Optimal Residential Demand Response in Distribution Networks

• Published in: IEEE journal on selected areas in communications Vol. 32; no. 7; pp. 1441 - 1450
• Main Authors: Shi, Wenbo, Li, Na, Xie, Xiaorong, Chu, Chi-Cheng, Gadh, Rajit
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.07.2014; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Aggregates; Computer simulation; Demand; Electric utilities; Electricity; Energy policy; Home appliances; Households; Load modeling; Networks; Optimization; Power distribution; Power flow; Power systems; Residential energy; Schedules; Simulation; optimal power flow (OPF); smart grid; distribution networks; distributed algorithms; Demand response (DR)
• ISSN: 0733-8716; 1558-0008
• DOI: 10.1109/JSAC.2014.2332131

Demand response (DR) enables customers to adjust their electricity usage to balance supply and demand. Most previous works on DR consider the supply-demand matching in an abstract way without taking into account the underlying power distribution network and the associated power flow and system operational constraints. As a result, the schemes proposed by those works may end up with electricity consumption/shedding decisions that violate those constraints and thus are not feasible. In this paper, we study residential DR with consideration of the power distribution network and the associated constraints. We formulate residential DR as an optimal power flow problem and propose a distributed scheme where the load service entity and the households interactively communicate to compute an optimal demand schedule. To complement our theoretical results, we also simulate an IEEE test distribution system. The simulation results demonstrate two interesting effects of DR. One is the location effect, meaning that the households far away from the feeder tend to reduce more demands in DR. The other is the rebound effect, meaning that DR may create a new peak after the DR event ends if the DR parameters are not chosen carefully. The two effects suggest certain rules we should follow when designing a DR program.