A Balancing Demand Response Clustering Approach of Domestic Electricity for Day-Ahead Markets

Abstract

This paper introduces a new clustering approach for multi-customer intelligent demand response for customers living in the same or closer smart grid locations using real electricity consumption data from smart meters. Most of the demand side management or customer tariffs focused on a single customer to optimize their usage discarding the others connected to the same grid. The proposed balancing clustering focus on the customers connected to the same or closest grid to optimize the smooth operating of the energy producers. This approach offers a triple win-win-win model for peak and low consumption customers as well as the balancing for the producer/ distributor utility companies for planning the day ahead markets. This paper uses the most widely used clustering method of k-means for finding similar customers on the opposing side peak, low consumption profiles and combines the most distinguished customers forming more uniform consumption for day-ahead market. This customer balancing and grouping them provides a better way toaggregate residential load data for power buy and sell for all sides and results in better load scheduling.