Various techniques for electricity customer classification are presented and discussed, with the focus on highlighting the behaviour of electricity customers. The surveyed techniques include classical approaches (applications of statistics and deterministic clustering algorithms), as well as methods based on artificial intelligence (neural networks and fuzzy systems). The classification techniques are illustrated by using various sets of features characterising the shape of the load patterns. Different approaches for feature selection, both in the time and in the frequency domain, are discussed. A number of specific metrics, some of which were originally developed by the authors, are applied in order to quantify the classification adequacy and to identify the most suitable classification techniques. Detailed results obtained from real life applications are provided.
Emergent electricity customer classification / G. CHICCO; R. NAPOLI; F. PIGLIONE; P. POSTOLACHE; M. SCUTARIU; C. TOADER. - In: IEE PROCEEDINGS. GENERATION, TRANSMISSION AND DISTRIBUTION. - ISSN 1350-2360. - STAMPA. - 152:2(2005), pp. 164-172.
|Titolo:||Emergent electricity customer classification|
|Data di pubblicazione:||2005|
|Digital Object Identifier (DOI):||http://dx.doi.org/10.1049/ip-gtd:20041243|
|Appare nelle tipologie:||1.1 Articolo in rivista|