3D-CNN-based feature extraction of ground-based cloud images for direct normal irradiance prediction

被引:117
作者
Zhao, Xin [1 ]
Wei, Haikun [1 ]
Wang, Hai [2 ]
Zhu, Tingting [1 ]
Zhang, Kanjian [1 ]
机构
[1] Southeast Univ, Sch Automat, Minist Educ, Key Lab Measurement & Control,CSE, Nanjing 210096, Jiangsu, Peoples R China
[2] St Marys Univ, Sobey Sch Business, Halifax, NS, Canada
基金
中国国家自然科学基金;
关键词
Direct normal irradiance; 3D convolutional neural network; Feature extraction; Ground-based cloud image; SOLAR IRRADIANCE; NEURAL-NETWORKS; FORECASTS; MODELS;
D O I
10.1016/j.solener.2019.01.096
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Cloud cover and cloud motion have a large impact on solar irradiance. One of the effective ways for direct normal irradiance (DNI) prediction is to use cloud features, which has been extensively studied. A Convolutional Neural Network (CNN) has the advantage of automatic features extraction by using strong computing capabilities. In this paper, a novel 3D-CNN method is proposed by processing multiple consecutive ground-based cloud (GBC) images in order to extract cloud features including texture and temporal information. The resulting features and the DNI data are then used to establish a DNI forecasting model. The experiments are carried out to evaluate the performance of the proposed forecasting method by using the data from January 1, 2013 to December 31, 2014. The experimental results show that the proposed method coupled with the multilayer perceptron (MLP) model achieves forecast skill of 17.06% for 10-minute ahead DNI prediction.
引用
收藏
页码:510 / 518
页数:9
相关论文
共 35 条
[1]  
[Anonymous], 2017, NEUROCOMP, DOI DOI 10.1109/TII.2017.2739340
[2]   A hybrid method based on a new clustering technique and multilayer perceptron neural networks for hourly solar radiation forecasting [J].
Azimi, R. ;
Ghayekhloo, M. ;
Ghofrani, M. .
ENERGY CONVERSION AND MANAGEMENT, 2016, 118 :331-344
[3]   Online short-term solar power forecasting [J].
Bacher, Peder ;
Madsen, Henrik ;
Nielsen, Henrik Aalborg .
SOLAR ENERGY, 2009, 83 (10) :1772-1783
[4]   Comparison of solar radiation models and their validation under Algerian climate - The case of direct irradiance [J].
Behar, Omar ;
Khellaf, Abdallah ;
Mohammedi, Kama .
ENERGY CONVERSION AND MANAGEMENT, 2015, 98 :236-251
[5]   Intra-hour direct normal irradiance forecasting through adaptive clear-sky modelling and cloud tracking [J].
Bone, Viv ;
Pidgeon, John ;
Kearney, Michael ;
Veeraragavan, Ananthanarayanan .
SOLAR ENERGY, 2018, 159 :852-867
[6]   Multimodel inference - understanding AIC and BIC in model selection [J].
Burnham, KP ;
Anderson, DR .
SOCIOLOGICAL METHODS & RESEARCH, 2004, 33 (02) :261-304
[7]   Multi-model solar irradiance prediction based on automatic cloud classification [J].
Cheng, Hsu-Yung ;
Yu, Chih-Chang .
ENERGY, 2015, 91 :579-587
[8]   Bi-model short-term solar irradiance prediction using support vector regressors [J].
Cheng, Hsu-Yung ;
Yu, Chih-Chang ;
Lin, Sian-Jing .
ENERGY, 2014, 70 :121-127
[9]   Short-term probabilistic forecasts for Direct Normal Irradiance [J].
Chu, Yinghao ;
Coimbra, Carlos F. M. .
RENEWABLE ENERGY, 2017, 101 :526-536
[10]   Hybrid intra-hour DNI forecasts with sky image processing enhanced by stochastic learning [J].
Chu, Yinghao ;
Pedro, Hugo T. C. ;
Coimbra, Carlos F. M. .
SOLAR ENERGY, 2013, 98 :592-603