A Demand Response Scheme in Smart Grid with Clustering of Residential Customers

Demand response (DR) is one of the crucial technologies in smart grid for its potential benefits to lower the peak load and smooth the residential demand profiles. The existing DR schemes in the literature mainly focus on optimizing customer's load profiles but not enough attentions are paid to important factors such as energy consumption patterns of residential appliances, electricity cost, users' satisfactory level, fairness and energy consumption habits. This paper proposes a flexible DR scheme in smart grid with clustering of residential customers and comprehensively considering the aforementioned factors. New features are extracted from historical data to depict customers' characteristics and clustering methods are applied to explore their electricity consumption habits. Then such information is further utilized to help schedule the residential appliances in a more flexible but effective manner. Numerical results based on real-world traces demonstrate that the proposed DR scheme performs well in reducing the system expenditure and lowering peak to average ratio (PAR). Our research further analyzes the impacts of various factors, including customers' preferences and energy consumption patterns, which sheds some illuminations on how to devise efficient DR strategies.