Hybrid model analysis and validation for PV energy production forecasting

Author

Gandelli, A. ; Grimaccia, F. ; Leva, S. ; Mussetta, M. ; Ogliari, E.

Author_Institution

Dept. of Energy, Politec. di Milano, Milan, Italy

fYear

2014

fDate

6-11 July 2014

Firstpage

1957

Lastpage

1962

Abstract

In this paper a forecasting method for the Next Day´s energy production forecast is proposed with respect to photovoltaic plants. A new hybrid method PHANN (Physical Hybrid Artificial Neural Network) based on Artificial Neural Network (ANN) and basic Physical constraints of the PV plant, is presented and compared with an ANN standard method. Furthermore, the accuracy of the two methods have been studied in order to better understand the intrinsic error committed by the PHANN, reporting some numerical results. This computing-based hybrid approach is proposed for PV energy forecasting in view of optimal usage and management of RES in future smart grid applications.

Keywords

neural nets; photovoltaic power systems; power engineering computing; ANN standard method; PHANN; PV energy production forecasting; PV plant; RES management; computing-based hybrid approach; hybrid model analysis; photovoltaic plants; physical hybrid artificial neural network; smart grid applications; Artificial neural networks; Forecasting; Predictive models; Production; Solar radiation; Training; Weather forecasting;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks (IJCNN), 2014 International Joint Conference on

Conference_Location

Beijing

Print_ISBN

978-1-4799-6627-1

Type

conf

DOI

10.1109/IJCNN.2014.6889786

Filename

6889786