STUDY ON POWER QUALITY ISSUES IN GRID CONNECTED

WIND FARMS AND THEIR REMEDIAL MEASURES

March 2014

Project No. RD-RD-190-10

A.D Thirumoorthy
Principal Investigator
Tamilnadu energy development Agency, Coimbatore

Dr. C. Chellamuthu
Co-Principal Investigator
RMK Engineering College, Kavaraipettai
 ACKNOWLEDGEMENTS

The Centre for Wind Energy Technology (C-WET) would like to

acknowledge the experts from Tamilnadu Energy Development Agency ,
Coimbatore and RMK Engineering College, Kavaraipettai who have
authored this report. C-WET also thanks the Ministry of New and
Renewable Energy (MNRE) for their financial and technical support for
the research and production of this report.

Major contributors in alphabetical order include :

Chellamuthu,C RMK Engineering College

Thirumoorthy, A.D Tamilnadu Energy Development Agency
 NOTICE

This report was prepared as an account of the work sponsored by

Centre for Wind Energy Technology (C-WET). Neither C-WET nor any of
their employees, makes any warranty, express or implied, or assumes
any legal liability or responsibility for the accuracy, completeness, or
usefulness of any information, apparatus, product, or process disclosed,
or represent that its use would not infringe privately owned rights.
Reference herein to any specific commercial product, process, or service
by trade name, trademark, manufacturer, or otherwise does not
necessarily constitute or imply its endorsement, recommendation, or
favoring by C-WET. The views and opinions of authors expressed herein
do not necessarily state or reflect those of C-WET.

No part of this report shall be either reproduced or otherwise be

used in any form without the written permission of the C-WET and the
proponent.

Available electronically at http://www.cwet.res.in

 CONTENT

CHAPTER TITLE PAGE NO.

NO.
1 Literature review
2 Data collection and analysis of wind farm located at 1
peedampalli substation
2.1 Introduction 5
2.2 Fixed speed Induction generator 5
2.3 Power quality analyzer- Dranetz PX5.8 6
2.4 Power quality analyzer- Fluke Serial No 435 12
2.5 Comparison of data collected from the power 13
quality analyzers
2.6 Conclusion 15
3 Simulation of the peedampalli substation
3.1 Introduction 16
3.2 Simulation model 17
3.3 Layout of Peedampalli substation 19
3.4 Simulation model of the peedampalli substation 21
3.5 Steady state analysis of peedampalli wind farm 23
3.5.1 Simulation result of wind farm for various 23
bus bar voltages
3.5.2 Frequency Variations 25
3.5.3 Simulation of wind electric system for 27
various wind speeds
3.6 Dynamic analysis of grid connected wind farm 28
3.6.1 Short Circuit Analysis 30
3.6.2 Load Variation 33
3.6.3 Flickering 34
 3.6.4 Harmonic load flow 36
3.7 Comparison between results of measurement and 36
simulation
3.7.1 Simulation of voltage sag 37
3.7.2 Transient voltage due to capacitor switching 38
3.7.3 Tripping of the wind generator 40
3.7.4 Swell 42
3.8 Conclusion 44
4 Data collection and analysis of wind farm located at
Pethappampatty wind farm
4.1 Introduction 46
4.2 Power Quality Analyzer – Dranetz BMI 51
4.3 Time Plot (Pethappampatty) 56
4.4 Power Quality Analyzer-Fluke 434 58
4.5 Power Quality Analyzer-Hioki 3390 60
4.6 Conclusion 62
5 Data collection and analysis of wind farm at
chinnaputhur substation
5.1 Introduction 64
5.2 Substation layout 64
5.3 Analysis of recorded events 67
5.3.1 Transients 67
5.3.2 Impulsive transients 69
5.3.3 Sag 70
5.4 Analysis of overall recorded events 71
5.4.1 Voltage time plot 71
5.4.2 Time plot of current 73
5.4.3 Flickering 74
 5.4.4 Frequency 76
5.5 Conclusion 77
6 Modelling and controlling concept of the variable
speed wind turbine with PMSG in DIgSILENT
6.1 General 78
6.2 Simulation tools 79
6.3 Simulation model of individual blk using dsl 80
6.3.1 Modelling of mechanical side block 81
6.3.1.1 Speed measurement 81
6.3.1.2 Synchronous generator model 83
6.3.1.3 Frame 84
6.3.1.4 Composite Model 85
6.3.2 Modelling of frequency converter 85
6.3.2.1 Frequency converter 87
6.3.2.2 Generator side controller 90
6.3.2.3 Grid side controller 93
6.3.2.4 Frame 96
6.3.2.6 Composite Model 97
6.4 Grid connection 98
6.5 Conclusion 100 100
7 Analysis and simulation of E2 wind farm feeder in
chinnaputhur substation
7.1 Introduction 101 101
7.2 Load flow studies 103
7.3 Steady state simulation of the e2 feeder connected 105
to grid
7.3.1 Simulation of E2 feeder system for various 105
bus bar voltages
 7.3.2 Frequency variations 108
7.3.3 Simulation of E2 feeder system for various 110
wind speeds
7.4 Dynamic simulation of the grid connected wind 113
electric system
7.4.1System performance under deterministic 113
wind speeds
7.4.2 Dynamic behavior of the wind turbine 117
driven PMSG under grid faults
7.4.3 Flickering 120
7.5 Comparison of simulation result with measurement 121
7.5.1 Introduction 121
7.5.2 Simulation of voltage sag 121
7.5.3 Tripping of the wind generator 123
7.5.4 Swell 124
7.6 Conclusion 125
8 Remedial measures and suggestions for improving
power quality
8.1 Improvement Techniques 127
8.2 Capacitor application 127
8.3 Re-Conductoring 128
8.4 Bifurcation 128
8.5 Load balancing 129
8.6 Feeder reconfiguration 129
8.7 Fact devices 130
8.8 Implication of poor power quality 130
8.9 Custom power solutions 131
8.10 Advantages of power conditioning 131
8.11Fault ride through 132
8.12 Suggestions 133
References 138
 ABBREVIATION
C Capacitance of DC-Link Capacitor
E Electromotive Force (Induced Voltage)
ids Direct Axis Stator Current
If Excitation Current
IGBT Insulated Gate Bipolar Transistor
iqs Quadrature Axis Stator Current
Is Stator Current
m Number of Phases
MPPT Maximum Power Point Tracking
N Total no Transformer
p Number of Pole Pairs
Peq Equavalent Real Power in p.u
Pgen Generator Active Power
PMSG Permanent Magnet Synchronous Generators
Pref Real Power Reference
Qeq Equavalent Reactive Power in p.u
Qgen Generator Reactive Power
Qref Reactive Power Reference
RMS Root Mean Square
RRF Rotor Reference Frame
Rs Stator Reactance
S Apparent Converter Power
Seq Equavalent Apparent Power in p.u
SFRF Stator Flux Reference Frame
Te Electromagnetic Torque
U DC Link Voltage
Uh Voltage Representing the Main Field
Us Stator Voltage
Vdc ref DC Voltage Reference
Xf Field Winding Reactance
Xh Main Reactance
Xsσ Stator Leakage Reactance
Zeq Equavelent Impedance
ωgen Generator Rotational Speed
δ Load Angle
ΨE DC Excited Flux
ϕ Power Factor
Ωgen Mechanical Generator Speed
∆UDC Allowed Voltage Ripple
ω Electrical Frequency of the Grid
Cp Power Coefficient
Hm Inertia Constant
ids Direct Axis Stator Current
iqs Quadrature Axis Stator Current
Ks Spring Constant
Li Inductance
p Pole
PCC Point of Common Coupling
Ps Stator Real Power
Pst Flicker Emission from the Wind Turbine
Qs Stator Reactive Power
RMS Root Mean Square
Rs Stator Resistance
Vwg(t) Gust Component
Vwr(t) Ramp Component
Vwt(t) Turbulence Component
β Blade Angle
λ Tip Speed Ratio
Ψk Network Impedance Angle
ωs Synchronous Speed
ωm Mechanical Speed
Ψi Flux Linkage
 1

