Application of genetic algorithm and neural network for sizing of hybrid photovoltaic wind power generation (HPVWPG) systems
In this paper, a methodology for sizing of hybrid photovoltaic-wind power generation (HPVWPG) system is described. The purpose of this methodology is to predict the PV-generator area, the capacity of the battery and the capacity of the wind generator in isolated (remote) areas, where the meteorological data (solar irradiation, air temperature, wind speed, etc) are not always available. For this purpose, a database of meteorological data (solar irradiation, ambient temperature and wind speed) for 40 sites located in Algeria has been studied. Firstly, based on the objective function minimisation using genetic algorithms (GA), the different optimal components (PV-generator area, battery capacity and wind generator capacity) for 40 sites were estimated. Subsequently, based on the optimal estimated components, an artificial neural network (ANN) is used for predicting the optimal components of the hybrid system in isolated areas from only the geographical coordinates of the sites. The results showed clearly the potential of the proposed methodology for predicting the optimal sizing components of HPVWPG systems in isolated area.
Keywords: hybrid photovoltaic-wind systems, sizing, genetic algorithms, GAs, artificial neural networks, ANNs, photovoltaics, wind energy, wind power, power generation, meteorological data, solar irradiation, ambient temperature, wind speed, solar power, solar energy
No comments were found for Application of genetic algorithm and neural network for sizing of hybrid photovoltaic wind power generation (HPVWPG) systems. Be the first to comment!