The artificial neural network for solar radiation prediction and designing solar systems: a systematic literature review. (1st October 2015)

• Record Type: Journal Article
• Authors: Qazi, Atika; Fayaz, H.; Wadi, A.; Raj, Ram Gopal; Rahim, N.A.; Khan, Waleed Ahmed
• Is Part Of: Journal of cleaner production. Volume 104(2015:Oct. 01)
• Page Start: 1
• Page End: 12
• Publication Date: 2015-10-01
• Subjects: Solar energy -- Solar radiation prediction -- Solar systems -- Data mining -- Artificial neural network; Factory and trade waste -- Management -- Periodicals; Manufactures -- Environmental aspects -- Periodicals; Déchets industriels -- Gestion -- Périodiques; Usines -- Aspect de l'environnement -- Périodiques; 628.5
• DOI: 10.1016/j.jclepro.2015.04.041
• Languages: English
• ISSNs: 0959-6526

Abstract

Abstract: Solar energy generated by sunlight has a non-schedulable nature due to the stochastic environment of meteorological conditions. Hence, power system control and the energy business require the prediction of solar energy (radiation) from a few seconds up to one week in advance. To deal with prediction shortcomings, various solar radiation prediction methods have been used. Predictive data mining offers variety of methods for solar radiation predictions where artificial neural network is one of the reliable and accurate methods. A systematic review of literature was conducted and identified 24 papers that discuss artificial neural network for solar systems design and solar radiation prediction. The artificial neural network techniques were employed for designing solar systems and predicting solar radiations to assess current literature on the basis of prediction accuracy and inadequacies. Specific inclusion and exclusion criteria in two distinct rounds were applied to determine the most relevant studies for our research goal. Further, it is observed from the result of this study that artificial neural network gives good accuracy in terms of prediction error less than 20%. The accuracy of solar radiation prediction models is found to be dependent on input parameters and architecture type algorithms utilized. Therefore, artificial neural network as compared to other empirical models is capable to deal with many input meteorological parameters, which make it more accurate and reliable. Highlights: Artificial neural network techniques for solar radiation prediction are described. Artificial neural networks for solar systems and prediction accuracy are examined. The comparison between the baseline and artificial neural network models is noted. It is deduced that neural networks (NNs) yield a better prediction outcome.