CHAPTER 1
LITERATURE REVIEW
The wind energy sector has grown significantly in India despite ups and
downs. Today, after having surmounted many problems, wind energy development
in India is known to be a very mature industry. In India, wind farms are
concentrated in rural areas, where transmission and distribution grids tend to be
very weak and often lack the capacity to tie-in large amounts of wind power. The
insufficient capacity of the power system can cause large variations in steady state
voltage and power outages. Substations are not always equipped with automatic
voltage regulators, further adding to the complications and instability. Inadequate
grid capacity and equipment affect both the efficiency and operation of the wind
turbines. Thus with the increasing penetration of wind power, there is a need to
study the impact of their interconnection to grid.
In India FSIG (Fixed Speed Induction Generator) is the most commonly
used machine. A few turbines are equipped with DDSG (Direct Drive Synchronous
Generator). So, this project will focus on the modeling of FSIG and DDSG
systems. Analysis will be carried out with detailed and reduced order models.
Complexities of the various parts of a wind electric system model, such as
aerodynamic conversion, drive-train and generator representation will be analyzed.
Machines may be represented with active stall and pitch mechanisms. The inverter
modeling will be taken up for DDSG systems. Finally, a comparison of both the
models for WECS will be done. A conclusion on the option of generator and its
parameters to allow a higher penetration of wind power into the grid may be
provided.
In this work, the steady state behavior of WECS will be studied using load
flow analysis. It is important to study the dynamic performance because the
performance of wind turbines is affected by disturbances in the grid frequency
variations, voltage quality and fluctuations in steady state voltage. Wind power
output is reduced during disturbances simply because turbines go off-line in an
 2

outage. Frequency variations lead to changes in rotor speed, subsequently

impacting performance and output. Voltage imbalances can cause the turbine to
trip off-line and lead to efficiency losses in induction generators. The weak Indian
grid networks can exacerbate these problems, making development difficult for
rural areas with insufficient grid capacity. In this project, the impact of connecting
FSIG and DDSG on the transient stability will be compared.
The impact of wind generation on power systems is no longer negligible if
high penetration levels are going to be achieved. It is very important to boost the
transfer capability of the transmission system while improving the utilization of
transmission assets to accommodate high penetration of wind power on the
network. Hence, this project may lead to certain recommendations to improve the
wind power participation.
Overall, this project will present a preliminary analysis of the impact of high
wind power penetration in the planning and operation of the Indian power system.
A consolidated study of the impact of wind power with the Indian data is not
available so far. The outcome of this project may be in terms of recommendations
for the appropriate integration of the new wind generation foreseen for the Indian
power system.
Over the past few years, several efforts have been made in the research and
development for the study of grid connected Wind Energy Conversion Systems
(WECS). As larger wind electric system are brought into service, it becomes
increasingly important to predict their effect on the utility grid. This requires
development of appropriate models for WECS. Work is being carried out on
mainly two types of generators that are commonly used with commercial wind
turbines. They are fixed-speed with directly grid coupled squirrel-cage induction
generator (FSIG), and variable-speed based on a direct-drive synchronous
generator (DDSG).
One of the problems with wind energy generation is the dependence of the
injected power on the wind speed. The wind speed cannot be predicted, but the
 3

probability of a particular wind speed occurring can be estimated. This can be done
by assuming a Weibull or a Rayleigh probability distribution. Once the wind speed
is known, the power injected into the grid can be calculated by means of the wind
turbine power curve. Therefore, assessment of steady state performance of
electrical networks with WECS is not as simple as with conventional generation
and can be planned from a probabilistic point of view only. Several researchers in
India and abroad have developed steady state models for various types of WECS
and demonstrated the application of these models for both deterministic and
probabilistic load flow analysis.
The fast growth of wind power in India has led to the concern about the
impact of WECS on the dynamics of electric power systems. With the increase in
wind power penetration into the utility grid, the power quality and the dynamic
security of the power system might be affected because of the following reasons:

i. The behavior of WECS may be different from conventional generation.

ii. The wind turbines are connected in clusters in the form of wind farm mainly
into the weak electrical grids and
iii. The present connection requirements of wind farms do not have the
capacity to handle such high penetration levels. Thus the dynamic analysis
and the power quality issues of grid connected WECS becomes very
important.
iv. Various wind farms are identified, to measure the power quality events at the
substation. Three power quality analyzer of Fluke 435, Dranetz PX5.8 and
Hioki are connected at the individual wind turbine and group circuit
breaker.
 4

CHAPTER 2
DATA COLLECTION AND ANALYSIS OF WIND FARM
LOCATED AT PEEDAMPALLI SUBSTATION
2.1 Introduction
The wind farm of fixed speed induction generator is identified at the
Coimbatore district in Tamilnadu for measuring the power quality issues. Two
power quality analyzer Fluke and Dranetz are connected at the individual wind
turbines and at the Group control circuit breaker. The various power quality events
such as sag, swell, impulsive transient, oscillatory transient and interruption are
measured.
2.2 Fixed Speed Induction Generator (FSIG)
TEDA has selected peedampalli substation Coimbatore district for recording
the power quality events with the help of power analyzers according to EN50160
standard. The substation consists of six fixed speed wind turbine units connected
through 11kv line. The schematic diagram of peedampalli wind farms, its location
and distances from the substation are shown in Figure 2.1.
The each wind turbine generates 600KW at 690V. The total capacity of this
wind farm is 3.6MW.The fixed speed squirrel cage Induction generator directly
connected to LV side of transformer rated at 800KVA, 690V/11KV. The HV side
of the transformer is connected to 11KV common feeder .The 11KV peedampalli
substation is connected to 110KV substation through transformer rated at 16MVA,
11KV/110KV
Two data loggers were installed at different location in wind farm substation
to record the Voltage quality or disturbance according to EN50160 European
standard. The Power quality Analyzer manufactured by Dranetz was installed at
group control circuit breaker at 11KVsubstation .Another power quality analyzer
manufactured by Fluke435 was installed at wind station, No: SE859 on the LV side
of 690V/11KV transformer. The period of recording of the events in the system is
from 24.07.2011 to 21.08.2011, (29 days).
 5

Figure 2.1: Layout of peedampalli substation

2.3 Power quality analyzer- Dranetz PX5.8
Dranetz power quality analyzer has been connected at wind farm feeder from
24.7.2011 to 21.8.2011, as the wind flow is at maximum during this period. About
140 events were recorded during this period and 75 events were found without any
repetition and they are taken for further analysis. From the recoded data, 34
impulse transient events, only 1 oscillatory transient event, 26 sag events, 3 swell
events and 2 interruption events were identified. The Figures 2.2 to 2.10 show a
few recorded voltage quality events.
The recorded voltage quality event were
1. Transient
 Impulse transient
 Oscillatory transient
2. Sag
 Short duration instantaneous sag
 Short duration momentary sag
 Short duration temporary sag
 Long duration under voltage
3. Swell
 Short duration instantaneous swell
 Short duration momentary swell
 6

4. Interruption
 Short duration temporary interruption
 Long duration interruption

Figure 2.2: Transient event 57786 on 30-7-11

Figure 2.3: Impulsive transient event 102067 on 2-8-11

 7

Figure 2.4: Oscillatory transient 230652 on 13-8-11

Figure 2.5: Short duration instantaneous sag 57787 on 30-7-11

 8

Figure 2.6: Short duration momentary sag 31291 on 28-7-11

Figure 2.7: Short duration temporary sag event 89953 on 1-8-11

 9

Figure 2.8: Short duration instantaneous swell event 180821 on 7-8-11

Figure 2.9: Long duration interruption event 112188 on 3-8-11

 10

Figure 2.10: Short duration temporary interruption event 303989 on 21-8-11

2.3 Power quality analyzer- Fluke Serial No 435

The second power quality analyzer used was Fluke 435. This was connected
to wind station, No: SE859 on the LV side of the transformer rated 80MVA,
690V/11KV. The Fluke 435 instrument recorded events only on 13 days out of 29
days, the remaining 16 days there were no issues to record in the system and the
system was operating normally on these days. The recorded graphs are shown in
Figure 2.11.
 11

