A review of Electricity Load Profile Classification methods

Abstract

With the electricity market liberalisation in Indonesia, the electricity companies will have the right to develop tariff rates independently. Thus, precise knowledge of load profile classifications of customers will become essential for designing a variety of tariff options, in which the tariff rates are in line with efficient revenue generation and will encourage optimum take up of the available electricity supplies, by various types of customers. Since the early days of the liberalisation of the Electricity Supply Industries (ESI) considerable efforts have been made to investigate methodologies to form optimal tariffs based on customer classes, derived from various clustering and classification techniques. Clustering techniques are analytical processes which are used to develop groups (classes) of customers based on their behaviour and to derive representative sets of load profiles and help build models for daily load shapes. Whereas classification techniques are processes that start by analysing load demand data (LDD) from various customers and then identify the groups that these customers' LDD fall into. In this paper we will review some of the popular clustering algorithms, explain the difference between each method.

Data science for building energy management: A review

“…In line with this, Prahastono et al [25] presented an overview of common clustering methods (e.g. hierarchical, k -means, fuzzy k -means, follow the leader and fuzzy relation) and compared their effectiveness in the classification of customers and the generation of electricity load profiles (in terms of average retail price and net generation).…”

A review on clustering of residential electricity customers and its applications

“…In this regard, distance measures are used to find similarity or dissimilarity between any pair of objects. For numeric attributes like electricity consumption data, measures like Minkowski (L p -norm), cosine similarity, and dynamic time warping (DTW) are usually used [4] [8,9].…”

Electricity load profile classification using Fuzzy C-Means method

“…The Fuzzy C-means (FCM) method is a clustering technique wherein each data point belongs to every cluster to some degree that is specified by a membership grade. The procedure is similar to standard K-means [1], the difference is that each data set has a degree of membership to each initial cluster [2], i.e. each data set belongs to all clusters to some degree.…”