Environmental Engineering Reference
In-Depth Information
n
i
P
i
K
e
E
∗
where
is the voltage drop ratio. Note that this voltage drop ratio is the overall effective
voltage drop ratio, which is much smaller than the drop ratio due to the droop effect and/or
the load effect, but the voltage drop ratio in both the conventional droop controller and the
controller in (Sao and Lehn 2005) is just the voltage drop ratio due to the droop effect and does
not include the voltage drop ratio due to the load effect. Although the controller in (Sao and
Lehn 2005) can compensate for the voltage drop due to the load effect, it cannot compensate
for the voltage drop due to the droop effect. The robust droop control strategy can compensate
for the voltage drop due to both effects and, hence, offers much better capability of voltage
regulation. The voltage drop here is no longer determined by the output impedance originally
designed as characterised in (19.1) but by the parameters
n
i
,
K
e
and the actual power
P
i
.
It can be considerably reduced by using a large
K
e
. If there are errors in the RMS voltage
measurement, then the trade-off between the voltage drop and the accuracy of power sharing
has to be made because the voltage drop is proportional to
n
i
K
e
but the sharing error is inverse
proportional to
n
K
e
.
Here is a calculation example. Assume that the voltage drop ratio at the rated power is
n
i
P
i
K
e
E
∗
=
10% and the error in the RMS voltage measurement is
V
o
E
∗
=
5%, whether
because the local voltages of inverters are measured or because the sensors are not accurate.
Then, the error in the real power sharing is
K
e
E
∗
n
i
S
i
0
.
V
o
E
∗
=
0
5%
10%
=
.
5%, which is reasonable.
It is worth noting that the robust droop control still contains the voltage droop function.
What is different from the conventional droop control is that the voltage droop is applied to the
output voltage
V
o
but not to the voltage set-point
E
, which is able to improve the performance
significantly.
19.6.4 Error Due to the Global Settings for E
∗
and
ω
∗
This sub-section is devoted to the sensitivity analysis of the error in the global settings
E
∗
and
ω
∗
for the robust droop controller.
Any small error
ω
i
would lead to the reactive power deviation (if still stable) of
ω
i
in
1
m
i
ω
i
,
Q
i
=−
Q
1
+
Q
2
,the
according to (19.6). For two inverters operated in parallel with
Q
1
+
Q
2
=
ω
=
ω
2
−
ω
1
=
ω
2
−
ω
1
is
relative reactive power sharing error due to the error
ω
∗
m
i
Q
i
Q
1
Q
1
−
Q
2
Q
2
=
Q
1
Q
1
−
Q
2
Q
2
ω
ω
∗
,
e
Q
%
=
=
ω
∗
m
i
Q
i
where
is the inverse of the frequency boost ratio at the rated reactive power. The smaller
the frequency boost ratio, the bigger the reactive power sharing error; the bigger the error
ω
,
the bigger the sharing error. For example, for a typical frequency boost ratio of
m
i
Q
i
ω
∗
=
1%,
Search WWH ::
Custom Search