Figure 2.11: Recorded graph for duration of 7/24/11 to 8/21/11

2.4 Comparison of data collected from the power quality analyzers

The power quality analyzer installed at wind farm feeder records the same
data recorded by the power quality analyzer installed at the wind farm feeder.
There is a time lag of data 2 microseconds between the two readings. 26 events of
fluke 435 match with the same 26 events of Dranetz analyzer. The Figures 2.12 and
2.13 shows the transient and short duration momentary sag events.
 12

Figure 2.12: Impulsive transient event 131264 on 4-8-11

Figure 2.13: Short duration momentary sag event 294708 on 20-8-11

 13

2.5 Conclusion
The power quality issues related to grid connected wind farm have been
recorded and the possible causes are identified. Two power quality analyzers of
different manufacturers are used at Peedampalli substation, Coimbatore district
from 24.7.11 to 21.8.11(29 days). The Fluke 435 series analyzer installed at wind
turbine on LV side of transformer and the Dranetz analyzer was installed at
substation. Totally 78 mutually exclusive events such as Transient, Sag, Swell and
Interruptions were recorded. 26 data recorded by the fluke 435 matches with 26
same data of Dranetz in spite of a microsecond delay. 34 Impulse transients of
time duration of nanosecond to microsecond were recorded. 10% to 20% voltage
spikes were due to Switching of capacitance, Switching of isolator and utility fault
clearing. It is observed that system stability was maintained even with these
disturbances. An oscillatory transient is a sudden change in the steady state
condition of the voltage, current or both, at both the positive and negative signal
limits, oscillating at the natural system frequency. Only 1 oscillatory transient has
been identified. Totally 27 events of Sag were recorded. The main reason for sag
was due to sudden increase in the load. It was seen from the recorded graph,
voltage drop varied from 40% to 60% of the nominal value. Totally 3 events of
Swell due to sudden removal of load and sudden change in wind speed were
recorded. Voltage rise was about 10% of nominal value. There were 2 events of
Interruption due to switching of wind turbine and sudden drop in wind speed for
the period ranging from second to minute.
It is observed from the various events, 35 events were transient in nature and
about 27 sag events, totaling to 78. Therefore, dynamic simulation of Peedampalli
substation in DIgSILENT was necessary to study the sag and transient events
 14

CHAPTER 3
SIMULATION OF THE PEEDAMPALLI SUBSTATION
3.1 Introduction
The steady and dynamic state stabilities are analyzed for the grid connected
Peedampalli wind farm located at Coimbatore district .The power quality analyzers
Dranetz and Fluke were connected at this substation to record the various power
quality issues. The general structure of fixed speed wind electric system is shown
in Figure 3.1

Figure 3.1: General structure of fixed speed wind electric system

3.2 Simulation model

The important task is to model the wind farm accurately to predict the
performance of wind electric system. DIgSILENT 14.6 power factory is more
suitable for analyzing the grid connected wind farm. The models available in the
 15

software provide the ability to simulate load flow, RMS fluctuations and transient
events in the same environment. DIgSILENT provides a comprehensive library
models for electrical components in the power system e.g. generators, motors,
power plant controllers, dynamic loads and various passive network elements, such
as lines, transformers, static loads and shunts. In the present study , we used the
grid model with the electrical components of wind turbine and built in
standard component models available in the DIgSILENT library. The models of
the wind speed, aerodynamic model, shaft model, and pitch angle control model of
the wind turbines are implemented in the Dynamic Simulation Language
(DSL).Each individual model are built with individual block definition (blk). The
mathematical equations of the individual model are written in the block definition
in DSL form. The values of parameter variable are entered with help of common
model. A Frame is a type of block definition which includes a number of slots for
integrating the various blocks defined. All the individual blocks such as turbine
blk, shaft model blk, pitch angle controller blk and built in model of Induction
machine (ASM blk) are connected together through the slot with help of Frame
definition. The Figure 3.2 shows the frame definition block. In order to connect
the wind electric system to the grid, the composite model is used. In the composite
model, a frame is selected for linking the individual slots to the variables of the
common model. The Figure 3.3 shows the composite model of wind electric
system. In this software, the simulation of EMT (Electromagnetic Transient), RMS
(Electromechanical Transient), Fault, Optimal load, Reliability and Harmonic load
flow including Flickering can be carried out.
 16

Figure 3.2: Frame definitions to form slots to integrate all the blocks

Figure 3.3: Composite model of fixed speed wind electric system

 17

3.3 Layout of Peedampalli substation

The layout of the active stall controlled peedampalli wind farm is shown in
Figure 3.4. It consists of six wind generator each with capacity 600kW at 690V is
taken for the analysis. The total capacity of this wind farm is 3.6MW.The shaded
part in the layout indicates the wind generator under erection work. Each wind
turbine is connected to an 11 kV bus bar. The induction generators, soft-starters
and the capacitor banks for reactive power compensation and step up transformer
are placed at the foot of wind turbine. The control of active and reactive power is
based on the measured reactive power at the point of common coupling. The wind
turbine controller must be able to adjust the production of wind turbine, according
to the demands imposed by the system operator. In case of normal operating
conditions, the wind turbine has to produce maximum power. In power limited
operation mode, the wind turbine has to limit its production to the rated power for
wind speed greater than rated values.
The voltage generated by the squirrel cage generator is directly connected to
LV side of a transformer rated at 800kVA, 690V/11kV. The HV side of the
transformer is connected to 11kV common feeder using underground cable at
voltage level of 11kV .The 11kV Peedampalli substation is connected to 110kV
substation through a transformer rated at 16MVA. At this location, different wind
farm are also connected to the substation as shown in Figure 3.5.
 18

Figure 3.4 Layout of peedampalli substation

Figure 3.5 Sketch of 11kV /110kV Substation

 19

3.4 Simulation model of the peedampalli substation

With help of power system component in DIgSILENT, the wind farm is
built similar to the substation layout diagram. The Figure 3.6 shows the simulated
load flow data of the peedampalli substation. Each generator generates 0.6MW real
power; total power flow at the110kV feeder is 3.6MW. The Figure 3.7 shows the
load flow analysis of the entire grid connected wind farm and shows the Real and
Reactive power flow in generators 4, 5 and 6. The Figure 3.8 shows the Real and
Reactive power flow produced by the generators 1, 2 and 3.

Figure 3.6 Simulation model of the Peedampalli substation in DIgSILENT

 20

Figure 3.7 Load flow diagram of peedampalli substation for generators 4, 5,

and 6
 21

Figure 3.8 Load flow data for generators 1, 2 and 3

3.5 Steady state analysis of peedampalli wind farm

The steady state stability is analyzed for variation in Voltage and Wind
speed as discussed in sec 3.3.
3.5.1 Simulation result of wind farm for various bus bar voltages
Due to load variation and fault in the power system, the grid bus bar voltages
may increase or decrease. The 110kV grid bus bar voltage is varied in steps. As the
grid voltage is increased, the generated power remains constant and the current
decreases. The generator absorbs more reactive power from the grid. The voltage
of the various bus bars increase above the nominal value as shown in Table 3.1. If
the grid voltage is decreased in steps the generated current increases and reactive
power absorption from the grid decreases as shown in Table 3.2 at rated wind
speed. The performance of generator G1 which is far away from the grid bus bar
and G6 which is near to the grid bus bar are considered for comparison.
 22

Table 3.1 Grid Bus bar Voltage Increased from 110kV to 121 kV

Si. Power system Rating RMS simulation of steady state

No component analysis
1. Generator 1 0.6MW,0.69kV 0.60MW,0.71kV,-0.33MVAR,
-0.33MVAR,0.546kA 0.5 kA
2. Generator 6 0.6MW,0.69kV, 0.60MW,0.71kV,
-0.33MVAR, ,0.546kA -0.33MVAR,0.50kA
3. Feeder 11kV 11kV 12.1kV
4. Feeder 110kV 110kV 121kV
5. Grid Frequency 50Hz 50Hz

Table 3.2 Grid Bus bar Voltage decreased from 110kV to 100 kV
Si. Power system RMS simulation of steady state
No component analysis
1. Generator 1 0.60MW,0.63kV,-0.29MVAR,0.6kA
2. Generator 6 0.60MW,0.63kV,-0.29MVAR,0.6kA
3. Feeder 11kV 10kV
4. Feeder 110kV 100kV
5. Grid Frequency 50Hz

3.5.2 Frequency variations

Frequency variation in the power system depends upon the power
demand. When the generated power and the load demand are equal, the value of
frequency is 50Hz. Under the light load condition, the excess generated power is
absorbed by the external grid. The frequency of the each bus bar remains constant
as shown in Table 3.3.When the load is increased above the generated power, the
 23

external grid supplies the power to compensate extra load. Therefore the bus bar
frequency is remaining constant as shown in Table 3.4. In this software, there is a
facility to increase or decrease the grid frequency. The grid frequency is increased
to 51Hz from 50Hz. It is observed that the generator real power and current is
increased but the generator bus bar voltage remain constant as shown in Table
3.5.As the grid frequency decreased, it is observed that generator real power and
current is decreased but voltage remain constant as shown in Table 3.5a.

