Title :
On the use of information theoretic mean shift for electricity load patterns clustering
Author :
Sumaili, Jean ; Keko, Hrvoje ; Miranda, V. ; Chicco, Gianfranco
Author_Institution :
Power Syst. Unit, INESC Porto-Inst. de Eng. de Sist. e Comput. do Porto, Porto, Portugal
Abstract :
This paper analyzes the application of the Information Theoretic (IT) Mean Shift algorithm for modes finding in order to provide the classification of Electricity Customer Load Patterns. The impact of the algorithm parameters is discussed and then clustering indices are used in order to make a comparison with the classical methods available. Results show a good capability of the modes found in capturing the data structure, aggregating similar load patterns and identifying the uncommon patterns (outliers).
Keywords :
energy consumption; information theory; pattern classification; pattern clustering; power markets; clustering indices; data structure; electricity customer load pattern classification; electricity load pattern clustering; information theoretic mean shift algorithm; Clustering algorithms; Cost function; Entropy; Indexes; Kernel; Minimization; Partitioning algorithms; clustering; information theoretic learning; load patterns; mean shift; modes finding; outliers;
Conference_Titel :
PowerTech, 2011 IEEE Trondheim
Conference_Location :
Trondheim
Print_ISBN :
978-1-4244-8419-5
Electronic_ISBN :
978-1-4244-8417-1
DOI :
10.1109/PTC.2011.6019390
