Machine learning models to quantify and map daily global solar radiation and photovoltaic power

Yu Feng, Weiping Hao, Haoru Li, Ningbo Cui, Daozhi Gong and Lili Gao

Renewable and Sustainable Energy Reviews, 2020, vol. 118, issue C

Abstract: Global solar radiation (Rs) reaching Earth's surface is the primary information for the design and application of solar energy-related systems. High-resolution Rs measurements are limited owing to the high costs of measuring devices, and their stringent operational maintenance procedures. This study evaluated a newly developed machine learning model, namely the hybrid particle swarm optimization and extreme learning machine (PSO-ELM), to accurately predict daily Rs. The newly proposed model was compared with five other machine learning models, namely the original ELM, support vector machine, generalized regression neural networks, M5 model tree, and autoencoder, under two training scenarios using long-term Rs and other climatic data taken during 1961–2016 from seven stations located on the Loess Plateau of China. Overall, the PSO-ELM with full climatic data as inputs provided more accurate Rs estimations. We also calculated the daily Rs at fifty other stations without Rs measurements on the Loess Plateau using the PSO-ELM model, as well as the potential photovoltaic (PV) power using an empirical PV power model, and then generated high-resolution (0.25°) Rs and PV power data to investigate the patterns of Rs and PV power. Significant reductions in Rs (−6.49 MJ m−2 per year, p < 0.05) and PV power (−0.46 kWh m−2 per year, p < 0.05) were observed. The northwestern parts of the study area exhibited more Rs and PV power and are therefore considered more favorable for solar energy-related applications. Our study confirms the effectiveness of the PSO-ELM for solar energy modeling, particularly in areas where in-situ measurements are unavailable.

Keywords: Machine learning; Solar radiation; Model comparison; Photovoltaic power; Loess Plateau of China (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (20)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S136403211930601X
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:rensus:v:118:y:2020:i:c:s136403211930601x

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/journaldescription.cws_home/600126/bibliographic
http://www.elsevier. ... 600126/bibliographic

DOI: 10.1016/j.rser.2019.109393

Access Statistics for this article

Renewable and Sustainable Energy Reviews is currently edited by L. Kazmerski

More articles in Renewable and Sustainable Energy Reviews from Elsevier
Bibliographic data for series maintained by Catherine Liu ().