Optimization of Home Energy Management System with Incentives Using Deep Reinforcement Learning
Optimization of Home Energy Management System with Incentives Using Deep Reinforcement Learning
カテゴリ: 部門大会
論文No: GS6-1
グループ名: 【C】2021年電気学会電子・情報・システム部門大会
発行日: 2021/09/08
タイトル(英語): Optimization of Home Energy Management System with Incentives Using Deep Reinforcement Learning
著者名: 孫 俊傑(筑波大学),安芸 裕久(筑波大学)
著者名(英語): Junjie Sun (The University of Tsukuba),Hirohisa Aki (The University of Tsukuba)
キーワード: 電力システム|深層強化学習|デマンドレスポンスデマンドレスポンス|Energy system|Deep reinforcement learning|Demand response
要約(日本語): With the development of smart grid and smart home, the structure of the power consumption side and the power generation side have undergone tremendous changes. Home energy management system (HEMS) is expected to cope with the increasing complexities and uncertainties in the prosumer side of the smart grid system. In this paper, we proposed to use deep q learning (DQN), a type of deep reinforcement learning (DRL) that combines the deep learning (DL) and reinforcement learning (RL) to perform online control for smart appliances and real-time dynamic electrical price scheme for HEMS and an aggregator. By using real-world database, design a model and evaluate the effectiveness of it.
PDFファイルサイズ: 1,206 Kバイト
受取状況を読み込めませんでした
