Professional Documents
Culture Documents
SUMMARY within the wind farms are connected directly to the DNOs
33kV distribution systems. Two types of turbine transformer
A typical wind farm consists of several wind turbines, each have been studied for the 2x10-turbine site; a standard design
operating at less than 1kV and connected to the MV grid transformer and an innovative and specially designed wind-
system via a turbine transformer. When the turbine turbine transformer. The specially designed transformer has a
transformers are energised, the grid system will see a significantly lower magnetising inrush magnitude than the
transient phenomenon called inrush current and voltage standard design transformer. The turbine transformers with
dips/fluctuations on the local grid system will result. Severe low magnetising inrush magnitude were proposed for the 17-
voltage dips can cause fluctuations of certain types of lights turbine site.
resulting in customer complaints. System studies of voltage
dips caused by energisation of turbine transformers for three Site measurement data were obtained for the operational wind
proposed MV wind farm sites connected to 33kV distribution farm site for verification purposes to compare with the results
networks have been undertaken using PSCAD software and obtained with PSCAD models.
are described in this paper. Site measurement data was
obtained for one of the three sites for verification purposes. 2x10-TURBINE SITE
The results of the system studies, recommended mitigation
measures to overcome any unacceptable voltage dips and This wind farm consists of 20 wind turbines, each rated at
energisation switching sequence for each wind farm site are 1.3MW. Each turbine is connected to the 33kV collection
contained in the paper. network via a turbine transformer, each rated at 1.5MVA,
8.3% leakage impedance. Two types of turbine transformers
Key words: Wind Farms, Turbine Transformers, Inrush were proposed for the project. One is a standard design
Current, MV Grid System, Energisation, Voltage transformer whose maximum peak inrush current is 7.35pu of
Dips/Fluctuations, Point of Connection (POC). rated current. The inrush current is damped to rated current
within 1.6 seconds after energisation. The other is a special
INTRODUCTION design turbine transformer which has a significantly lower
magnetising inrush magnitude than the standard design
As MV wind farms become more prevalent, utilities are transformer. The maximum peak inrush current is 4.04pu of
becoming more concerned about their impact on power rated current, only 55% of that of the standard design
systems. One effect of wind farm generation on a weak transformer, and the inrush current is damped to rated current
distribution system is voltage fluctuation upon energisation of within 1.0 second.
the turbine transformers and cables.
The wind farm substation consists of two POC busbars
A typical wind farm consists of several wind turbines, each running split, each having 10 turbines connected, and no
operating at less than 1kV and connected to the MV (10kV to 33kV interconnections between these two 10-turbine circuits.
150kV, 33kV connection is mainly considered in this paper) The wind farm has two separate 33kV grid connections via
grid system via a turbine transformer. When the turbine these two busbars. Each of the two grid connections is
transformers are energised, the grid system will see inrush independently supplied and controlled. Thus, this wind farm
current and voltage dips/fluctuations on the local grid system. is referred to as a 2x10-turbine site.
When many turbine transformers are energised
simultaneously these voltage dips may be severe. Severe The local distribution network, the wind farm collection
voltage dips can cause the end users equipment to operate system and the wind farm substation were modelled and
inefficiently, e.g. fluctuations of certain types of lights. studied using the proprietary software PSCAD. The voltage
Thus, the voltage fluctuation at the point of connection (POC) dips due to the energisation of the wind farm 33kV collection
and local grid system may be required to be as low as 3% system, i.e. turbine transformers and cables, are assessed in
depending on the interpretation of requirements such as the compliance with the British Engineering Recommendation
British Engineering Recommendation P28 [1], or other P28 assuming the lower limit of 3%. Generally, the
National requirements used by the Distribution Network maximum voltage dip occurs at the instant of energisation,
Operator (DNO) concerned. referred to as voltage dip measured at 0ms in this paper. It is
recommended in P28 that voltage dips due to direct-on-line
System studies of voltage dips caused by energisation of starting of induction motors are measured at 30ms after
turbine transformers for two proposed wind farm sites (a energisation to assess the voltage changes due to motor
2x10-turbine site and a 17-turbine site) and an operational starting. Since energisation of turbine transformers is a
wind farm have been undertaken using PSCAD software [2, similar transient phenomenon to the direct-on-line starting of
3] and described in this paper. Each wind turbine operates at induction motors, the voltage dips due to energisation of
690V and is connected to a 33kV collection system via a turbine transformers measured at 30ms are used to compare
33/0.69kV turbine transformer. The 33kV collection systems with the required limit for this site as well as for the other two
sites in the paper.
