Fuzzy classification of electrical load demand profiles-a case study

This paper proposes a new method for the automatic classification of large sets of electrical demand profiles using fuzzy relations. With the introduction of profiling and intelligent metering, in the near future, for all customers, it is expected that there will be a set of tariffs that will be mapped to a matrix of customer profiles. This process will require a method of profile clustering for the design of tariffs and a process of classifying any given pattern into these clusters. This paper describes the method developed using fuzzy clustering to identify different groups of similar individual profiles which may be used by supply businesses for tariff development and end user costing.