A Soft Computing Methodology for Estimation and Forecasting of Daily Global Solar Radiation (DGSR)
Energy is one of the most crucial building blocks of economic development. The energy sector in India has a rapid growth in recent years. The factors that took the nation to a very acute energy crisis are increase in population, transportation, urbanization, industrialization, high standard of living and fast depleting fossil fuels. Energy demand problems have increased all over the world, and renewable energy sources are more crucial to solve these problems. Solar energy is a stepping stone to satisfy the growing energy demands in India and across the globe. In this research work, Coimbatore location was considered for daily global solar radiation (DGSR) analysis and the meteorological parameters used for this research are minimum air temperature, global solar radiation, maximum air temperature, sunshine hours, wind speed, mean air temperature, extraterrestrial radiation, average atmospheric pressure, average precipitation and relative humidity. Statistical models and artificial intelligence computational technique with ANFIS, i.e. adaptive neuro-fuzzy inference system, were used for forecasting and estimation of DGSR. The results of ANFIS model were found to be the best fit for forecasting and estimation of DGSR in any region.
KeywordsSolar energy Estimation and forecasting of solar radiation Statistical models ANFIS ANFIS chaotic time series prediction
- 2.Dwivedi, A., Bari, A., Dwivedi, G.: Scope and application of solar thermal energy in India—a review. Int. J. Eng. Res. Technol. 6(3), 315–322 (2013)Google Scholar
- 9.Krishnaiah, T., Srinivasa Rao, S., Madhumurthy, K., Reddy, K.S.: Neural network approach for modelling global solar radiation. J. Appl. Sci. Res. 3(10), 1105–1111 (2007)Google Scholar
- 18.Swartman, R.K., Ogunlade, O.: Solar radiation estimates from common parameters. In: Solar Energy Conference, Tempe, Arizona (1967)Google Scholar