CIRED2005
Session No 2
CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005
The PSCAD multiple run facility was used to vary the system can be reduced within 3% by an 80 PIR inserted for
switching angle for point-on-wave to search for the worst 10ms, added to each of the two main circuit breakers. No
voltage dips. The results of studies for the 2x10-turbine site additional switchgear is required within the wind farm
are given in Figures 1 (standard design transformers) and 2 collection system to facilitate sequential energisation.
(special design transformers). Voltage dips at three points of Although PIR switchgear is currently not generally available
the system are shown in the figures, the points being POC 1, at 33kV, it is believed that PIR technology is feasible and
POC 2, and an 11kV primary substation which is located would be of a design similar to neutral earthing resistors
downstream of (i.e. fed from) POC 2 and referred to as Grid commonly used on distribution systems. Such resistances are
point_11kV. Voltage dips measured at 0ms (i.e. peak values) rated at well over 1kA for as long as 30 seconds. Switchgear
are shown in light lines whereas the voltage dips measured at used to switch in and out PIRs would be similar to typical
30ms after the peak are highlighted in heavy lines. distribution switchgear and, in fact, may be cheaper as it
would not need to break fault current.
15 POC 1_0ms
POC 2_0ms
Grid point_11kV_0ms 17-TURBINE SITE
12 POC 1_30ms
POC 2_30ms
This wind farm consists of 17 wind turbines, each rated at
Voltage dip (%)
Grid point_11kV_30ms
9
850kW. Each turbine is connected to the 33kV collection
6
network via a turbine transformer, each rated at 1.0MVA,
5.9% leakage impedance. The turbine transformers with low
3 magnetising inrush magnitude were proposed for the wind
farm. The maximum peak inrush current is 4.63pu of rated
0 current. The inrush current is damped to rated current within
0 1 2 3 4 5 6 7 8 9 10 1.0 second.
Number of transformers energised simultaneously
Figure 1. Voltage dips varying with the number of standard design This wind farm substation consists of one busbar and is
transformers energised simultaneously (2x10-turbine site) proposed to be looped into an existing 33kV circuit which
runs between two existing substations. These two existing
15
POC 1_0ms substations are fed from the same 132kV substation. Under
POC 2_0ms normal operating conditions, the 33kV circuit is fed from both
12 Grid point_11kV_0ms
POC 1_30ms ends.
Voltage dip (%)
POC 2_30ms
9 Grid point_11kV_30ms
The voltage dips due to the energisation of turbine
6
transformers and cables of this 17-turbine site are assessed in
compliance with the British Engineering Recommendation
3 P28, i.e. permitted limit of voltage step changes is 3%.
CIRED2005
Session No 2
CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005
Figure 4 (a) shows the 3-ph wind farm energisation currents simultaneously, voltage dips at POC and grid system can be
when the first 8 transformers are energised simultaneously. reduced within 3% by a 40 PIR inserted for 10ms. The
Figure 4 (b) shows the 3-ph energisation currents when 6 wind farm can be energised under both normal operating
transformers are energised simultaneously with the first 8 conditions and minimum fault level conditions with the 33kV
initially energised transformers. These figures show the grid circuit single-end fed.
effects of initially energised transformers on the decay of the
energisation inrush current. The energisation inrush current is VERIFICATION OF PSCAD MODEL
more rapidly damped when there are no initially energised
transformers (Figure 4 (a)) than with initially energised Verification of the PSCAD model is based on measured data
transformers (Figure 4 (b)). The slow decay of the inrush from an existing wind farm having a comparable number of
current with initially energised transformers prolongs the turbines to the two proposed projects. Each turbine is
voltage dip, as indicated in Figure 3. The voltage dips connected to the 33kV collection network via a turbine
measured at 0ms increase regularly with the increase in transformer, each rated at 1.5MVA, 5.9% leakage impedance.
number of transformers energised, but the voltage dips Standard design transformers were used for the wind farm.
