Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.14279/13364
Title: A Neural Network Approach for short-term forecasting of PV Generation in Dwellings
Authors: Georgiou, Giorgos S. 
Christodoulides, Paul 
Kalogirou, Soteris A. 
Major Field of Science: Engineering and Technology
Field Category: Environmental Engineering
Keywords: Artificial Neural Networks;Buildings;Forecasting;Photovoltaics;Renewable Energy Generation
Issue Date: Sep-2018
Source: 53rd International Universities Power Engineering Conference, 2018, 4-7 September, Glasgow, UK
Conference: International Universities Power Engineering Conference 
Abstract: The use of renewable energy, especially in buildings, has continuously significantly been increasing, due to the need of reducing the demand from energy grids. Hence, forecasting the renewable energy generation is beneficial, even in buildings, as the energy demand can be optimized for a certain time horizon, resulting in the further reduction of the energy bills, as compared to traditional ways such as Net-Metering. This paper shows the preliminary results of an undergoing research regarding the 24-hour prediction of a PV production, in a dwelling in Cyprus. For the given data and the case studied, the results exhibit an overall correlation of 99% approx. and a Mean Squared Error of 1.4% approx. for a cloudy day.
URI: https://hdl.handle.net/20.500.14279/13364
DOI: 10.1109/UPEC.2018.8541925
Rights: © 2018 IEEE.
Type: Conference Papers
Affiliation : Cyprus University of Technology 
Publication Type: Peer Reviewed
Appears in Collections:Δημοσιεύσεις σε συνέδρια /Conference papers or poster or presentation

CORE Recommender
Show full item record

Page view(s) 1

393
Last Week
0
Last month
4
checked on Nov 6, 2024

Google ScholarTM

Check

Altmetric


Items in KTISIS are protected by copyright, with all rights reserved, unless otherwise indicated.