Queensland load profiling by using clustering techniques

Author

Colley, Daven ; Mahmoudi, Nadali ; Eghbal, Daniel ; Saha, Tapan K.

Author_Institution

Sch. of Inf. Technol. & Electr. Eng., Univ. of Queensland, Brisbane, QLD, Australia

fYear

2014

fDate

Sept. 28 2014-Oct. 1 2014

Firstpage

1

Lastpage

6

Abstract

Load profiling is essential in power systems operation and planning. Accurate load profiles lead to a better load scheduling as well as load and price forecasting. Clustering techniques are used to provide an enhanced knowledge on electrical load patterns. This paper deals with clustering methods to analyze Queensland´s load data. The K-means clustering method is used here, where its accuracy is measured using the clustering dispersion indicator (CDI). This method is applied on the Queensland load curves in 2013, where distinct monthly and yearly load profiles are obtained. In addition, the characteristic of each load profile depending on the day type and weather conditions are analyzed.

Keywords

load forecasting; power system planning; K-means clustering method; Queensland load curves; Queensland load profiling; clustering dispersion indicator; clustering techniques; electrical load patterns; load forecasting; load scheduling; power systems operation; power systems planning; price forecasting; Algorithm design and analysis; Australia; Clustering algorithms; Clustering methods; Dispersion; Educational institutions; Power engineering; K-means; clustering dispersion indicator; clustering technique; electrical load profiling; load pattern analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Power Engineering Conference (AUPEC), 2014 Australasian Universities

Conference_Location

Perth, WA

Type

conf

DOI

10.1109/AUPEC.2014.6966554

Filename

6966554