Spatiotemporal feature encoded deep learning method for rooftop PV potential assessment

2025年9月1日徐剑, 郭直灵, Qing Yu, 董克川, 谭鸿骏, Haoran Zhang, Jinyue Yan

Spatiotemporal feature encoded deep learning method for rooftop PV potential assessment

概要: 编码多源时空特征的深度学习框架，可预测阴影投射并计算屋顶光伏潜力（年发电潜力 MAPE 2.85%，遮挡评估 mIoU 89.23%），速度约为仿真方法的 158 倍。
类型: 期刊论文
发表于: Applied Energy, 394, 126171
DOI: 10.1016/j.apenergy.2025.126171

Spatiotemporal feature encoded deep learning method for rooftop PV potential assessment — 图 1：遮挡建筑周边的建筑高度与年屋顶辐照量。

引用本文

@article{xu2025spatiotemporal,
  title   = {Spatiotemporal feature encoded deep learning method for rooftop PV potential assessment},
  author  = {徐剑 and 郭直灵 and Qing Yu and 董克川 and 谭鸿骏 and Haoran Zhang and Jinyue Yan},
  journal = {Applied Energy},
  year    = {2025},
  doi     = {10.1016/j.apenergy.2025.126171}
}