基于地基云图图像特征的光伏功率预测Photovoltaic Power Prediction Based on Image Features of Ground Cloud Image
路志英,周庆霞,李鑫,王泽涵
摘要(Abstract):
光伏输出功率主要受太阳辐照度的影响,而天空中云的生成、运动以及消融会使太阳辐照度呈现随机性和波动性。地基云图可实时记录天空状况,因此获取地基云图的图像特征是光伏功率准确预测的关键步骤。对全天空成像仪采集的地基云图展开研究。首先,修复地基云图;然后,利用图像处理技术提取影响太阳辐照度变化的图像特征,包含光照强度、高频分量、透射率、天顶距离以及云因子特征;最后,将图像特征作为输入数据,光伏功率作为输出数据,利用梯度提升决策树算法构建光伏功率预测模型,实现光伏功率的预测。实验结果表明,采用从地基云图提取的图像特征构建的光伏功率预测模型,使得光伏功率预测的均方根误差可低于1%,为光伏功率的准确预测提供了一种技术手段。
关键词(KeyWords): 太阳辐照度;光伏功率;地基云图;特征提取;预测模型
基金项目(Foundation): 国家自然科学基金资助项目(51677123)
作者(Author): 路志英,周庆霞,李鑫,王泽涵
DOI: 10.19635/j.cnki.csu-epsa.000457
参考文献(References):
- [1]徐正阳,路志英,刘洪(Xu Zhengyang,Lu Zhiying,Liu Hong).基于泄漏积分型回声状态网络的在线学习光伏功率预测(Online-learning PV power forecasting based on leaky-integrator echo state network)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2018,30(2):1-7.
- [2] Strzalka A,Alam N,Duminil E,et al. Large scale integration of photovoltaics in cities[J]. Applied Energy,2012,93:413-421.
- [3]张景景(Zhang Jingjing).基于BP神经网络的太阳辐射度预测(Forecast solar radiation based on BP neural network)[J].工业控制计算机(Industrial Control Computer),2014,27(6):39-40,43.
- [4]贺文,齐爽,陈厚合(He Wen,Qi Shuang,Chen Houhe).蚁群BP神经网络的光伏电站辐照强度预测(Prediction of radiation intensity for photovoltaic power plants based on ant colony BP neural network)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2016,28(7):26-31.
- [5]刘剑,曹美燕,高治军,等(Liu Jian,Cao Meiyan,Gao Zhijun,et al).一种基于随机森林的太阳能辐射预测模型(A solar radiation prediction model based on radom forest)[J].控制工程(Control Engineering of China),2017,24(12):2472-2477.
- [6]陶仁峰,李凤婷,李永东,等(Tao Renfeng,Li Fengting,Li Yongdong,et al).基于云层分布规律与太阳光跟踪的光伏电站MPPT策略(MTTP strategy of photovoltaic station based on cloud distribution pattern and sunlight tracking)[J].电力系统自动化(Automation of Electric Power Systems),2018,29(5):113-116.
- [7]朱想,周海,丁杰,等(Zhu Xiang,Zhou Hai,Ding Jie,et al).全天空云图图像复原算法(All-sky cloud map image restoration algorithm research)[J].计算机辅助设计与图形学学报(Journal of Computer-Aided Design&Computer Graphics),2014,26(6):896-902.
- [8]刘海波,王建芳,于海芹,等(Liu Haibo,Wang Jianfang,Yu Haiqin,et al).太阳能工程中几种相关角度的计算及应用(Several correlative angle calculation of solar energy project)[J].阳光能源(Solar&Renewable Energy Sources),2010,12:67-70.
- [9]路志英,周庆霞,李鑫(Lu Zhiying,Zhou Qingxia,Li Xin).基于改进Criminisi算法的地基云图修复方法(Ground-based cloud image inpainting method based on improved Criminisi algorithm)[J].数据采集与处理(Journal of Data Acquisition and Processing),2019,34(1):16-25.
- [10] He Kaiming,Sun Jian,Tang Xiaoou. Single image haze removal using dark channel prior[J]. IEEE Trans on Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353.