Table 3.3 Frequency variation at each bus bar with light load
Si.No Element Name Frequency Value
1 Generator Bus 1 50
2 Generator Bus 2 50
3 Generator Bus 3 50
4 Generator Bus 4 50
5 Generator Bus 5 50
6 Generator Bus 6 50
7 Feeder 11 kV 50
8 Feeder 110 kV 50

Table 3.4 Frequency variation at each bus bar when the load is more than
generation
Si.no Element Name Frequency Value
1 Generator Bus 1 50
2 Generator Bus 2 50
3 Generator Bus 3 50
4 Generator Bus 4 50
5 Generator Bus 5 50
 24

6 Generator Bus 6 50
7 Feeder 11kV 50
8 Feeder 110kV 50

Table 3.5 Increase in grid frequency from 50Hz to 51Hz

Si. Power system RMS simulation of steady state analysis
No component
1. Generator 1 0.66MW,0.69kV,-0.33MVAR,0.6kA,51Hz
2. Generator 6 0.66MW,0.69kV,-0.33MVAR,0.6kA,51Hz
3. Feeder 11kV 11kV
4. Feeder 110kV 110kV
5. Grid Frequency 51Hz

Table 3.5a Decrease in grid frequency from 50Hz to 49Hz

Si. Power system RMS simulation of steady state analysis
No component
1. Generator 1 0.58MW,0.69kV,-0.30MVAR,0.52kA,49Hz
2. Generator 6 0.58MW,0.69kV,-0.30MVAR,0.52kA,49Hz
3. Feeder 11kV 11kV
4. Feeder 110kV 110kV
5. Grid Frequency 49Hz

3.5.3 Simulation of wind electric system for various wind speeds

If the wind velocity decreases below its rated value (12m/s), the generated
power also decreases. The fixed speed wind generator is also known as constant
voltage source, generates constant voltage from rated wind velocity 12m/s to cut in
speed 3.5m/s as shown in Tables from 3.6 to 3.7. As the wind velocity is reduced
gradually, the generated power and current reduce accordingly. The speed and real
power variations depend upon the speed of the wind. When the wind velocity
 25

comes nearer to the cut in speed of 3.5m/s, the induction generator behaves like
an induction motor and starts absorbing the reactive and real power from the grid
as shown in the Table 3.8 (negative sign in real and reactive powers indicate its
absorbs the power ).
Table 3.6 Wind velocity (V=12m/s)
Si. Power system RMS simulation of steady state
No Component analysis
1 Generator 1 0.6MW,-0.3MVAR, 0.526kA
2. Generator 6 0.6MW,-0.3MVAR, 0.526kA
3. Feeder 11kV 11kV
4. Feeder 110kV 110kV

Table 3.7 Wind velocity (V=8m/s)

Si. Power System RMS simulation of steady state
No Component analysis
1 Generator 1 0.35MW,-0.15MVAR, 0.21kA
2. Generator 6 0.35MW,-0.15MVAR, 0.21kA
3. Feeder 11kV 11kV
4. Feeder 110kV 110kV

Table 3.8 Wind velocity (V=3m/s) Motor action

Si. Power System RMS simulation of steady state
No Component analysis
1 Generator 1 -0.15MW,-0.1MVAR, 0.12kA
2. Generator 6 -0.15MW,-0.1MVAR, 0.12kA
3. Feeder 11kV 11kV
4. Feeder 110kV 110kV
 26

3.6 Dynamic analysis of grid connected wind farm

In the dynamic analysis, the wind speed is assumed to vary from cut in
speed to cut out speed (3.5m/s to 20m/s) as shown Figure 3.9. As the wind turbines
are a few Km apart, the wind speed at individual wind turbines are assumed to have
different values in the simulation. The wind generator 1which is a few km away
from grid bus bar and the generator 6 which is nearer to the grid bus bar are taken
for the analysis. Various scenarios such as Transient stability analysis, three phase
fault, and sudden increase of load, load flow harmonics and flickering are
simulated. The electromechanical transient (RMS) simulation is executed for 600
seconds and the various data such as real power, reactive power, current, electrical
and mechanical torques are recorded.
The pitch controlled wind turbine used in this model. When the speed of the
wind is varying, the power generated is equal to or less than that of the rated value
as shown in Figure 3.10.When the speed is below the rated value, the pitch control
mechanism is deactivated. The induction generator absorbs the reactive power from
the grid or from the capacitor banks.

Figure 3.9 Variations of wind speed for different wind turbines

 27

Figure 3.10 Active power generation of different generator

3.6.1 Short circuit analysis

The three-phase symmetrical fault is applied at 11kV bus bar t=5 secs for the
duration of 500ms in the simulation time of 20 seconds. The voltage at 11kV bus
bar drops to zero and other bus bar voltages are decreased drastically. The Figures
3.11 and 3.12 show the active power and the voltage dip in the generators 1 and 6.
The Figure 3.13 shows that the generator current increases by 2 to 5times of the
rated current. Depending on the severity of the fault and location of the fault, the
generator current may increase to a high value. The difference between the
mechanical input and electrical output power causes acceleration in the rotor and
therefore, the rotor speed increases as shown in Figure 3.14
 28

Figure 3.11: Variation of active power, reactive power, voltage and current at
generator 6 during the short circuit

Figure 3.12: Voltage dip at generator bus bar and the active power of
generator 1 during the short circuit
 29

Figure 3.13: Variation in generator 1 current during the short circuit

Figure 3.14: Speed and torque oscillation during the short circuit interval
 30

3.6.2 Load variation

At the 11kV bus bar, an industrial load of 3.6MW is suddenly switch on
at t=300secs for a short span of time. During this interval, there is a sudden dip in
the voltage level at 11kV feeder and generator buses as shown in Figures 3.15 and
3.16.

Figure 3.15: Voltage sag at the bus bars during sudden increase in the load
 31

Figure 3.16: Line voltages of Generator bus bars during sudden increase in
load

3.6.3 Flickering
The IEC 61000-4-15 standard specifies the function and design of apparatus
for the measurement of flicker, termed the "Flicker meter‖. The Power
Factory provides a Flicker meter command for the calculation of the short-term and
long-term flicker according to IEC 61000-4-15.
The continuous and switching operations give different types of flickering.
With the help of harmonic load flow in the DIgSILENT software the long term and
the short term flickering due to continuous and switching operations can be
calculated with the constant wind speed of 12m/s and with SCC equal to 15MVA
for various grid impedance angles (30˚, 50˚, 70˚, 85). The Table 3.9 shows the
flickering data at the various generator bus bar for 30˚ impedance value at wind
speed Vw=12m/s and SSC=15MVA .The value of Pst, and Plt for continuous and
switching operations for 30˚ are plotted and shown in Figures 3.17.
 32

Table 3.9 Flickering data at the bus bar for 30˚ impedance value for
Vw=12m/s and SSC=15MVA
Short-Term Long-Term Short-Term Short-Term
Flicker Flicker Flicker Flicker
Disturbance Disturbance Disturbance Disturbance
Factor Factor Factor Factor
Si. Continuous Continuous Switching Switching
No Element Operation Operation Operation Operation
Generator 1
1. bus bar 0.085264 0.085264 0.223803 0.214897
Generator 2
2. bus bar 0.085264 0.085264 0.223803 0.214896
Generator 3
3. bus bar 0.085264 0.085264 0.223803 0.214896
Generator 4
4. bus bar 0.085264 0.085264 0.223803 0.214896
Generator 5
5. bus bar 0.085263 0.085264 0.223802 0.214896
Generator 6
6. busbar 0.08526 0.085263 0.223799 0.214893
Transformer
7. busbar 1 0.084779 0.08526 0.223793 0.214887

8. Feeder 11kV 0.066404 0.084779 0.222508 0.213653

Feeder 110
9. kV 0.047914 0.066404 0.174282 0.167346
 33

0.25 Variable
Pst continous operation
Plt continous operation
Pst switching operation
Plt switching operation
0.20
Flickering

0.15

0.10

0.05

100 200 300 400 500 600

Time in sec

Figure 3.17: Plot of Pst, and Plt for continuous and switching operations for
30˚
The RMS simulation is executed for minimum of 600sec.The values of Pst and
Plt measured with help flicker meter are 0.084563 and 0.225468 respectively.

