Grid-connected Virtual Synchronous Machine using Virtual Inertia Control

Abstract

System frequency stability becomes more difficult to maintain as system inertia steadily declines due to quick spread of inverter based renewable energy. Virtual Synchronous Generator (VSG) control is one of major solutions to this problem since it makes grid-connected inverters mimic inertial and damping properties of traditional synchronous machines. This paper is about modeling, control design and dynamic performance study of VSG inverter using virtual impedance-based architecture in MATLAB/Simulink. The control structure involves incorporating virtual inertia and damping through active power-frequency droop, reactive power-voltage regulation as well as inner voltage-current control loops. Virtual inertia constant (J) and damping coefficient (D) are varied one after the other during step power reference changes. It has been seen that a bigger value of J causes the transient swings to be more intense and the time to come back to normal level to be longer. Whereas with a low D the active power can overshoot up to 34% and the reactive power troughs can be almost three times deeper than those with high damping ones. Eventually all the responses reach the same steady states implying that J and D are purely dynamic parameters. Fast, well damped responses with minimum overshoot are the result of moderate inertia and intermediate damping that corresponds to optimal performance. Such results reveal the very basic inertia-damping trade-off in VSG designs and show that coordinated adjusting of the parameters is a must for a stable transient and a reliable incorporation of renewable energy.