|
Clustering methodology for smart metering data based on local and global features Host Publication: 1st International Conference on Internet of Things and Machine Learning Authors: L. Arco Garcia, G. Casas and A. Nowé Publisher: Association for Computing Machinery (ACM) Publication Date: Oct. 2017 Number of Pages: 13
Abstract: In order to develop real intelligent smart grids, understanding the patterns hidden in the smart grid data is crucial. More precisely, the detection of the preferences, behavior and characteristics of consumers and prosumers is crucial. In this work, we introduce a general methodology that groups energy consumption and production of households, based on global as well as local features, allowing the characterization of load and production profiles for consumers and prosumers. Our methodology is illustrated using a two-level clustering approach. The theoreticalresults are applicable in other areas, and have utility in a general business analysis.
|