3.6.4 Harmonic load flow

With help of harmonic load flow, the harmonic level can be calculated in a
grid connected wind farm. The fixed speed wind farm did not inject any harmonic
into the grid. In this software, harmonic any order can be manually entered into the
load or with help of rectifier circuit. The diode rectifier circuit is connected at the
generator bus bar. The injected harmonic by the rectifier circuit is estimated at the
grid. The measured harmonics at the grid are depicted as shown on Figure 3.18.
 34

Figure 3.18 Harmonics measured at the grid

3.7 Comparison between results of measurement and simulation
The comparison between the results of the simulations and measurements
was done to validate the model developed. The power quality instrument
manufactured by Dranetz was connected at the common breaker of the wind farm.
During this study, power quality issues such as voltage sag, Interruption,
Oscillatory transient and Interruption were recorded.
3.7.1 Simulation of voltage sag
Voltage sag is defined as the decrease (between 10% and 90%) in RMS
voltage at the power frequency for the duration of 0.5 cycles to 1 minute. In the
network design considered, the motor at the 11kV bus bar is not started until eighth
second. The switching on of this motor results in a sag in the voltage on the 11kV
line. The motor on the 11kV bus bar is started on the eighth second and it has
caused a sag in voltage in the 11kV bus bar. The sags are momentary in nature as
shown in the Figure 3.19. The Voltage sag measured at the wind feeder is depicted
in Figure 3.20. The dip in voltage is due to the sudden load variation and the
starting of motor. .
 35

Figure 3.19 Simulated voltage sag in 11kV bus bar at peedampalli substation

Figure 3.20 Measured voltage sag at peedampalli sub –station

 36

3.7.2 Transient voltage due to capacitor switching

The network is considered to be in the normal operation before a capacitor
is switched into the 11kV bus bar. At the eighth second, the capacitor is switched
off. This results in oscillations in the line voltage of the generators 1 and 6 as
shown Figures 3.21 and 3.22. A transient-oscillation is noted. It can be seen that
the waveforms of Figures 3.21 to 3.23 look alike. Therefore, capacitive switching
is considered as one of the cause of the oscillatory transient in the network.

Figure 3.21 Simulated transient line voltage waveform of generator 1

 37

Figure 3.22 Simulated transient phase voltage waveform of generated 6

Figure 3.23 Measured transient voltage waveform of generator in peedampalli

 38

3.7.3 Tripping of the wind generator

The simulation assumes that the generator 1 has been subjected to fault in
the wind electric system between the intervals 3 to 4 seconds. Thus, this generator
is switched out from the network during the concerned interval. At the fourth
second, the fault is cleared and the generator is brought back to the normal
operation. The simulation results for such a situation show that the variation in
phase voltage as given Figures 3.24.
The noteworthy observations are: The line-to-line voltage of generator 1
suddenly starts decaying and becomes zero at some point in time. It is restored
back to the normal operation after the clearance of fault. The Figure 3.25 shows
that the measured voltage interruption wave is very similar to the simulated wave
form in the DIgSILENT.

Figure 3.24: Simulated voltage waveform generators during tripping

 39

Figure 3.25: Measured voltage waveform during tripping

3.7.4 Swell
Voltage swell is an increase (between 10% and 90%) in RMS voltage at a
power frequency for duration from 0.5 cycles to 1 min. In the designed network,
the low voltage loads which are connected to the 11kV bus bar are disconnected.
The voltage thus rises from its nominal value of 1 p.u to1.01 p.u shown in Figures
3.26 to 3.27. The Figure 3.28 shows the measured waveform which is similar to the
simulated one.
 40

Figure 3.26 Waveform of simulated voltage swell at generator busbar

Figure 3.27 Waveform of simulated voltage swell at transformer busbar

 41

Figure 3.28 Waveform of measured voltage swell

3.8 Conclusion
The peedampalli wind farm located at Coimbatore district have been
taken for the power quality studies. The wind farm contains 6 fixed speed wind
turbines each connected to the squirrel cage induction generator which generates
600kW at 690V. The total generated power of the wind farm is 3.6MW. The steady
state and dynamic analysis were performed with the help of RMS simulation in
DIgSILENT. From the load flow analysis, wind farm total generated power and
power demand were identified and the voltage profiles at different feeders are
recorded.

Under the steady state analysis, the wind speed is considered as constant.
The performances of wind farm are observed under the variations in grid voltage
and wind speed. If the wind velocity is decreased below its rated value of 12m/s,
the generated power and current of the generator 1 and 6 are reduced. At 3m/s, the
induction generator acts as a motor absorbing the real power from the grid. The
110kV grid bus bar voltage is varied in steps. As the grid bus bar voltage is
 42

increased, the generated power remains constant and the current decreases. The
generators absorb more reactive power from the grid. If the grid voltage is
decreased, the generated real power again remains constant, current increases and
reactive power absorption from the grid decreases. The system frequency remains
constant with increase or decrease in load demand since the system connected to
external grid which behaves like an infinite bus bar.

Under dynamic analysis, wind speed is assumed to vary from cut in speed to
cut out speed (3.5m/s to 20m/s). As the wind turbines are a few kilometers apart,
the wind speed at individual wind turbines are assumed different values in the
simulation. Various scenarios such as Transient stability analysis, three phases
Fault, sudden increase of load, load flow harmonics and wind variation were
simulated for 600 sec. The wind generator 1 is a few km away from grid bus bar
and the generator 6 which is nearer to the grid bus bar are taken for the analysis.
The wind generator is pitch controlled to generate constant real power for speeds
above the rated value. The three-phase symmetrical fault is applied at 11kV bus bar
t=5 sec for the duration of 500ms for the simulation time of 20 seconds. The
voltage at 11kV bus bar drops to zero and the other bus bar voltages near to the
fault are decreased. At the 11kV bus bar, industrial load was suddenly included for
a short span of time about 300 milliseconds. During this interval, there is a sudden
60% dip in the voltage level at 11kV feeder and generator bus bars. The harmonics
were injected at the generator bus and its effects were identified at different bus
bars with the help of harmonic load flow analysis. The values of P st and Plt for
continuous and switching operation are calculated using Flicker meter for various
impedance angle 30˚, 50˚, 70˚and 85˚ at the point of measurement.
The comparison between the results of the simulations and measurements
was carried out to validate the model developed for the peedampalli substation.
During this study, power quality issues such as voltage sag, interruption, oscillatory
transient and interruption were recorded. The sags are momentary in nature and
 43

they occur due to sudden inclusion of motor at 11kV bus bar. The 30% voltage dip
is observed in the simulated and measured result for 5 cycle duration. The
capacitive switching is considered as one of the causes of the oscillatory transient
in the network. The capacitor is switched off after eight sec, the oscillations in the
line voltage of the generators 1 and 6 are observed. The measured and simulated
waveforms are well matched with each other. The interruption waveform occurs
due to tripping of the generator and the short circuit. The generator 1 is tripped for
a short span of time; the line-to-line voltage of generator 1 suddenly starts decaying
and becomes zero. It is restored back to the normal operation after the closure of
the circuit breaker. The measured voltage during interruption is very similar to the
simulated voltage.
 44

