电力系统自动化2017，Vol.41Issue(22)：96-101,6.DOI:10.7500/AEPS20170517017
考虑时空相关性的光辐照度序列估计方法
Estimation Method of Solar Irradiance Sequence Considering Temporal-Spatial Correlation
摘要
Abstract
Complete and reliable solar irradiance data is the prerequisite of photovoltaic generation research.A new method is proposed to estimate or reconstruct the solar irradiance sequence with constraints of sparse meters and high ratio of outliers.The temporal and spatial relationship of ground irradiance metering from various metering stations in Hexi Corridor in Gansu Province is analyzed by the correlation coefficient method on sunny and cloudy days.In consideration of the prevailing wind,the area correlation model is developed by Voronoi polygon method.By using time phasor delay,the data reconstruction model is built based on the asynchronous sequence.The absent irradiance sequence could be recovered by choosing the metering data with the highest correlation and equivalent time delay in the prevailing wind as inputs.The data test has shown that both the accuracy of the area correlation model and reconstruction model have reach a relatively high level.关键词
光伏发电/光辐照度/相关性分析/数据重构/神经网络Key words
photovoltaic power generation/solar irradiance/correlation analysis/data reconstruction/neural network引用本文复制引用
李兢,乔颖,龚莺飞..考虑时空相关性的光辐照度序列估计方法[J].电力系统自动化,2017,41(22):96-101,6.基金项目
This work is supported by National Natural Science Foundation of China (No.51677099).国家自然科学基金面上项目(51677099). （No.51677099）
