Targeted demand response for flexible energy communities using clustering techniques

The present study explores the use of clustering techniques for the design and implementation of a demand response (DR) program for commercial and residential prosumers. The goal of the program is to alter the consumption behavior of the prosumers pertaining to a distributed energy community in Italy. This aggregation aims to: a) minimize the reverse power flow at the primary substation, that occurs when generation from solar panels in the local grid exceeds consumption, and b) shave the system wide peak demand, that typically occurs during the hours of late afternoon. Regarding the clustering stage, three popular machine learning algorithms for electrical load clustering are employed -namely k-means, k-medoids and an agglomerative hierarchical clustering- alongside two different distance measures -namely euclidean and constrained dynamic time warping (DTW). We evaluate the methods using multiple validation metrics including a novel metric -namely peak performance score (PPS)- that we propose in the context of this study. The best model is employed to divide daily prosumer load profiles into clusters and each cluster is analyzed in terms of load shape, mean entropy, and load type distribution. These characteristics are then used to distinguish the clusters that have the potential to serve the optimization objectives by matching them to appropriate DR schemes including time of use (TOU), critical peak pricing (CPP), and real-time pricing (RTP). The results of this study can be useful for network operators, utilities, and aggregators that aim to develop targeted DR programs for groups of prosumers within flexible energy communities.