Advanced Contextual-targeted Building Flexibility Based on Signature Labelling for Demand Response

The residential sector is constantly rising the energy demand and results in stressing the power grids while increasing the need for more power generation units. Consequently, building flexibility is of key importance for the energy supply and response of the grid. In this paper, a novel method to identify potential flexible events within the end-users energy consumption patterns is proposed. A contextual-based unsupervised algorithm is exploited to detect flexible motifs. Consequently, this a priori knowledge is utilized to predict the load patterns that are flexible applying a supervised technique. Eventually, experimental results show case that the proposed method is feasible and accurate for building flexibility and may be utilized for demand response schemes.