期刊
SOLAR ENERGY
卷 112, 期 -, 页码 446-457出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.solener.2014.12.014
关键词
Intraday solar forecasting; Machine learning techniques; Statistical models
In this paper, we propose a benchmarking of supervised machine learning techniques (neural networks, Gaussian processes and support vector machines) in order to forecast the Global Horizontal solar Irradiance (GHI). We also include in this benchmark a simple linear autoregressive (AR) model as well as two naive models based on persistence of the GHI and persistence of the clear sky index (denoted herein scaled persistence model). The models are calibrated and validated with data from three French islands: Corsica (41.91 degrees N; 8.73 degrees E), Guadeloupe (16.26 degrees N; 61.51 degrees W) and Reunion (21.34 degrees S; 55.49 degrees E). The main findings of this work are, that for hour ahead solar forecasting, the machine learning techniques slightly improve the performances exhibited by the linear AR and the scaled persistence model. However, the improvement appears to be more pronounced in case of unstable sky conditions. These nonlinear techniques start to outperform their simple counterparts for forecasting horizons greater than 1 h. (C) 2014 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据