Synchronverters: Grid-friendly Inverters that Mimic Synchronous Generators - Control of Power Inverters in Renewable Energy and Smart Grid Integration - page 273

Because there is no delay involved in the frequency (speed) loop, the time constant τ f can be

made much smaller than that of a physical synchronous generator. In order to make sure that
the frequency loop has a quick response so that it can track the frequency reference quickly, τ f

should be made small. Hence, for a given frequency droop coefficient D p , J should be made

small. It is not necessary to have a large inertia as for a physical synchronous generator, where
a larger inertia means that more energy is stored mechanically. The energy storage function
of a synchronverter can, and should, be decoupled from the inertia. This is the opposite of the
approach proposed in (Driesen and Visscher 2008). The short-term energy storage function
can be implemented with a synchronverter using the same storage system, e.g. batteries, that is
used for long-term storage. Usually, a synchronverter would be operated in conjunction with
a distributed power source and an energy storage unit that would be connected to the DC bus
via a DC/DC converter.

18.3.2 Regulation of Reactive Power and Voltage Droop Control
The regulation of reactive power Q flowing out of the synchronverter can be realised similarly.
Define the voltage droop coefficient D q as the ratio of the required change of reactive power

Q to the change of voltage v , i.e.,

D q =

Q
v

=

Q
Q n

v n
v

Q n
v n ,

where Q n is the nominal reactive power, which can be chosen as the nominal power, and v n is

the nominal amplitude of the terminal voltage v . Again, note that in much of the literature, e.g.

(Sao and Lehn 2005), D q is defined as v

Q . The regulation mechanism for the reactive power
can be realised as shown in the lower part of Figure 18.4. The difference between the reference
voltage

v fb is amplified with the voltage droop
coefficient D q before adding to the difference between the set point Q set and the reactive power
Q , which is calculated according to (18.9). The resulting signal is then fed into an integrator
with a gain 1 K to generate M f i f . Here, K is dual to the inertia J . Note that there is no need to
measure reactive power Q because it is available internally.
Similarly, the regulation mechanism of the reactive power shown in the lower part of
Figure 18.4 also has a cascaded control structure, if the effect of the LC filter is ignored or
compensated, which means

v r and the amplitude of the feedback voltage

e . The inner loop is the (amplitude) voltage loop and the
outer loop is the reactive power loop. The time constant

v fb ≈

τ v

of the voltage loop is

K K

τ v =

D q ≈

˙ ˙

θ θ n D q

as the variation of ˙ θ is very small. Hence, K can be chosen as

˙

= θ n D q τ v .

K