A Hierarchical Scheduling Strategy of Thermostatically Controlled Loads in Smart Grid

Abstract

A hierarchical scheduling strategy of thermostatically controlled loads (TCLs) is proposed to relieve the mismatch between supply and demand in smart grid, which considers the consumer discomfort, the potential capacity of multi-aggregator and the optimal allocation of frequency regulation signal to each load aggregator. The hierarchical scheduling framework is composed of power company layer, agent layer and aggregator layer. First, in power company layer, the power company issues adjustment tasks and generates frequency regulation signals for load agent. Second, in the agent layer, when receiving the frequency regulation signal from the power company, the agent divides the the regulation signal into sub-regulation signals and allocate them to each aggregetor, in order to minimize the economic cost including management cost, discomfort cost and deviation cost. In the aggregator layer, we design the sliding mode controller based on the reaching law to control heterogeneous TCL aggregators. Furthermore, the genetic algorithm (GA) is applied to solve the optimization problem of the agent layer and optimize the parameters of the controller. Finally, simulation results show that the hierarchical scheduling strategy achieves optimization of economical cost as well as the improvement of consumer comfort, and provides the promising auxiliary service.