the limitations of digital simulation and the advantages ... · pdf filelimitations of digital...
TRANSCRIPT
Panos Kotsampopoulos: Smart RUE-National Technical University of Athens
Felix Lehfuss: Austrian Institute of Technology - AIT
Georg Lauss: Austrian Institute of Technology - AIT
Benoit Bletterie: Austrian Institute of Technology - AIT
Nikos Hatziargyriou: Smart RUE-National Technical University of Athens
Limitations of Digital Simulation and Advantages of PHIL Testing
for DG providing Ancillary Services
DERlab General Assembly-Workshop, Paris, March 2015
Evolution of power system/component
analysis, testing and validation
Full System Hardware Testing
Component Hardware Testing Pure Simulation Methods
Hardware in the Loop (HIL): System Testing
Full system digital simulationNumeric analysis
Grid Simulator
InverterDuT
PV arraySimulator
Power Amplifier
InverterDuT
Power Amplifier
Network, Substation
PV arrays
AC side DC side
AC side DC side
Comm. Comm.
InverterDuT
RTS:PV arrays
RTS: Network,
Substation
PowerPower
PHIL Fault-Ride-Through tests
on PV inverters
Simulated network according to
Standards for Fault-Ride-Through
Fault in the simulated network in the RTDS
The hardware PV
inverter is disconnected
following the relevant
standard
The settings of the hardware-software relays are adapted to the network
conditions (e.g. DG on or off)
Adaptive protection testing via
HIL simulation
PHIL testing of ancillary services by
inverter-based DG
Inverter-based DG are particularly difficult to model
accurately
Therefore the “classic” pure digital simulation can face
limitations
PHIL makes use of the real DG and can therefore reveal the
actual system behaviour Software Hardware
grid-connected
PV inverter
DC sourceline impedance
transformer
programmable
voltage source
HV network MV line 10km
OLTC control
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
DC +
DC -
+ DC
- DC
Ssc=1 GVA
R/X=1/6110/30kV, 30 MVA
Ux=11.92%, Ur=0.37%
R=0.253 Ω/km
X=0.355 Ω/km
measurement: V
command:
tap change
New scaling and stability-accuracy method
PHIL allows scaling of the software and hardware. A 2kVA hardware LV PV
inverter is used to evaluate the integration of a 15MVA PV park connected
to the MV. The grid voltage control algorithm of the full and
reduced-scale DGs are exactly the same.
Shifting part of the software impedance on the hardware side is proposed.
Stability can be achieved without compromising accuracy (if necessary a
“smaller” feedback filter can be used)
e-sΤ1
Gfilter(s)
GLPF(s)
(optional)
1/α
RMV·b/α -
Rsh·α
ΧMV·b/α -
Χsh·α
Rsh Χsh
Inverter
filter
Shifted Hardware
Impedance
Equivalent part of
the MV line
Software: MV Hardware: LV
Vcn_s
+
-
A+
iHUT
Rtransf·b/α
Xtransf·b/α
e-sΤ2
Vcn_h
V se
cond
ary
V in
verte
r
+
-
+
-
HW
SW
V
Va
redHuT
fullHuT
S
Sb
_
_
.)(
)()(
sZa
sZa
bsZ
sh
SWSW
Accuracy Improvement with the proposed
method
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 200010
-3
10-2
10-1
100
101
102
103
frequency (Hz)
Err
or
(%)
No shifting, LPF 500 Hz
Shifting, No LPF
GLPF(s)ZSW(s) 1/ZHW(s)
Vs e-sTd Gfilter(s)+VzHW
-
GLPF(s)ZSW(s)−a·Zsh(s) 1/ZHW(s)
Vs e-sTd Gfilter(s)+VzHW
1/a- ZHW(s)+Zsh(s)
ZHW(s)
(a)
(b)
1/a
No need for feedback filter
with the shifting impedances
→ Much higher accuracy
Formula for the minimum
shifted impedance to achieve
stablity was derived
Down-scaling of the voltage
improves the stability
Up-scaling of the current
deteriorates the stability
DG and OLTC interactions: cosφ(P)
DG operates with cosφ(P) control
DG active power increases, stays constant and then decreases
Recurring tap-changes occur
Good accuracy of the PHIL test due to the proposed method
Software Hardware
grid-connected
PV inverter
DC sourceline impedance
transformer
programmable
voltage source
HV network MV line 10km
OLTC control
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
DC +
DC -
+ DC
- DC
Ssc=1 GVA
R/X=1/6110/30kV, 30 MVA
Ux=11.92%, Ur=0.37%
R=0.253 Ω/km
X=0.355 Ω/km
measurement: V
command:
tap change
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
0.97
0.98
0.99
1
1.01
1.02
1.03
1.04
1.05
1.06
time (s)
Vo
lta
ge
p.u
.
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
-4
-3
-2
-1
0
1
time (s)
tap
po
sitio
n
tap PHIL
tap simulation
Vinverter simulation
Vsecondary simulation Vsecondary PHIL
Vinverter PHIL
DG and OLTC interactions: Q(U)
Active power of the DG increases → DG voltage increases → reactive
power absorption by the DG increases (Q(U)) → Voltage of the secondary of
the transformer decreases → tap-change occurs
Recurring tap-changes occur
Instability of the Q(U) controller (i.e. Oscillations): not visible at
the pure digital simulation
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
0.99
1
1.01
1.02
1.03
time (s)
Vo
lta
ge
p.u
.
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80-3
-2
-1
0
1
2
time (s)
tap
po
sitio
n
tap PHIL
tap simulation
Vinverter simulation Vinverter PHIL
Vsecondary PHIL
Vsecondary simulation
DG and OLTC interactions: Q(U)
Additional reactive power flow
Instability of the Q(U) controller (i.e. Oscillations):
not visible at the pure digital simulation
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80-1.2
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
time (s)
P/P
no
m, Q
/Pn
om
Pinverter: PHIL & simulation
Qinverter simulation
Qinverter PHIL
DG and OLTC interactions: Q(U)
Voltage drop at the HV network
Similarly, the oscillations are not shown at the pure-digital
simulation
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0.95
0.96
0.97
0.98
0.99
1
1.01
1.02
1.03
1.04
time (s)
Vo
lta
ge
p.u
.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15-5
-4
-3
-2
-1
0
1
time (s)
tap
po
sitio
n tap PHIL
tap simulation
Vsecondary simulation Vsecondary PHIL
Vinverter simulation Vinverter PHIL
Q(U) controller instability
Closed-loop. Internal delays in the controller (e.g., signal
processing, average filtering of the voltage measurement,
delay in communication between processors) of the
commercial inverter are unknown. Could not be represented
in pure simulation.
PHIL was able to show the true behaviour.
Q(U)
Inverterdynamics
Grid
Measurement
QRef QInvURef U
Delay e-TsURMS
Conclusions
PHIL is a valuable tool that merges component testing
and power system simulation: Power system testing
New interfacing method was proposed
PHIL could reveal interactions of OLTC and DG that
were not visible in pure-digital simulation
Potential problems of OLTC and DG were identified
Thank you for your attention
P. Kotsampopoulos, F. Lehfuss, G. Lauss, B. Bletterie, N. Hatziargyriou,
"The limitations of digital simulation and the advantages of PHIL
testing in studying Distributed Generation provision of ancillary
services", Accepted for publication at the IEEE Transactions on Industrial
Electronics
www.smartrue.gr www.ait.ac.at