基于XGB算法及全天空成像仪图像的超短期DNI预测ULTRA-SHORT-TERM DNI PREDICTION BASED ON XGB ALGORITHM AND ALL-SKY IMAGER IMAGES
胡中,倪杭飞,李伟,宓霄凌,白帆,王伊娜
摘要(Abstract)：
吸热器是塔式太阳能热发电站的核心设备，但云层短时间的遮挡和离开会使吸热器接收的太阳直接辐射（DNI）发生随机波动，造成吸热器表面受热发生剧烈变化，会影响塔式太阳能热发电站的安全运行。本文提出了一种基于极值梯度提升（eXtreme Gradient Boosting,XGB）算法及全天空成像仪图像的超短期DNI预测方法。通过使用基于太阳位置的全天空成像仪标定方法，完成全天空成像仪的姿态校正和畸变矫正；对晴空背景参数进行拟合并建立晴空库，为云层的检测提供基础；通过对太阳及其临近区域进行插值修复，使1 min内的DNI预测成为可能；根据云的运动矢量，利用外推法得到未来云层可能会遮挡太阳的图像点，并提取图像特征，用于基于XGB算法的超短期DNI预测模型的训练，最后利用训练后的模型完成超短期DNI预测。
关键词(KeyWords)： 塔式太阳能热发电站;全天空成像仪;晴空库;超短期DNI预测;XGB算法;云层检测
基金项目(Foundation):
作者(Author): 胡中,倪杭飞,李伟,宓霄凌,白帆,王伊娜
DOI: 10.19911/j.1003-0417.tyn20201224.02
参考文献(References)：
- [1] DARULA S, KITTLE R. CIE general sky standard defining luminance distributions[C]//eSim 2002 Proceedings:The Canadian Conference on Building Simulation, September,2002, Montreal, Canada.
- [2] CHAUVIN R, NOU J, THIL S, et al. Cloud detection methodology based on a sky-imaging system[J]. Energy procedia, 2015, 69:1970-1980.
- [3] BERNECKER D, RIESS C, ANGELOPOULOU E, et al. Continuous short-term irradiance forecasts using sky images[J]. Solar energy, 2014, 110:303-315.
- [4] INMAN R H, PEDRO H T C, COIMBRA C F M. Solar forecasting methods for renewable energy integration[J].Progress in energy and combustion science, 2013, 39(6):535-576.
- [5] QUESADA-RUIZ S, CHU Y, TOVAR-PESCADOR J,et al. Cloud-tracking methodology for intra-hour DNI forecasting[J]. Solar energy, 2014, 102:267-275.
- [6] TZOUMANIKAS P, NIKITIDOU E, BAIS A F, et al. The effect of clouds on surface solar irradiance, based on data from anall-sky imaging system[J]. Renewable energy, 2016,95(C):314-322.
- [7] PENG Z, YU D, HUANG D, et al. 3D cloud detection and tracking system for solarforecast using multiple sky imagers[J]. Solar energy, 2015, 118:496-519.