Neural Computing and Applications

, Volume 23, Issue 3, pp 1197–1204

A linear genetic programming approach for the prediction of solar global radiation

Original Article

DOI: 10.1007/s00521-012-1039-6

Cite this article as:
Shavandi, H. & Saeedi Ramyani, S. Neural Comput & Applic (2013) 23: 1197. doi:10.1007/s00521-012-1039-6
  • 261 Downloads

Abstract

In this article, the linear genetic programming (LGP) is utilized to predict the solar global radiation. The solar radiation is formulated in terms of several climatological and meteorological parameters. Comprehensive databases containing monthly data collected for 6 years (1995–2000) in two nominal cities in Iran are used to develop LGP-based models. Separate models are established for each city. To verify the performance of the proposed models, they are applied to estimate the solar global radiation of test data of database. The contribution of the parameters affecting the solar radiation is evaluated through a sensitivity analysis. The results indicate that the LGP models give precise estimations of the solar global radiation and significantly outperform traditional angstrom’s model.

Keywords

Solar global radiation Linear genetic programming Climatological parameters Prediction 

Copyright information

© Springer-Verlag London Limited 2012

Authors and Affiliations

  1. 1.Department of Industrial EngineeringSharif University of TechnologyTehranIran

Personalised recommendations