DBPapers
DOI: 10.5593/sgem2017H/63/S27.096

APPLICATION OF THE ARTIFICIAL INTELLIGENCE IN THE ENERGY SOLVING OF BUILDING SERVICES OF INTELLIGENT BUILDINGS AREAS

B. Garlik, P. Hlavacek
Thursday 23 November 2017 by Libadmin2017

References: 17th International Multidisciplinary Scientific GeoConference SGEM 2017, www.sgemviennagreen.org, SGEM2017 Vienna GREEN Conference Proceedings, ISBN 978-619-7408-29-4 / ISSN 1314-2704, 27 - 29 November, 2017, Vol. 17, Issue 63, 767-776 pp; DOI: 10.5593/sgem2017H/63/S27.096

ABSTRACT

Currently, a great emphasis is put on the environment, energy intensity, and therefore on applications of the renewable energy sources (RES). Putting a large amount of RES into the modern distribution systems represents certain problems we can address, inter alia, the variable energy outputs of RES aggregated together, including their accumulation managed by respective energy management system (EMS). The result may be the optimization unit commitment of RES distributed in the micro-grids of a fictitious intelligent city composed of an intelligent buildings (IB) complex. For this purpose, a computer program has been implemented to optimize operating costs to cover energy consumption of intelligent buildings, based on predicted load profiles. The underlying basis of EMS is, among other things, the optimization unit commitment of RES distributed in the micro-grid, distant from a fictitious intelligent city composed of an intelligent buildings complex. For this purpose, a computer program has been implemented to optimize operating costs to cover energy consumption of intelligent buildings, based on predicted load profiles.
As an example of the self-organizing neural network used for cluster analysis (CA), we demonstrate its efficiency in the process of identifying type daily energy consumption diagrams of an intelligent buildings complex combined in the electric micro-grid for a typical working day and typical day off based on network’s annual history. The aforementioned type daily diagrams can be used to predict the power consumption.

Keywords: artificial neural network, cluster analysis, self-organizing map, simulated annealing, unit commitment