CHAPTER 4
DATA COLLECTION AND ANALYSIS OF WIND FARM
LOCATED AT PETHAPPAMPATTY WIND FARM

4.1 Introduction
The Substation consists of four wind farm feeder namely Pukkulam, SV
Patty, Eluppanagaram and Ponneri connected to the 33kV feeder. The station was
constructed and commissioned by M/s Gamesa during the year of 2010-11. The
wind turbines installed in the wind farm are variable speed doubly fed induction
generators (DFIG). The wind farm layout is shown in Figure 4.1
The Pukkulam feeder, consisting of 12 units of variable speed wind turbine
connected to doubly fed induction generator with rating 600 kW at 11kV is shown
in Figure 4.2.The total capacity of this feeder is 7.2 MW. The DFIG injects the
power into the grid from the stator and the rotor with the help of inverter and
converter. The voltage generated by the DFIG is directly connected to LV side of
the transformer rated at 25MVA at 11/33kV. The HV side of the transformer is
connected to the common 33kV feeder.
SV Patty Feeder consists of 17 units of variable speed wind turbine
connected to DFIG which generates 600kW at 11kV as shown in Figure 4.3. The
total capacity of this wind feeder is 10.2MW. The DFIG is directly connected to
the LV side of the 11/33kV transformer. The HV side of the transformer is
connected to common 33kV feeder.
Eluppanagaram feeder has 18 units of variable speed wind turbine connected
to DFIG as shown in Figure 4.4. The total power capacity of this wind feeder is
10.8MW.The Ponneri feeder consists of 9 units of wind electric system having a
total power capacity of 5.4MW. All the DFIG are connected to the LV side of
11/33 kV transformer as shown in Figure 4.5.
 45

All the four wind farm feeders are connected to a common 33kV feeder. The
power transformers, 33/110 kV and 110/230kV are used to step up the voltage
from 11kV to 230 kV for connecting the wind electric system to the 230kV grid.
The Power Quality analyzers were installed at various locations in the
Pethappampatty substations from 13/04/2012 to 24/05/2012 (43 days).Three
different Power Quality meter was installed: The Fluke 435 meter was installed at
wind station connected to Pukkulam feeder using the current transformers. The
other Power Quality analyzer manufactured by Dranetz was installed at feeder
breaker in Pukkulam substation. The third Power Quality analyzer, HIOKI was
installed at 110kV circuit breaker point.

Figure 4.1: Pethappampatty wind farm layout

 46

Figure 4.2: Layout of wind farm connected to pukkulam feeder

Figure 4.3: Layout of wind farm connected to SV patty feeder

 47

Figure 4.4: Layout of wind electric generators connected to eluppanagaram

feeder
 48

Figure 4.5: Layout of wind electric generators connected to ponneri feeder

4.2 Power Quality Analyzer – Dranetz BMI
Dranetz power quality analyzer was installed at feeder breaker at 33kV
Pukkulam substation from 13/04/2012 to 24/05/2012 (43 days). A set of 522 data
was recorded during this period. A set of 170 data was taken for analysis as the
remaining data were found to be repetitive. Out of 170 data, 88 data are impulsive
transients, 24 data are oscillatory transients, 54 data are transients and remaining 4
data are sag and swell. The Figures 4.6 to 4.13 shows the recorded power quality
events.
 49

Figure 4.6: Transient event 18970 on 24-5-12

Figure 4.7: Transient event 14432 on 18-5-12

 50

Figure 4.8: Impulsive transient event 16599 on 21-5-12

Figure 4.9: Impulsive transient event 16638 on 21-5-12

 51

Figure 4.10: Impulsive transient event 14605 on 19-5-12

Figure 4.11: Oscillatory transient event 12518 on 21-5-12

 52

Figure 4.12: Short duration instantaneous sag event 14431 on 18-5-12

Figure 4.13: Short duration momentary swell event 14582 on 18-5-12

4.3 Time plot (Pethappampatty)

Dranetz Power Quality Analyzer recorded the time plot for the period of
21.05.12 to 13.04.12 (39 days). The time plots measure the RMS Voltage,
Flickering, and Harmonic distortions. The recorded waveform events are shown in
Figures 4.14 to 4.15.
 53

Figure 4.14: Recorded events from 14.5.12 to 24.5.12

 54

Figure 4.15 Recorded events from 13.4.12 to 16.4.12

4.4 Power Quality Analyzer-Fluke 434

The Fluke Power Quality Analyzer was installed in wind station at
Pukkulam wind farm (3 wattmeter connections). From 13/04/2012 to 024/05/2012,
the various power quality events such as dip, swell, frequency distortion, current
waveform and voltage waveform were recorded. The recorded waveforms are
shown from Figures 4.16 to 4.19.
 55

Figure 4.16 Frequency variation plot

Figure 4.17 Current variation plot

 56

Figure 4.18 Voltage variation plot

Figure 4.19 Recorded voltage plot on 20/04/12

4.5 Power Quality Analyzer-Hioki 3390
The Power Quality Analyzer manufactured by HIOKI was installed at group
control breaker point at 110kV common feeder of Pethappampatty substation with
help of step down current transformer. Various power quality issues were recorded
such as Voltage, Flickering, and Harmonic in time plot as shown in Figures 4.20 to
4.22.
 57

Figure 4.20: Voltage and Current waveform

Figure 4.21: Time plot – Voltage waveform

 58

Figure 4.22: IEC Flicker variation

4.6 Conclusion
The power quality issues related to grid connected wind farm have been
recorded and the possible causes are identified. Three power quality analyzers of
different manufacturers are used at Pethappampatty Substation, Coimbatore district
from 04/13/2012 to 05/24/2012 (43 days). The Power Quality Analyzer, Fluke 435
was installed at individual wind turbine of Pukkulam wind farm with the help of
current transformer (3 wattmeter connections). The other Power Quality Analyzer,
Dranetz was installed at feeder breaker at Pukkulam substation with the help of
step down current transformer. The other Power Quality Analyzer, HIOKI was
installed at 110 group control breaker point with help of step down current
transformer. About 522 data were recorded during this period and 324 events were
found to be repetitive .The remaining 178 events were taken for analysis. Out of
178, about 96 events were Impulsive Transients. About 25 events were Oscillatory
Transients, and 53 events were Transients .The remaining 4 events were Sag and
Swell. About 96 Impulse transients of time duration of nanosecond to microsecond
were recorded. Switching of capacitance, Switching of isolator and Utility fault
clearing are the causes for 10% to 20% of voltage spikes. It is observed that the
system stability was maintained even with these disturbances. An oscillatory
 59

transient is a sudden change in the steady state condition of the voltage, current or
both, at both the positive and negative signal limits, oscillating at the natural
system frequency. Only 25 oscillatory transient has been identified. Totally 2
events of Sag and Swell were recorded. The main reason for sag was the sudden
increase in the load. It is seen from the recorded graph, voltage drop varied from
40% to 60% of the nominal value. Totally 2 events of Swell due to sudden removal
of load and sudden change in wind speed were recorded. Voltage rise was about
10% of nominal value. Two events of Interruption occurred at wind generator
station due switching of wind turbine and sudden drop in wind speed for the period
ranging from second to minute.
 60

CHAPTER 5
DATA COLLECTION AND ANALYSIS OF WIND FARM AT
CHINNAPUTHUR SUBSTATION
5.1 Introduction
The Chinnaputhur substation was constructed and commissioned by Enercon
Pvt Limited in the year 2008 and is located at Dharapuram district. The wind
turbines installed in this wind farm are variable speed Permanent Magnet
Synchronous Generators (PMSG).
5.2 Substation layout
The layout of the substation of Chinnaputhur is shown in Figure 5.1. It is
seen that the substation has fourteen wind farm feeders (E1- E14) of different
rating that are connected to a 22kV bus bar. Each turbine of the wind farm is
capable of generating 0.85MW at 0.44kV and is connected to a transformer of
rating 0.950MVA, 0.44kV/22kV. This connection is realized with a converter and
an inverter control system.
The HT side of the unit transformer of the wind generator is connected to the
22kV bus bar. This 22kV bus bar is connected to 110kV bus bar through the two
identical transformers capacity 50MVA, 22/110kV.The capacity of all the 14 wind
power feeders (E1- E14) is given in the Table 5.1. The total power generation of
the substation is 123.0MW (Table 5.1).
 61

