Centralized versus Decentralized Frameworks for Residential Demand Response Potentials Coordination

Coordination of residential demand response (DR)potentials is crucial to avoid severe peak rebounds. In this paper,two frameworks are presented for domestic demand sidemanagement and benefits and drawbacks of them are comparedto each other. The two proposed frameworks are based oncentralized and decentralized approaches. In the centralizedframework, distribution system operator (DSO) gathersinformation of appliances and preferences of all customers andoptimizes the operation of responsive appliances. The optimizedoperation parameters of responsive appliances are then sent byDSO as control signals to the customers. On the other hand, inthe decentralized framework, each customer improves theoperation of its responsive loads by considering its preferences inrespect to received signals from DSO. Then, the improved loadprofiles are sent back to DSO for checkout. In this paper, it isassumed that the operation of responsive appliances is controlledsuch that more evenly distributed load profile is achieved. Inorder to investigate the performance of the two frameworks, theyare applied to several case studies with different numbers ofconsumers who participated in the DR program. The resultsshow that both methods are effective in modifying system loadprofile. It is observed that the modification is more significant asmore consumers are enrolled in the DR program. However, inthe centralized framework, the execution time and the requiredmemory increase exponentially by growing the number ofenrolled customers.

Keywords:

demand response , smart distribution network , centralized method , decentralized method , distribution system operator