arXiv cs.AI· Jiayang Tu, Bozhen Jiang, Bojun Du, Yunxiao Ling, Ershun Du, Qin Wang·· 1 天前AI 评分22
LLM 增强深度强化学习结合人在回路校准，用于实时最优潮流
LLM-Enhanced Deep Reinforcement Learning with Human-in-the-Loop Calibration for Real-Time Optimal Power Flow
AI 导读
研究提出一种 LLM 增强的深度强化学习框架，通过人在回路（HITL）校准解决实时最优潮流（RT-OPF）中难以用数值奖励编码的运行偏好与安全要求。该方法在离线训练时用稀疏语言评价反馈增强基于物理的奖励，并引入选择性 HITL 校准机制检查纠正可疑的 LLM 判断。
正文 · 原文
Abstract:Real-time optimal power flow (RT-OPF) requires fast and reliable decision-making under time-varying operating conditions. However, many practical operating preferences and safety related requirements are difficult to encode using conventional numerical rewards alone, which limits the effectiveness and controllability of deep reinforcement learning (DRL) for RT-OPF. To address this issue, this paper proposes an LLM-enhanced DRL framework with human-in-the-loop (HITL) for RT-OPF. The proposed method augments the physics-based reward with sparse language evaluative feedback during offline training, so that semantic operational requirements can be incorporated into policy learning. To improve reliability in RT-OPF, a selective HITL calibration mechanism is further introduced to inspect and correct questionable LLM judgments before they continuously shape the learning process. Case studies on the IEEE 30-bus and IEEE 118-bus systems show that LLM feedback improves policy safety over vanilla TD3, while further HITL calibration achieves the best overall performance and more stable training behavior. These results demonstrate the effectiveness of combining language guidance and expert calibration for RT-OPF.
| Subjects: | Systems and Control (eess.SY); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.15003 [eess.SY] |
| (or arXiv:2611.15003v1 [eess.SY] for this version) | |
| https://doi.org/10.48550/arXiv.2611.15003 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jiayang Tu [view email]
[v1]
Wed, 11 Nov 2026 04:01:40 UTC (1,100 KB)
来源：arXiv cs.AI · arxiv.org