類似仮想ビル選択による即時運用可能な空調電力管理システムの強化学習制御

カテゴリ: 論文誌(論文単位)

グループ名: 【D】産業応用部門

発行日: 2023/01/01

タイトル(英語): Reinforcement Learning Control for HVAC Energy Management System with Instant Operation by Selecting Virtual Building with Similar Environment

著者名: 青木　佳史（岐阜大学），髙橋　祐輔（岐阜大学），蜷川　忠三（岐阜大学），森川　純次（三菱重工サーマルシステムズ（株））

著者名(英語): Yoshifumi Aoki (Gifu University), Yusuke Takahashi (Gifu University), Chuzo Ninagawa (Gifu University), Junji Morikawa (Mitsubishi Heavy Industries Thermal Systems, Ltd.)

キーワード: ビルマルチ空調機，強化学習，転移学習_x000D_　　building multi-type air-conditioner，reinforcement learning，transfer learning

要約(英語): This paper proposes a novel method for building multi-type air-conditioners, wherein the power management control system can automatically adapt to various air-conditioning environments using reinforcement learning. Our previous study reduced the learning period by pre-training on a virtual building and simulated the dynamic power characteristics of air-conditioners and room temperature. However, advantages of decreasing the learning period diminish when the difference between the virtual and actual buildings is significant. Therefore, our proposed method first performs pre-training on multiple virtual buildings with different environments. Subsequently, it selects the one whose environment is closest to that of the actual building based on the difference in average rewards.

本誌: 電気学会論文誌D（産業応用部門誌） Vol.143 No.1 （2023）

本誌掲載ページ: 27-34 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejias/143/1/143_27/_article/-char/ja/