Main content area

A novel estimation approach for the solar radiation potential with its complex spatial pattern via machine-learning techniques

Koo, Choongwan, Li, Wenzhuo, Cha, Seung Hyun, Zhang, Shaojie
Renewable energy 2019 v.133 pp. 575-592
artificial intelligence, cities, decision making, environmental impact, equipment, geographic information systems, kriging, models, prediction, solar energy, solar radiation, China
As a clean and sustainable energy resource with lower environmental impact, the Chinese government encourages the application of solar energy system. The global solar radiation on the horizontal surface in the specific site should be investigated in advance so that the solar energy system could be implemented properly and efficiently. However, the monthly average daily solar radiation (MADSR) in China has complex spatial patterns, and its observation stations are still lacking due to the high cost of equipment. To address these challenges, this study aimed to develop a novel estimation approach for the MADSR with its complex spatial pattern over a vast area in China via machine-learning techniques (i.e. a clustering method (k-means) and an advanced case-based reasoning (A-CBR) model). The MADSR and the relevant information were collected from 97 cities in China for 10 years (from 2006 to 2015). The average prediction accuracy of the proposed approach was determined at 93.23%, showing a promising way. The proposed novel approach is expected to be generalized via the interpolation methods (e.g. kriging method in a geographical information system) so that decision-makers (e.g. construction manager or facility manager) can determine the appropriate location, size and form in implementing the solar energy system.