AI-Based Customized Simulation Setup, Process Control, and Result Processing Technology for Distribution Networks

doi:10.3390/pr14142333

2026-07-17

DOI：10.3390/pr14142333

原文链接

原文求助

摘要

En 中文

为解决配电网数字仿真系统中“多样化仿真需求与标准化配置”和“海量仿真输出与稀疏决策信息”的两对基本矛盾，本文基于现有的基于公共信息模型（CIM）的自动仿真模型生成和动态电压调节仿真工作，进一步构建了一个用户导向的智能仿真服务层。该层由协同组成的Orchestration_Agent（仿真编排代理）和Analysis_Agent（结果分析代理）构成，承担三项职责：基于多级仿真粒度（L0–L3）和仿真模板库，利用大型语言模型（LLM）实现自然语言需求解析和自动工作流编排；基于仿真知识图谱，实现参数推荐、验证及过程控制的异常自适应恢复；基于规则模板、统计分析与因果图模型混合架构，实现自动结果分析、根本原因推理及结构化报告生成，并通过案例反馈驱动知识库迭代。在包含91个配电变压器区域的实际10 kV馈线30天连续数据（2880个时间截面）上进行验证：需求解析综合准确率达96.3%，自动参数配置覆盖率94.7%，异常识别召回率/精确率94.0%/96.9%，根本原因推理准确率92.1%，单次仿真端到端时间中位数从手动模式下的约36分钟缩短至4.7分钟。结果表明，所提服务层为配电网仿真从工具导向向服务导向的演进提供了一条可行的工程技术路径。

Keyword:

customized distribution network simulation

large language model agents (LLM agents)

natural language requirements parsing

automatic workflow orchestration

simulation process control

causal reasoning

simulation knowledge graph

root cause analysis

期刊

Processes

IF:

2.8

论文数:

7.2K

被引数:

3.7W

机构

暂无机构信息

引用论文

ChatGrid: Intelligent Knowledge Q&A for Power Dispatching Control Based on Large Language Models and Retrieval-augmented Generation

2024 IEEE 7th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC)

IF0

2024-09-20

PREAI

Ni,Mingjian; Zhang,Jing; Fu,Cong; Wang,Jiaqi; Ning,Xin; Li,Siyuan

Graph-Based Approaches and Functionalities in Retrieval-Augmented Generation: A Comprehensive Survey 朱志；黄涛；王凯；叶俊；陈欣；罗少. 基于图的方法和功能在检索增强生成中的应用：一项全面综述. ACM Comput. Surv. 2026, 58, 261. [Google Scholar] [CrossRef] [Scilit]

ACM Computing Surveys

IF0

2026-07-31

PREAI

Zhu,Zulun; Huang,Tiancheng; Wang,Kai; Ye,Junda; Chen,Xinghe; Luo,Siqiang

Gu, J.; Guo, M.; Li, R.; Xu, J.; Ye, Y.; Hu, Z.; Cui, S. Real-time cooperative voltage regulation strategy of distribution network based on approximate dynamic programming. Front. Energy Res. 2024, 12, 1512832. [Google Scholar] [CrossRef]基于近似动态规划的配电网实时协同电压调节策略

IF0

PREAI

A Data–Physics-Driven Modeling Approach of Key Equipment for Large-Scale Distribution Network Simulation 基于数据-物理驱动的面向大规模配电网仿真关键设备建模方法

Electronics

IF0

PREAI

Qiu,Rui; Bai,Hao; Yao,Ruotian; Liu,Chengxi; Xu,Min; Chen,Qi; Yang,Weichen

A Review of Knowledge Graph-Based Reasoning Technology in the Operation of Power Systems 电力系统运行中基于知识图谱的推理技术综述

Applied Sciences

IF0

2023-03-29

OAAI

Rui Liu; Rong Fu; Kang Xu; Xuanzhe Shi; Xiaoning Ren

A survey on LoRA of large language models 大型语言模型的LoRA研究综述

FRONTIERS OF COMPUTER SCIENCE

IF4.6

2024-12-14

OAAI

Mao, Yuren; Ge, Yuhang; Fan, Yijiang; Xu, Wenyi; Mi, Yu; Hu, Zhonghao; Gao, Yunjun

Parameter-efficient fine-tuning of large-scale pre-trained language models 大规模预训练语言模型的参数高效微调

NATURE MACHINE INTELLIGENCE

IF23.9

2023-03-02

139

OAAI

Ding, Ning; Qin, Yujia; Yang, Guang; Wei, Fuchao; Yang, Zonghan; Su, Yusheng; Hu, Shengding; Chen, Yulin; Chan, Chi-Min; Chen, Weize; Yi, Jing; Zhao, Weilin; Wang, Xiaozhi; Liu, Zhiyuan; Zheng, Hai-Tao; Chen, Jianfei; Liu, Yang; Tang, Jie; Li, Juanzi; Sun, Maosong

Optimised operational management of distribution grids by utilising flexibilities through automation of electrical assets using a multi-agent-system approach Riebesel, D.; Schramm, S.; Mehlmann, G.; Luther, M. 通过利用自动化电气资产的灵活性并采用多代理系统方法，优化配电网的运行管理。发表于第27届国际电力分配会议（CIRED 2023），意大利罗马，2023年6月12–15日。 [Google Scholar] [CrossRef] [Scilit]

IET Conference Proceedings

IF0

2023-07-04

PREAI

Riebesel,D.; Schramm,S.; Mehlmann,G.; Luther,M.

A Retrieval-Augmented Generation Approach for Data-Driven Energy Infrastructure Digital Twins

SMART CITIES

IF5.5

2024-10-24

OAAI

Ieva, Saverio; Loconte, Davide; Loseto, Giuseppe; Ruta, Michele; Scioscia, Floriano; Marche, Davide; Notarnicola, Marianna