Figure 5.1: Substation layout of Chinnaputhur

Table 5.1: Feeder capacity of the wind farm (E1-E14)

Si.No Name of Location of Capacity of No of wind Total
the feeder equivalent individual turbine capacity
generator from wind MW
SS (KM) turbine MW
1. E1 8.25 0.85 12 10.2
2. E2 12.25 0.85 11 9.35
3. E3 9 0.85 10 8.5
 62

4. E4 6.6 0.85 10 8.5

5. E5 9.30 0.85 14 12
6. E6 7.40 0.85 10 8.5
7. E7 11.80 0.85 11 9.35
8. E8 7.75 0.85 7 5.95
9. E9 6 0.85 10 8.5
10. E10 14 0.85 12 10.2
11. E11 8.2 0.85 9 7.65
12. E12 6.7 0.85 7 5.95
13. 5.5 0.85 10 8.5
E13
0.30 5 1.5
14. E14 8 0.85 10 8.5
Total 123.00MW

In order to analyze the power quality disturbances in the grid connected wind
farm, two power quality analyzers namely Fluke 435 and Dranetz meters were
installed in the substation from 16/07/13 to 24/08/13 (40 days).The Dranetz was
installed at the 22kV feeder E2 and fluke 435 meter was installed at the group
control breaker of the 110kV bus bar. The specification as well as the connection
details of the both the power quality analyzers were clearly explained in the
previous reports 2 and 3. The various power quality events such as sag, swell,
transient, interruption, flickering and harmonics were found to occur during the
measurement period.
5.3 Analysis of recorded events
Analysis is carried out for the data recorded by the Power Quality
measurement instruments. The events recorded are classified as transients, sag,
swell and interruption
 63

5.3.1 Transients
The various transient events recorded are shown in Figures 5.3- 5.4.The 15%
of dip in the voltage occurs during the transient period for the duration of 0.01ms..
The transient occurs due to the switching operation of isolator switch and
generator.

Figure 5.3: Transient Event 9792 on 31/7/13

 64

Figure 5.4: Transient Event 9799 on 31.7.13

5.3.2 Impulsive transients
The ten Impulsive transient events recorded are shown in Figures 5.5- 5.6.
The impulse transient occurs for the duration less than 50usecs due to switching of
capacitor and isolator switch.

Figure 5.5: Impulsive Transient Event 9793 on 31.7.13

 65

Figure 5.6: Impulsive Transient Event 9794 on 31.7.13

5.3.3 Sag
The six sag events recorded is shown in Figure 5.7. The instantaneous sag
occurs for the duration of two to four cycles due to the symmetrical or
unsymmetrical fault at the substation level.

Figure 5.7: Short duration instantaneous sag event 9798 on 31.7.13

 66

5.3.4. Swell
The only swell event recorded is shown in Figure.5.8. The swell may occur
due sudden removal of load for the duration of 0.09 sec.

Figure 5.8: Instantaneous Swell event 03 on 20.7.13

5.4 Analysis of overall recorded events

The overall power quality events are generated using the Dranetz software.
The report specifies the variation of the magnitudes of voltage for events such as
Transients, Sag and Swell. The variation in magnitude of the current, flickering and
harmonics are also given.
5.4.1 Voltage time plot
The Figures 5.9 and 5.10 show the variation of voltage magnitude for the
observation period as specified according to EN50160 standard. The nominal
voltage is identified as 62V.The dotted red line above and below the nominal
 67

voltage specifies the upper voltage and lower voltage limits whose values are 67V
and 55V respectively.
During forty days of observation, the magnitude of voltage decreases below
the lower limit for six times and increases above the upper limit only once. The
drop in voltage below the lower voltage limit is called as Sag and the increase in
the voltage above the upper limit is called as Swell. As per the compliance report
EN50160 standard, the sags have sustained for the duration of 10 to 100
milliseconds and a swell have sustained for the duration of 10 milliseconds. The
occurrence of sag is due to the varying power demand and a fault in the substation
and swell may occur due to the removal of load at the distribution level.
There was no interruption during the period of study. Nineteen transient
events were observed which might have been caused due to the switching of
capacitor bank or isolator.

Volts

65

64

63

62

61

60

A Vrms

5% 95% 99%
C re a ted with Dr a nVie w 6 .8.1

Figure 5.9: Bar chart of phase voltage, A

 68

Figure 5.10: Time plot of phase A voltage magnitude

5.4.2 Time plot of current

The Figure 5.11 shows the variation of current magnitude during the
observation period in compliance with EN50160 standard. The current value rises
above 0.75A four times during the period of study and this increase is predicted
due to the occurrence of sag.
 69

Figure 5.11: Variation of Current Magnitude

5.4.3 Flickering
The low frequency fluctuation in voltage is referred as flickering. Short term
flickering and long term flickering may occur due to the continuous and switching
operations in the substation. The acceptable fluctuation limits Pst and Plt is less than
one. The Figure 5.12 shows the flickering events. The Pst and Plt values increase
above one for six times during the observation period.
 70

Figure 5.12: Variation of flickering values

5.4.4 Frequency
The Figure 5.13 shows the bar chart which depicts the consistency of
frequency during the period of study. It was found that 99% of the duration the
frequency is maintained below 50.4Hz and 5% of the duration the frequency is
below 49.8Hz which may be due to increase in demand.

Figure 5.13: Bar diagram of frequency variation

 71

5.5 Conclusion
The various power quality events of variable speed permanent magnet
synchronous generator are analyzed. The Chinnaputhur substation located at
Coimbatore district consist of variable speed wind turbines. It has fourteen (E1-
E14) wind farms feeders of different capacity which are connected at 22kV bus
bar. To analyze the power quality disturbances in grid connected wind farm, two
power quality analyzers namely Fluke 435 and Dranetz meters were installed from
16/07/13 to 24/08/13 (40 days).The meter Dranetz was installed at the feeder E2 of
wind farm and fluke 435 was installed at the group control breaker of the 110kV
bus bar. The various power quality events such as Sag, Swell, Transient and
interruption and Flickering were recorded according to EN51620 standard. Totally
22 events were recorded out of which 6 events were sag, 1 event was swell and
the remaining events were transient events. The flickering occurs due to
continuous and switching operation of the wind turbines. The recorded flickering
graph shows the Pst, Plt values are less than 1 for most of the time and it exceeded
the value unity almost six times during the measuring period. Therefore, variable
speed generator produces less flickering. From the recorded report, the variable
speed generator has less number of power quality events than the fixed speed
generator.
 72

CHAPTER 6
MODELLING AND CONTROLING CONCEPT OF THE
VARIABLE SPEED WIND TURBINE WITH PMSG IN
DIgSILENT

6.1 General
A comprehensive dynamic simulation model of the PMSG wind turbine is
implemented in the power system simulation software DIgSILENT Power Factory
and a control strategy for the entire turbine system is developed. The general
structure of the variable speed wind turbine type with full-scale converter is shown
in Figure 6.1.

Figure 6.1: Variable speed multi-pole PMSG wind electric system

6.2 Simulation tools

The models of the wind speed, the mechanics, aerodynamics and the
control systems of the wind electric system are written in the dynamic simulation
 73

language (DSL) of the DIgSILENT. The DSL makes it possible for the users to
create their own blocks either as modifications to the existing models or as
completely new models. These new models can be collected into a library, which
can be easily used further in the modeling of other wind farms and wind turbines.
The DIgSILENT provides a comprehensive library of models of electrical
components in power systems. The library includes models of generators, motors,
power electronic controllers, dynamic loads and various passive network elements
such as lines, transformers, static loads and shunts. Therefore, in the modeling, the
grid model and the electrical components of the wind turbine model are taken from
the standard components in the existing library.
The Figure 6.2 shows the modeling scheme of the PMSG wind turbine and
its control concept. The whole model includes the models for both the aero
dynamical and mechanical system as well as for the electrical system and its
control structure. The built-in models for all the electrical components
(generator model, converter models, transformer model, grid model) are
provided in the DIgSILENT library and they are marked by the yellow blocks.
The models for all mechanical components (wind, aerodynamics, drive train
model) and for the total control system developed by the user are marked in
Figure 6.2 by the grey blocks.
The frequency converter control of the PMSG is divided into two types: a
control for the grid side converter and a control for the generator side converter.
The frequency converter control is coordinated with the pitch angle control of the
rotor blades.
 74

