Document Type

Conference Paper

Rights

This item is available under a Creative Commons License for non-commercial use only

Disciplines

Applied mathematics, Statistics, Probability, Electrical and electronic engineering, Energy and fuels

Publication Details

CIRED Workshop, 29-30 May 2012, Lisbon, Portugal

Abstract

This paper investigates a method of classifying domestic electricity load profiles through Self Organising Maps (SOMs). Approximately four thousand customers are divided into groups based on their electricity demand patterns. Dwelling and occupant characteristics are then investigated for each group. The results show that SOMs are an effective way of classifying customers into groups in terms of their electrical load profile and that certain dwelling and occupant characteristics are significant factors in determining which group they end up in.