Home // ENERGY 2012, The Second International Conference on Smart Grids, Green Communications and IT Energy-aware Technologies // View article
Authors:
Christian Schwarz
Felix Leupold
Tobias Schubotz
Keywords: energy pattern recognition; machine learning; in-memory database;
Abstract:
Todays energy companies mainly use generalized demand sets to predict the required amount of energy of their customers on a high aggregation level. This is sufficient in an energy consumption oriented power grid, having enough resources to produce and transmit the requested and produced amount of energy. With the increasing amount of renewable energy sources, the power grid evolves from a purely consumption controlled supply network to a production controlled grid. In that environment the need for detailed short term energy demand predictions increases. A first step to predict the demand of energy is to find generalizable patterns within the energy consumption data that can later on be used for early predictions on real time data. To study the possibility to predict energy patterns nearly real-time, we created an environment, where metering data is collected every second and used for real-time pattern matching. We developed and implemented pattern recognition algorithms that use the abilities of in-memory databases with the collected metering data in order to detect energy consumption patterns.
Pages: 7 to 12
Copyright: Copyright (c) IARIA, 2012
Publication date: March 25, 2012
Published in: conference
ISSN: 2308-412X
ISBN: 978-1-61208-189-2
Location: St. Maarten, The Netherlands Antilles
Dates: from March 25, 2012 to March 30, 2012