Empirical and machine learning models for predicting daily global solar radiation from sunshine duration: A review and case study in China

RENEWABLE & SUSTAINABLE ENERGY REVIEWS

卷:100

Review

Fan, Junliang¹  Wu, Lifeng^1,2  Zhang, Fucang^1,3  Cai, Huanjie^1,3  Zeng, Wenzhi⁴  Wang, Xiukang⁵  Zou, Haiyang¹ 

[1] Northwest A&F Univ, Minist Educ, Key Lab Agr Soil & Water Engn Arid & Semiarid Are, Yangling 712100, Shaanxi, Peoples R China

[2] Nanchang Inst Technol, Sch Hydraul & Ecol Engn, Nanchang 330099, Jiangxi, Peoples R China

[3] Northwest A&F Univ, Inst Water Saving Agr Arid Areas China, Yangling 712100, Shaanxi, Peoples R China

[4] Wuhan Univ, State Key Lab Water Resources & Hydropower Engn S, Wuhan 430072, Hubei, Peoples R China

[5] Yanan Univ, Coll Life Sci, Yanan 716000, Peoples R China

关键词: Solar radiation;    Machine learning;    Model comparison;    Global performance index (GPI);    Computational time;    Memory usage;   

DOI  :  10.1016/j.rser.2018.10.018

来源: Elsevier

【 摘 要 】

Accurate estimation of global solar radiation (R-s) is essential to the design and assessment of solar energy utilization systems. Existing empirical and machine learning models for estimating R-s from sunshine duration were comprehensively reviewed. The performances of 12 empirical model forms and 12 machine learning algorithms for estimating daily Rs were further evaluated in different climatic zones of China as a case study, i.e. the temperate continental zone (TCZ), temperate monsoon zone (TMZ), mountain plateau zone (MPZ) and (sub) tropical monsoon zone (SMZ). The best-performing model at each station and the overall best model for each climatic zone were selected based on six statistical indictors, a global performance index (GPI) and computational costs (computational time and memory usage). The results revealed that the machine learning models (RMSE: 2.055-2.751 MJ m(-2) d(-1); NRMSE: 12.8-21.3%; R-2: 0.839-0.936) generally outperformed the empirical models (RMSE: 2.118-3.540 MJ m(-2) d(-1); NRMSE: 12.1-27.5%; R-2: 0.834-0.935) in terms of prediction accuracy. The cubic model (M3), modified linear-logarithmic model (M5) and power model (M10) attained generally better ranks among empirical models based on GPI. M3 was the top-ranked model in TMZ and MPZ, while general best performance was obtained by M5 and M2 in SMZ and TCZ, respectively. ANFIS, ELM, LSSVM and MARS obtained generally better performance among machine learning models, with the overall best ranking by ANFIS in TCZ and SMZ and by ELM in MPZ and SMZ. XGBoost (8.1 s and 74.2 MB), M5Tree (11.3 s and 29.7 MB), GRNN (12.3 s and 295.3 MB), MARS (14.4 s and 42.6 MB), MLP (22.4 s and 41.3 MB) and ANFIS (29.8 s and 23.1 MB) showed relatively small computational time and memory usage. Comprehensively considering both the prediction accuracy and computational costs, ANFIS is highly recommended, while MARS and XGBoost are also promising models for daily R-s estimation.

【 授权许可】

Free