A methodology based on dynamic artificial neural network for short-term forecasting of the power output of a PV generator

•The output of the majority of renewables energies depends on the variability of the weather conditions.•The short-term forecast is going to be essential for effectively integrating solar energy sources.•A new method based on artificial neural network to predict the power output of a PV generator on...

Full description

Saved in:
Bibliographic Details
Published in:Energy conversion and management 2014-09, Vol.85, p.389-398
Main Authors: Almonacid, F., Pérez-Higueras, P.J., Fernández, Eduardo F., Hontoria, L.
Format: Article
Language:English
Subjects:
Acceptability
Applied sciences
Artificial neural network
Artificial neural networks
Dynamical systems
Dynamics
Economic forecasting
Energy
Equipments, installations and applications
Exact sciences and technology
Forecasting
Generators
Methodology
Natural energy
Photovoltaic conversion
PV generator
Short-term forecasting
Solar energy
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•The output of the majority of renewables energies depends on the variability of the weather conditions.•The short-term forecast is going to be essential for effectively integrating solar energy sources.•A new method based on artificial neural network to predict the power output of a PV generator one hour ahead is proposed.•This new method is based on dynamic artificial neural network to predict global solar irradiance and the air temperature.•The methodology developed can be used to estimate the power output of a PV generator with a satisfactory margin of error. One of the problems of some renewables energies is that the output of these kinds of systems is non-dispatchable depending on variability of weather conditions that cannot be predicted and controlled. From this point of view, the short-term forecast is going to be essential for effectively integrating solar energy sources, being a very useful tool for the reliability and stability of the grid ensuring that an adequate supply is present. In this paper a new methodology for forecasting the output of a PV generator one hour ahead based on dynamic artificial neural network is presented. The results of this study show that the proposed methodology could be used to forecast the power output of PV systems one hour ahead with an acceptable degree of accuracy.
ISSN:0196-8904
1879-2227
DOI:10.1016/j.enconman.2014.05.090