A Discussion of Mixed Integer Linear Programming Models of Thermostatic Loads in Demand Response

Abstract

In smart grids, it is expected that electricity retailers will offer time-differentiated tariffs with significant price variations in short periods. Consumers are then encouraged to engage in demand response strategies by making the most of their flexibility in the operation of end-use appliances to minimize the electricity bill without compromising on comfort. Air conditioning systems are particularly suited to be controlled, by profiting from the thermal inertia of building units. This paper presents novel mixed integer linear programming formulations to optimize the operation of a thermostatically-controlled air conditioning system, thoroughly discussing their main features and associated computational difficulties resulting from their combinatorial nature.