シミュレーション最適化技術を用いた風車配置最適化

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2023/02/01

タイトル(英語): Development of a Simulation-Based Optimization Method for Wind Farm Layout on Complex Terrain

著者名: 桐淵　大貴（（株）東芝 研究開発センター），吉田　琢史（（株）東芝 研究開発センター），岡山　芙有子（東芝エネルギーシステムズ（株） エネルギーアグリゲーション事業部），金野　夏奈（東芝エネルギーシステムズ（株） エネルギーアグリゲーション事業部），松田　匠（東芝エネルギーシステムズ（株） エネルギーアグリゲーション事業部）

著者名(英語): Daiki Kiribuchi (Corporate Research & Development Center, Toshiba Corporation), Takufumi Yoshida (Corporate Research & Development Center, Toshiba Corporation), Fuko Okayama (Energy Aggregation Division, Toshiba Energy Systems & Solutions Corporation), Kana Konno (Energy Aggregation Division, Toshiba Energy Systems & Solutions Corporation), Takumi Matsuda (Energy Aggregation Division, Toshiba Energy Systems & Solutions Corporation)

キーワード: 風車，風力発電，マイクロサイティング，風車配置最適化，シミュレーション最適化，整数線形計画　　wind turbine，wind power，micro-siting，wind farm layout optimization，simulation optimization，integer linear programming

要約(英語): To improve the efficiency of wind power plants, multiple wind turbines must be placed in a way that generates a large amount of power while satisfying various constraints. For complex terrain such as mountainous regions in Japan, long simulation runs are required in order to calculate how much power is generated and determine whether the constraints are satisfied. Consequently, the manual processes and conventional simulation methods for deciding wind farm layouts are time-consuming. To automate and accelerate the process of wind farm placement, we propose a simulation-based optimization method for wind farm layout. In this method, simulation data are used to construct a prediction model, based on which the wind farm layout is then obtained by the optimization method. We conducted numerical experiments to evaluate our method and confirmed that it could find a wind farm layout that satisfied all given constraints and increased power generation by more than 30%, whereas the manual method could not find a layout that satisfied all the constraints in the same amount of time.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.143 No.2 （2023） 特集：エネルギー分野へ適用されたAI・IoT技術

本誌掲載ページ: 151-158 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/ieejeiss/143/2/143_151/_article/-char/ja/