Environmental Engineering Reference
In-Depth Information
transient power drawn from the synchronverter. When the power commands were applied, the
synchronverter responded as if there was no load connected. The local voltage
recovered a bit
because of the power delivered by the synchronverter. After the droop mechanism was enabled
at
t
v
4 s, the synchronverter continued supplying the same real power to the local load but
slightly less reactive power because the local voltage is slightly higher than the nominal value.
When the grid voltage dropped at
t
=
=
5 s by 5%, the local voltage dropped immediately to
about 2
5% below the nominal value, which caused the synchronverter to send an extra 50 Var
of reactive power on top of the command. The responses when
R
.
=
1000
was not changed
are also shown in the right column of Figure 18.5 for comparison.
18.5 Experimental Results
The theory and simulations developed above were verified on an experimental synchronverter.
The parameters of the experimental synchronverter were largely the same as those given in
Table 18.1 and the control parameters were the same. The synchronverter was connected
to a three-phase 400 V 50 Hz grid via a circuit breaker and a step-up transformer. A PLL
was adopted to synchronise the synchronverter with the grid. The sampling frequency of the
controller was 5 kHz and the switching frequency was 12 kHz.
18.5.1 Performance of Power Flow Control
Two experiments were carried out to test the performance of power flow control, one when the
grid frequency was lower than the rated frequency and the other when the grid frequency was
higher than the rated frequency. Both experiments were carried out according to the following
sequence of events:
1. Start the system so that it synchronises with the grid, but keep all the IGBTs off.
2. Start operating the IGBTs, roughly at
t
=
6 s with
P
set
=
Q
set
=
0.
3. Turn the circuit breaker on, roughly at
t
=
11 s.
4. Apply
P
set
=
70 W, roughly at
t
=
16 s.
5. Apply
Q
set
=
21 s.
6. Enable the droop mechanism, roughly at
t
30 Var, roughly at
t
=
=
26 s.
7. Stop data recording, roughly at
t
=
30 s.
18.5.1.1 With a Grid Frequency Lower than 50 Hz
During this experiment, the grid frequencywas lower than 50Hz. The synchronverter frequency
followed the grid frequency very well, with noticeable transients after each action, as shown in
Figure 18.6(a). The local terminal voltage synchronised with the grid voltage very quickly once
the inverter output was enabled, roughly at
t
6 s, as shown in Figures 18.6(b) and 18.6(c).
The connection to the grid went very smoothly and there were no noticeable transients in
the frequency or power, as shown in Figure 18.6. The power remained around 0, but with
bigger spikes. After the real power demand was raised, it took less than 10 cycles to reach the
setpoint, which is very fast, and the overshoot was very small; see Figure 18.6(d). This action
=
Search WWH ::
Custom Search