Name
Title | ||
|---|---|---|
A Guided Deep Reinforcement Learning Method For Distribution Voltage Regulation via Battery Systems |
Abstract | ||
|---|---|---|
The rapid growth of photovoltaic penetration leads to serious voltage issues in distribution grids. While the battery energy storage system has been used for voltage regulation, its effectiveness is limited by the assumption of adequate battery sizes. In this paper, we first propose a battery scheduling model to relax the battery size assumption. Afterward, a deep reinforcement learning-based method is proposed to solve this high-dimensional scheduling model. The method includes a guided training framework, which combines reward shaping and curriculum learning techniques to guide the training. The proposed method has been implemented to the IEEE 13-bus test feeder. It shows that the guided training framework accelerates the training and improves the convergence. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1109/ISGT49243.2021.9372224 | 2021 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT) |
Keywords | DocType | ISSN |
Renewable energy integration,battery size,voltage regulation,deep reinforcement learning | Conference | 2167-9665 |
ISBN | Citations | PageRank |
978-1-7281-8898-0 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xiaoge Huang | 1 | 0 | 0.34 |
| Zhenhuan Ding | 2 | 0 | 1.01 |
| Ziang Zhang | 3 | 0 | 0.34 |