Integration of Generalized Regression Neural Network (GRNN) and Particle Swarm Optimization (PSO) for solar radiation prediction

Measured simple atmospheric parameters between 1985 and 2005 for Esfahan city in Iran (latitude N 32°67', longitude E 51°67', elevation 1550.4 m), were used for the estimation of global solar radiation (GSR) in future time domain using an Integrated Generalized Regression Neural Network (GRNN) and Particle Swarm Optimization (PSO) method. To achieve this, month of the year, air temperature and relative humidity were used as inputs to the neural networks and the GSR used as output. Measured weather data from 1988 to 2001, was used in order to train the networks while the measured data from 2002 to 2005 was used for validating the trained networks. This study confirms the ability of the GRNNs to predict solar radiation values.

کلیدواژه ها:

Solar Radiation ، Relative humidity ، Generalized Regression Neural Networks (GRNNs) ، Air temperature ، Particle Swarm Optimization (PSO)