Scholarlink

The crucial role of photovoltaic (PV) systems in the transition towards sustainable energy goals has intensified due to the growing global interest in renewable energies. Accurate forecasting of PV generation is essential to optimise energy management and ensure grid stability, which motivates the exploration of advanced predictive methodologies. This study leverages PVGIS, a comprehensive and freely available solar energy and radiation database, to develop and evaluate an intelligent time series forecasting model for PV power generation. By employing machine learning techniques specifically tailored to the time series data, this approach achieves high accuracy in predicting short-term PV production under varying weather conditions. The results demonstrate the model’s potential as a reliable tool for proper energy management in PV installations, allowing operators to optimise resource allocation and improve system efficiency with greater predictability. This research highlights the value of integrating open databases with intelligent forecasting methods to advance the operational efficiency of solar energy systems.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Power Prediction System for Photovoltaic Panels Using Time Series Neural Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Noel Freire-Mahía,
  • Álvaro Michelena,
  • Antonio Díaz-Longueira,
  • Héctor Quintián,
  • Esteban Jove

摘要

The crucial role of photovoltaic (PV) systems in the transition towards sustainable energy goals has intensified due to the growing global interest in renewable energies. Accurate forecasting of PV generation is essential to optimise energy management and ensure grid stability, which motivates the exploration of advanced predictive methodologies. This study leverages PVGIS, a comprehensive and freely available solar energy and radiation database, to develop and evaluate an intelligent time series forecasting model for PV power generation. By employing machine learning techniques specifically tailored to the time series data, this approach achieves high accuracy in predicting short-term PV production under varying weather conditions. The results demonstrate the model’s potential as a reliable tool for proper energy management in PV installations, allowing operators to optimise resource allocation and improve system efficiency with greater predictability. This research highlights the value of integrating open databases with intelligent forecasting methods to advance the operational efficiency of solar energy systems.