标调度研究项目"(U2340221). （U2340221）