measured at 30ms, when energising 6 transformers with 8 The maximum peak inrush current is 7.35pu of rated current
initially energised transformers, are equal to these when and the inrush current is damped to rated current within 1.6
energising 8 transformers without initially energised seconds. The wind farm substation consists of one busbar
transformers. and is directly connected to a 33kV substation via a dedicated
overhead line
Main : Graphs
I_Windfarm_A I_Windfarm_B I_Windfarm_C This wind farm has been commissioned and is in operation;
0.80 site measurement data has been obtained. Figures 5 (a) to 5
0.60 (c) are the voltage, current and rms voltage waveforms plotted
from the site measurement data. Figures 6 (a) to (c) are the
0.40 corresponding waveforms obtained by the PSCAD modelling
0.20 study. Table 1 shows a comparison of the voltage dips
0.00 obtained by site measurement with the PSCAD modelling
kA
-0.20 study.
-0.40 Figures 5 and 6 and Table 1 show that the PSCAD modelling
results are very similar to those obtained by site
measurements, which verifies the PSCAD model. The
0.000 0.050 0.100 0.150 0.200 0.250 0.300 voltages obtained with PSCAD modelling contain more
harmonic components within the first 100ms after
Figure 4 (a) Inrush current when energising eight transformers
energisation. Site measured currents decay rather more
(with no initially energised transformers) quickly than those obtained with PSCAD modelling. This
implies higher damping within the actual system than the
Main : Graphs PSCAD model.
I_Windfarm_A I_Windfarm_B I_Windfarm_C The site measured voltage waveforms also indicate that the
0.60
0.50 three phases of the turbine transformers were energised
0.40 non-simultaneously. The PSCAD modelling results shown in
0.30 Figures 6 were obtained when C-phase was closed at voltage
0.20 zero (1800), A-phase closed 1.1ms after C-phase and B-phase
0.10 closed 1.6ms after B-phase. This mimics the site
kA
0.00 measurements.
-0.10
-0.20 Figure 7 and Table 2 are the results of studies on three-phase
-0.30 non-simultaneous energisation. Each point on the black
-0.40
B-phase voltage dip curve in Figure 7 is the maximum
0.000 0.050 0.100 0.150 0.200 0.250 0.300 voltage dip when the B-phase is energised at a certain angle
(X-axis, varying between -900 to 300, 1800 is the 00 reference
for the angle) and the A and C phases are energised between
Figure 4 (b) Inrush current when energising six transformers 0 to 10ms after the B-phase is energised. The corresponding
(with eight initially energised transformers) energisation time for the A and C phases at which the
maximum voltage dip is obtained is referred to as delta (t) ms
The maximum voltage dip measured at POC 30ms after and is plotted as a pink curve in Figure 7. Figure 7 shows that
energisation is 3.9% when energising all 17 transformers the maximum voltage dip occurs when one phase is energised
simultaneously, exceeding the required limit of 3%. Voltage before voltage zero (-230) and the other two phases are
drops at the wind farm substation and the local grid system energised about 5ms later. Table 2 shows that the voltage dip
are within 3% at normal system operating conditions if at the POC is 19.5% when all three phases of all transformers
transformers are energised in the sequence of 8 turbine are energised simultaneously. Then the conclusion can be
transformers energised first, then 6 and finally 3. drawn that the voltage dip increases if the voltage of one
phase is around zero (around -600 to 150) when the system is
The effect on voltage dips of a PIR added to the wind farm energised and the three phases are energised
main circuit breaker has also been studied for this site. The non-simultaneously.
results show that to energise all 17 turbine transformers
CIRED2005
Session No 2
CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005
kV
-10 -10
-20 -20
-30 -30
-40 -40
0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 0.200 0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 0.200
kA
-0.50 -0.50
-1.00 -1.00
-1.50 -1.50
-2.00 -2.00
0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400
p.u.
0.925 0.925
0.900 0.900
0.875 0.875
0.850 0.850
0.825 0.825
0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400
0
TABLE 2 PSCAD modelling results maximum voltage dip -90 -75 -60 -45 -30 -15 0 15 30
Percentage percentage of Voltage dip at B-phase angle (degrees)
of initially POC (%)
transformer energised 3 phases Figure 7. Maximum voltage dips due to non-simultaneous energisation
energised transformers energised 0 ms 30 ms of 3- phase turbine transformers
CIRED2005
Session No 2
CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005
CIRED2005
Session No 2