Home // SMART 2013, The Second International Conference on Smart Systems, Devices and Technologies // View article

Advanced Street Lighting Control through Neural Network Ensembling

Authors:
Fabio Moretti
Stefano Pizzuti
Mauro Annunziato
Stefano Panzieri

Keywords: Lighting Efficiency; Energy Management Systems; Adaptive Control; Neural-Network models

Abstract:
In this work, we propose an innovative street lighting energy management system in order to reduce energy consumption. The main goal is to provide ‘energy on demand’ such that energy, in this case light, is provided only when needed. In order to achieve this purpose it is critical to have a reliable demand model, which in the case of street lighting turns out to be a traffic flow rate forecasting model. Several methods has been compared in order to find out one hour prediction model. In our case studies, Artificial Neural Networks performed best results. Moreover, several control strategies have been tested and the one which gave the best energy savings is the adaptive one we carried out. Experimentation has been carried out on two different case studies. In particular we focused our experimentation on public street of a small and a medium sized cities. Our studies show that with the proposed approach it is possible to save up to 50% of energy compared to no regulation systems

Pages: 76 to 81

Copyright: Copyright (c) IARIA, 2013

Publication date: June 23, 2013

Published in: conference

ISSN: 2308-3727

ISBN: 978-1-61208-282-0

Location: Rome, Italy

Dates: from June 23, 2013 to June 28, 2013