Scholarlink

<p>This study proposes a GVSAO-CNN-BiGRU-Attention system for optimizing photovoltaic (PV) plant output prediction under complex high-altitude meteorological conditions. The system integrates convolutional neural networks (CNN), bi-directional gated recurrent units (BiGRU), attention mechanisms, and genetic algorithm optimization (GVSAO). The CNN extracts spatial features, while the BiGRU enhances temporal dependency modeling, effectively capturing short- and long-term patterns. The attention mechanism focuses on key features, improving prediction accuracy, and GVSAO optimizes model parameters for superior performance. The experimental results demonstrate the system’s efficacy, with an MSE of 0.018, an MAE of 0.072, and an R<sup>2</sup> of 0.99, indicating the model's high accuracy. This multimodal deep learning approach demonstrates strong adaptability and stability, offering a practical solution for high-altitude PV power plants and supporting future smart grid development.</p>

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

Design of an intelligent optimization system for high-altitude photovoltaic power plant output power prediction and enhancement using GVSAO-CNN-BiGRU-Attention

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年5月8日

查看全文
  • Shuai Yuan,
  • Hui Zhou,
  • Kai Zhao

摘要

This study proposes a GVSAO-CNN-BiGRU-Attention system for optimizing photovoltaic (PV) plant output prediction under complex high-altitude meteorological conditions. The system integrates convolutional neural networks (CNN), bi-directional gated recurrent units (BiGRU), attention mechanisms, and genetic algorithm optimization (GVSAO). The CNN extracts spatial features, while the BiGRU enhances temporal dependency modeling, effectively capturing short- and long-term patterns. The attention mechanism focuses on key features, improving prediction accuracy, and GVSAO optimizes model parameters for superior performance. The experimental results demonstrate the system’s efficacy, with an MSE of 0.018, an MAE of 0.072, and an R2 of 0.99, indicating the model's high accuracy. This multimodal deep learning approach demonstrates strong adaptability and stability, offering a practical solution for high-altitude PV power plants and supporting future smart grid development.