Demand response, or load control, enables interactions between end-users and the grid through adapting end-users’ energy consumption to time-based pricing signals. This paper deals with the scheduling issue of demand response in residential distribution networks. The utility company considers a cost function representing the cost of providing energy to end-users. End-users’ smart appliances that can perform demand response include air-conditioning units and chargers of plug-in electric vehicles. In addition, operation of smart appliances away from desired power levels can lead to user dissatisfactions. The key problem is to minimize the electricity provider cost plus the dissatisfaction across users. The paper develops a distributed algorithm to solve the problem. The utility company and the end-users exchange messages through the Advanced Metering Infrastructure (AMI)—a two-way communication network—to obtain the optimal amount of electricity production and optimal end-user consumption schedules. The algorithm computes near-optimal schedules even when AMI messages are lost, which can happen due to cyber-attacks or malfunctions in the AMI network. An additional desirable feature is that end-user privacy is preserved, because user preferences with respect to appliance operation times need not be revealed to the utility company. The developed algorithm may facilitate the adoption of smart grid technologies, such as smart meters with processing units and communication capabilities, and demand response programs.
Residential Load Control: Distributed Scheduling and Convergence With Lost AMI Messages
Statements and opinions given in a work published by the IEEE or the IEEE Communications Society are the expressions of the author(s). Responsibility for the content of published articles rests upon the authors(s), not IEEE nor the IEEE Communications Society.