Fast-timescale Control Strategies for Demand Response in Power Systems.

Abstract

Concerns over climate change have spurred an increase in the amount of wind and solar power generation on the grid. While these resources reduce carbon emissions, the physical phenomena that they rely on - wind and sunlight - are highly stochastic, making their generated power less controllable. Demand-side strategies, which modulate load in a controllable manner, have been proposed as a way to add flexibility to the grid.

Resources with innate flexibility in their load profile are particularly suited to demand response (DR) applications. This work examines two such loads: heating, ventilation, and air conditioning (HVAC) systems, and plug-in electric vehicle (PEV) fleets.

HVAC systems can vary the timing of power consumption due to the thermal inertia inherent in their associated building(s). The first part of this thesis explores the efficacy of using commercial HVAC for DR applications. Results are presented from an experimental testbed that quantify performance, in terms of accuracy in perturbing the load in a desired manner, as well as the efficiency of this process.

PEVs offer very fast response times and may eventually represent a significant load on the power system. The second part of this thesis develops several control strategies to manage PEV power consumption in an environment where communication resources are limited, both to prevent detrimental system effects such as transformer overload, and to provide ancillary services such as frequency regulation to the grid.