サポートベクターマシンを用いた東京の日射量予測

カテゴリ: 全国大会

論文No: 7-061

グループ名: 【全国大会】平成23年電気学会全国大会論文集

発行日: 2011/03/05

タイトル(英語): Application of Support Vector Machine to Forecast Insolation in Tokyo, Japan

著者名: Joao Gari da Silva Fonseca Junior (産業技術総合研究所),大関 崇(産業技術総合研究所),荻本 和彦(東京大学)

著者名(英語): Joao Gari da Silva Fonseca Junior (National Institute of Advanced Industrial Science and Technology),Takashi Oozeki(National Institute of Advanced Industrial Science and Technology),Kazuhiko Ogimoto(Tokyo University)

キーワード: 日射量予測|気象数値予報|太陽電池

要約(日本語): The objective of this study is to forecast insolation at Tokyo, Japan with support vector machines and numerically predicted cloudiness. Insolation was forecast one day ahead, for a period of one year. The results showed that the proposed method was able to forecast insolation with a good level of accuracy. For example, the determination coefficient between forecast and measured values of insolation was 0.815. Furthermore, the insolation forecasts had an hourly root mean square error of 0.108 kWh/m2 and a mean absolute error of 0.065 kWh/m2.

原稿種別: 日本語

PDFファイルサイズ: 2,061 Kバイト