Bio-inspired Aging Model Particle Swarm Optimization Neural Network Training for Solar Radiation Forecasting Eduardo Rangel 1 , Alma Y. Alan´ ıs 1 , Luis J. Ricalde 2 , Nancy Arana-Daniel 1 , and Carlos L´ opez-Franco 1 1 CUCEI, Universidad de Guadalajara, Apartado Postal 51-71, Col. Las Aguilas, C.P. 45080, Zapopan, Jalisco, Mexico almayalanis@gmail.com 2 UADY, Faculty of Engineering, Av. Industrias no Contaminantes por Periferico Norte, Apdo. Postal 115 Cordemex, Merida, Yucatan, Mexico Abstract. This paper deals with a novel training algorithm for a neu- ral network architecture applied to solar radiation time series prediction. The proposed training algorithm is based in a novel bio-inspired aging model-particle swarm optimization (BAM-PSO). The BAM-PSO based algorithm is employed to update the synaptic weights of the neural net- work. The size of the regression vector is determined by means of the Cao methodology. The proposed structure captures efficiently the com- plex nature of the solar radiation time series. The proposed model is trained and tested using real data values for solar radiation. 1 Introduction The limited existing reserves of fossil fuels and the harmful emissions associ- ated with them have led to an increased focus on renewable energy applications recently. Among renewable energy sources, solar energy is one of the most im- portant techniques. However, in practice the integration of solar energy into the existing electricity supply system is a real challenge because its availability mainly depends on meteorological conditions, which cannot directly be changed by human intervention. For this reason it is important to have a reliable estima- tion of solar radiation. Integration of solar radiation forecast and output power is a good way to improve the performance in scheduling for microgrids. Solar radiation forecast- ing is not an easy task; solar radiation has a stochastic nature with high rate of change. Solar radiation time series present highly nonlinear behavior with no typical patterns and a weak seasonal character [1]. Several methods have been proposed to accomplish solar radiation forecasting like numerical weather prediction systems, statistical approaches and artificial neural networks using feedforward or recurrent structures ([2], [3]). Soft Computing methods are more suitable for short term predictions; these methods are based on time series his- torical data in order to build a mathematical model which approximates the input-output relationship. E. Bayro-Corrochano and E. Hancock (Eds.): CIARP 2014, LNCS 8827, pp. 682–689, 2014. c  Springer International Publishing Switzerland 2014