Short-Term Solar Irradiance Forecasting Using Cloud Motion Vector and Cellular Automata-Based Prediction System

カテゴリ: 研究会(論文単位)

論文No: CMN25028

グループ名: 【C】電子・情報・システム部門 通信研究会

発行日: 2025/02/28

タイトル(英語): Short-Term Solar Irradiance Forecasting Using Cloud Motion Vector and Cellular Automata-Based Prediction System

著者名: Telang Geet Tapan(Hitachi India Private Limited),Kumar Abhishek(Hitachi India Private Limited),Kumar Vinoth(Hitachi India Private Limited)

著者名(英語): Geet Tapan Telang(Hitachi India Private Limited),Abhishek Kumar(Hitachi India Private Limited),Vinoth Kumar(Hitachi India Private Limited)

キーワード: Cloud Motion Vector|Solar Irradiance |Cellular automata|Forecasting|Geospatial analysis|Hybrid Modelling|Cloud Motion Vector|Solar Irradiance |Cellular automata|Forecasting|Geospatial analysis|Hybrid Modelling

要約(日本語): Global demand for renewable energy, especially solar power, has been revised upward by 30%, with capacity between 2022 and 2027 set to rise by nearly 2400 GW. To address solar radiation’s intermittent nature, we present a short-term solar irradiance forec

要約(英語): Global demand for renewable energy, especially solar power, has been revised upward by 30%, with capacity between 2022 and 2027 set to rise by nearly 2400 GW. To address solar radiation’s intermittent nature, we present a short-term solar irradiance forecasting approach in Hiriyur, Karnataka, India, using deep learning and cellular automata with geospatial and weather data. Our deep neural network achieved RMSEs of 0.808 (latitude) and 1.283 (longitude) using LSTM-based cloud-position forecasting. For Global Horizontal Irradiance, RMSE was 83 W/m2 with 86% accuracy. Cloud dynamics were also successfully modeled via cellular automata to predict cloud presence in the short future.

本誌: 2025年3月3日-2025年3月4日通信研究会

本誌掲載ページ: 35-40 p

原稿種別: 英語

PDFファイルサイズ: 2,057 Kバイト