Load power estimation based secondary control for microgrids

The well-known active power-frequency and reactive power-voltage amplitude droop control is widely used in the coordinative control of parallel inverters in microgrids. However, this conventional droop method may cause frequency and voltage deviation, which affects the accuracy of power supply. This paper proposes a novel secondary control strategy to compensate this deviation. This strategy mimics the Master-Slave control but requires no communication lines among the parallel inverters. The master inverter adopts conventional droop method (using power to control frequency and voltage amplitude) and is controlled as a voltage source while the slave inverters adopt reversed droop method (using frequency and voltage amplitude to control power) and are controlled as current sources. The droop characteristic bias of slave inverters is designed online based on the estimation of load power demand. Through this method, frequency and voltage deviation can be eliminated and power sharing can be realized among all the slave inverters. Simulation and experimental results are provided to prove the effectiveness of the proposed control strategy.