Load pattern clustering based on the shape of the electricity consumption is a key tool to provide enhanced knowledge on the nature of the consumption and assist meaningful customer partitioning. This paper presents new developments to group the load patterns using an initial set of centroids specified according to a user-defined centroid model. The original Electrical Pattern Ant Colony Clustering (EPACC) algorithm is illustrated, highlighting its characteristics and parameters, with centroids evolution during the iterative process until stabilization. The EPACC results are compared with those obtained from the classical k-means algorithm to group the representative load patterns taken from a set of non-residential customers in typical weekdays.
Electrical Load Pattern Grouping Based on Centroid Model with Ant Colony Clustering / Gianfranco Chicco; Octavian-Marcel Ionel; Radu Porumb. - In: IEEE TRANSACTIONS ON POWER SYSTEMS. - ISSN 0885-8950. - STAMPA. - 28:No. 2, May 2013(2013), pp. 1706-1715. [10.1109/TPWRS.2012.2220159]
|Titolo:||Electrical Load Pattern Grouping Based on Centroid Model with Ant Colony Clustering|
|Data di pubblicazione:||2013|
|Digital Object Identifier (DOI):||http://dx.doi.org/10.1109/TPWRS.2012.2220159|
|Appare nelle tipologie:||1.1 Articolo in rivista|