计及雾霾影响因素的光伏发电超短期功率预测

Super Short-term Photovoltaic Power Forecasting Considering Influence Factor of Smog

近年来频繁出现的雾霾天气,对光伏发电的输出功率造成了显著影响。首先,针对太阳辐射削弱情景,采用机器学习方法对大气气溶胶光学厚度进行估计,进而利用经云遮系数修正的晴空太阳辐射模型计算光伏电池面板接收的辐射强度;其次,针对光伏电池面板积灰情景,开展功率测量实验,构造颗粒物累积浓度——功率衰减率样本集,从而通过相似日筛选估计不同积灰状态下的功率输出衰减率;最终,结合光电转换模型,基于地面颗粒物浓度预测以及精细天气预报信息,提出一种光伏发电超短期功率预测方法。实验结果表明,所提出的预测方法具有较高的精度,且易于工程实现。

Smog weather has been occurring frequently in recent years, which affects the output power of photovoltaic generation obviously. First of all, aim at the weakening effect on solar irradiance, estimation models of atmospheric aerosol optical thickness（AOD） were established with machine learning method, and irradiance received by photovoltaic panels was calculated using clear sky model corrected by cloud cover coefficient. Secondly, aim at the aggravating effect on dust deposition on PV panels, the sample set of ＂accumulated particle matter（PM） concentration — power attenuation rate＂ was constructed through power measurement experiment, then power attenuation rate under different dust deposition state was estimated with similar-day choosing method. Finally, based on photoelectric conversion model, PM concentration prediction and fine weather forecast information, super short-term photovoltaic power forecasting method was presented. Experimental results show that the presented forecasting method has higher accuracy, and is easy to implement in engineering project.