Selecting and prioritizing the electricity customers for participating in demand response programs

Abstract

Abstract Demand response (DR) provides an opportunity for customers to play an important role in the operation of the electricity grid by reducing or shifting their electricity usage during peak periods. However, selecting customers to participate in DR programs is challenging. To solve this problem, typical load profiles should be characterized by data mining techniques such as clustering algorithms. Traditional clustering algorithms manually determine the centre of clusters and that the selected centre of clusters may fall into a local optimum. Here, to overcome these issues, a new clustering algorithm based on the Density Peak Clustering algorithm (DPC) and Artificial Bee Colony algorithm (ABC) which is called A‐DPC, is implemented to optimally determine the representative load curves. Moreover, by introducing a new priority index, the eligible residential customers are selected for participating in DR programs. Also, to meet the sufficient load reduction in a DR event, the proposed approach suggests a plenty number of residential customers to be called. The result evidence that A‐DPC has a stronger global search ability to optimally select the centre of clusters if compared to other clustering algorithms.