计算机科学
可靠性(半导体)
可靠性工程
网格
分布(数学)
背景(考古学)
电网
区间(图论)
工程类
数学优化
作者
Mengmeng Zhuang,Andrija Sadikovic
出处
期刊:IEEE Power Engineering Society General Meeting
[Institute of Electrical and Electronics Engineers]
日期:2026-07-01
卷期号:: 1-5
标识
DOI:10.1109/pesgm58988.2026.11693356
摘要
This paper presents a human-in-the-loop LLM–assisted multi-agent framework for distribution grid reliability planning. Traditional reliability planning relies on deterministic simulations and manual engineering workflows, which can be difficult to scale for large feeder portfolios and evolving grid conditions. The proposed framework integrates large language model (LLM) reasoning with deterministic reliability analysis and optimization tools within a coordinated multi-agent architecture. Specialized agents perform data ingestion, reliability evaluation, planning optimization, and compliance reporting, while numerical simulation tools validate candidate planning actions generated by the LLM. Case studies on 100 real-world utility feeders demonstrate that the framework reduces planning time and manual engineering effort while maintaining acceptable planning accuracy for large-scale screening and scenario analysis.
科研通智能强力驱动
Strongly Powered by AbleSci AI