НЕВРОННА МРЕЖА ЗА ПРОГНОЗИРАНЕ НА КОЛИЧЕСТВОТО ГЕНЕРИРАНА ОТ СЛЪНЧЕВА РАДИАЦИЯ ЕНЕРГИЯ
NEURAL NETWORK FOR PREDICTING THE ENERGY GENERATED BY SOLAR RADIATION

Author(s): Penka Georgieva
Subject(s): Economy, Energy and Environmental Studies
Published by: Бургаски свободен университет
Keywords: artificial neural network; micro PV system; intelligent building management.
Summary/Abstract: This article proposes an approach for short-term forecasting of electricity production from a micro grid connected PV system using the means of artificial intelligence. An artificial neural network is used to create the model. The forecasting model has three input variables: solar radiation, wind speed and air temperature, and one output variable - the amount of electricity produced by the installed photovoltaic panels in the microsystem. This study is a part of a project for optimization of the energy consumption of a building in case that independent alternative renewable energy sources are used. The model is implemented and tested on real data collected in 5 minutes from a specially designed meteorological station installed in the building of the Burgas Free University.

• Details
• Contents