Power fluctuation damping of thermostatically controlled loads based on state estimation

As a kind of important demand side resource, the thermostatically controlled loads (TCLs) play an important role in the peak load shifting and load balancing programs. This paper establishes a state-queuing (SQ) model for the large-scale TCLs, and estimates the system states based on the Kalman-filtering. Based on state estimation, the control strategy, whose parameters are optimized by the genetic algorithm (GA), is designed to damp the power fluctuation of the TCLs. Finally, the effectiveness of the proposed method is verified.