Comparison of regression and artificial neural network models for estimation of global solar radiations. (December 2015)

• Record Type: Journal Article
• Authors: Kumar, Rajesh; Aggarwal, R.K.; Sharma, J.D.
• Is Part Of: Renewable & sustainable energy reviews. Volume 52(2015:Dec.)
• Page Start: 1294
• Page End: 1299
• Publication Date: 2015-12
• Subjects: Solar radiation -- Regression model -- Artificial neural network -- Means absolute percent error; Renewable energy sources -- Periodicals; Power resources -- Periodicals; Énergies renouvelables -- Périodiques; Ressources énergétiques -- Périodiques; 333.794
• DOI: 10.1016/j.rser.2015.08.021
• Languages: English
• ISSNs: 1364-0321

Abstract

Abstract: Various models based on regression as well as artificial neural networks have been studied for the estimation of monthly average global solar radiations. Most of the regression models generally used sunshine hour data for the estimation of global solar radiations on the horizontal surfaces, whereas maximum artificial neural network models have used multilayer feed forward network sigmoid trained with Levenberg–Marquardt back propagation algorithm with different input terminals and different hidden layer neurons. Artificial neural networks have been successfully employed in solving complex problems in various fields such as function approximation, pattern association and pattern recognition, associative memories and generation of new meaningful pattern. Comparison of regression and artificial neural network models have shown that the performance values of the artificial neural network models are better than the regression models. The mean absolute percent error (MAPE) values of the artificial neural network models are lower than those of the regression models. In addition, the R values of the artificial neural network models are higher than those of regression models. The artificial neural network offers an alternative method which cannot be underestimated.