13 November 2006 in Data & Knowledge Engineering
23 July 2018 in Soft Computing
1 April 2013 in IET Generation, Transmission & Distribution
GCGuilherme Augusto Barros Conde
Guilherme Augusto Barros Conde
FSFábio Correa dos Santos
MTMaria Emília de Lima Tostes
Maria Emília de Lima Tostes
The Brazilian electrical utility companies must meet continuity indicators for energy supply, which are represented by the indices of average interruption duration and frequency, according to targets established by the Brazilian Regulatory Agency for Electrical Energy (ANEEL). In a nationwide base, ANEEL has defined 30 clusters, each one having specific targets for Customer Average Duration Interruption Index and Customer Average Frequency Interruption Index; still, very frequently the utility distribution companies are financially penalised for not meeting these indicator targets. This study proposes a decision support system based on machine learning techniques so that the utility distribution companies can emulate the characteristics and procedures used by the ANEEL, and help in obtaining more adequate customer groups to evaluate the duration and frequency indicators. The proposed system was applied in a case study of a distribution utility whose supply area is located in the Brazilian Amazonia. The methodology proved to be adequate for seeking better customer grouping configurations that could result in a decrease in goal violations as well as providing more consistent goals, considering the specific characteristics of each distribution utility. Although this methodology was applied to a Brazilian scenario it also can be applied to other distribution utilities worldwide.