Figure 6.2: Modelling scheme and control concept of the variable speed wind
turbine with PMSG
Vw- Wind Velocity Us- Stator voltage of the generator
Ωrot-Angular velocity of the Turbine rad/sec Trot-Wind Turbine Torque
Ωgen- Angular velocity of the Generator rad/sec Tgen-Generator Torque
Pgrid – Real power at the Grid side
Qgrid- Reactive power at the Grid side

The DIgSILENT has the ability to simulate load flow, RMS fluctuations
and transient events in the same software environment. It provides models on
different levels of the detailing. It combines the models of electromagnetic
transient simulations of the instantaneous values with the models of the
electromechanical simulations of RMS values. This makes the models useful for
the studies of grid fault, the power quality and control issues.
 75

The developed model will be able to support the analysis of the interaction
between the mechanical structure of the wind turbine and the electrical grid both
during normal operation of the wind turbine and during the transient grid fault
events. The dynamic interaction between a wind turbine/wind farm and a power
system are simulated by using RMS simulation in the DIgSILENT. In the recent
years, the trend has been moved from the installations of a few wind turbines to the
large wind farms with several hundred MW of capacity. This situation means that
future wind farms must be able to replace conventional power stations, and thus be
active controllable elements in the power supply network. In other words, wind
farms must develop power plant characteristics. The TNEB utility which is
responsible for the transmission systems in Tamil Nadu has issued requirements
that focus on the influence of wind farms on the grid stability and the power quality
and on the control capabilities of the wind farms.. The E2 wind power feeder of
Chinnaputhur substation is simulated to study the various power quality issues and
their causes.
6.3 Simulation model of individual blk using dsl
In this simulation model, mechanical side control and converter side control
of variable speed PMSG are modeled independently. The mechanical side control
consists of wind speed blk, shaft blk, pitch angle control blk, speed measurement
blk and machine blk and are connected together to generate the electrical energy.
The generated electrical energy is controlled with help of generator side converter.
The converter side control consists of converter blk, inverter blk, maximum power
point tracking blk, grid side controller blk and generator side controller and they
are connected together. The converter system controllers generate a constant
stator voltage, real power and necessary reactive power at all the power system
condition such as load variation and fault. In this section, modeling of individual
blocks in the DSL and their interconnection together are explained.
 76

6.3.1 Modeling of mechanical side block

The mechanical side control consist of wind speed blk, shaft blk, pitch angle
control blk, speed measurement blk and synchronous machine blk. The model of
wind speed blk, shaft blk, and pitch angle blk were explained in the report 4
already submitted. With the help of frame model, the individually developed blk
are integrated together. The composite model is used connect the wind electric
system to the grid.
6.3.1.1 Speed measurement
The speed measurement block definition is included in this model to
measure the speed of the generator. This speed variable is used as the input data for
the maximum power point tracking control. In this block the variable, speed_gen
represents the rotor speed of generator and hpi represents the rotor angle and these
variables are the input signals to the speed measurement blk. The signal, speed is
the output variable as shown in Figure 6.3.The parameter variable fnom, nominal
frequency of the generator, is also entered.

Figure 6.3: Speed Measurement block for defining the variable

The mathematical equations developed for the speed measurement block are
entered in the block definition as shown in Figure 6.4.
 77

Figure 6.4: Speed Measurement block for entering the equation

6.3.1.2 Synchronous Generator Model

The synchronous generator namely, Sym Elm blk, available in library has
been selected for the wind electric generator. The synchronous generator is
modeled in the DIgSILENT using the models described in the literature (Kundur
P., 1994). The synchronous generator parameter data such as steady state reactance,
sub transient reactance and transient reactance values are entered as shown in
Figure 6.5. The acceleration time constant value is used to control the rotor
damping. In this SymElm blk (synchronous machine block), Pt, turbine power and
ve , the excitation voltage are given as the input and xspeed ,the speed of machine
in p.u and hpi,the rotor angle, are taken as the output.
 78

Figure 6.5: Synchronous machine windows for setting the generator

parameter
6.3.1.3 Frame
A Frame is a type of block definition which includes a number of slots for
integrating the various blocks defined. The variable speed wind electric system is
developed with generator PMG and PWM converter frames. The PWM converter
frame connects the power electronic component blk, grid side and generator side
controller blks .The Generator PMG frame connects the built in blocks from the
wind source to generator side. All the individual blocks such as Turbine blk, Shaft
model blk, Pitch angle controller blk, built in model of synchronous Machine (Sym
blk) are connected together as shown in Figure 6.6.When a slot in the frame
definition is selected the project block gets opened where in the number of
functional block already defined are displayed.
 79

Figure 6.6: Frame definition to form slots to integrate the Generator PMG
blocks

6.3.1.4 Composite model

In the frame definition, the wind electric generator system is modeled to
integrate all the blocks that have been discussed in the previous section. In order to
connect the wind electric system to the grid, the composite model is used. In the
composite model a frame is selected for linking the individual slots to the variables
of the common model. The various slots and their linking to the common block
variables are shown in Figure 6.7.
 80

Figure 6.7: Composite model of Generator PMG

6.3.2 Modeling of frequency converter
The PWM converter frame consists of converter blk, inverter blk,
maximum power point tracking blk, grid side controller blk and generator side
controller. The individual built in blk are integrated together with help of frame.
The Figure 6.8 shows the overall gird side and generator side controller block.
 81

Figure 6.8: Grid and Generator side controller block in DIgSILENT

6.3.2.1 Frequency converter
The power converters used in wind turbine application are usually
realized by self commutated pulse-width modulated circuits. The voltage source
AC/DC converter equivalent circuit is shown in Figure 6.9.
 82

Figure 6.9: Generic PWM converter model

These circuits are built with six switches with turn-off capability and six
anti parallel diodes. The switches are typically realized by IGBTs because they
allow for higher switching frequencies than the classical devices. The general
model of the PWM converter that usually operated as a voltage source converter is
illustrated in Figure 6.10.

Figure 6.10: PWM converter –General model

 83

In load flow analysis, the PWM converter model supports several control
conditions.
The supported control conditions are as follows:
 Vac-phi – Specifies magnitude and phase of the AC terminal voltage. It is
typical control mode is for the motor-side converters in variable speed
applications.
 Vdc-phi – Specifies the DC-voltage and the AC-voltage phase. No specific
application.
 PWM-phi – Load flow setup without control. The PWM index and phase are
directly set.
 Vdc-Q – Specifies DC-voltage and reactive power. Typical applications:
STATCOM, UPFC, grid-side converter for doubly-fed induction machine.
 Vac-P - Specifies AC-voltage magnitude and active power. This mode is
equivalent to a PV characteristic of synchronous generators. Typical
applications: Grid-side converter control driven synchronous machines and
VSC.
 P-Q – Specifies P and Q at the AC-side. This control is equivalent to a PQ
characteristic of synchronous machines. Typical applications: Grid-side
control of converter driven synchronous machines, VSC.

Since the model is based on the fundamental frequency approach, the load
flow calculations will assume always a frequency as defined in the external grid.
Then, the voltage angle for each bus bar is calculated with respect to the global
reference. For all the control conditions in the load flow analysis the frequency is
not a controlled variable.
 84

The control variables for the stability model are defined in 4 ways depending on
the applications (DIgSILENT, 2006):
A. Pmr, Pmi– Real and imaginary part of the PWM index. The reference system
in this case is the global reference frame, which is usually defined by a
reference-machine, external network, voltage source or a PWM converter.
This set of inputs must be always used in combination with phase
measurement devices e.g. Phase Locked Loop (PLL) and transformation
between reference frames.
B. Pmd, Pmq, cosref, sinref– This set of input is used in grid connected
applications. The PWM Index-vector is specified with reference to a
reference system, which is defined by cosref and sinref. For example the
outputs from the current controllers are connected to Pmd and Pmq while the
voltage is measured using a PLL and its output gives the cosref and sinref
signals.
C. Pm_in, dphiu – magnitude and phase of the PWM Index. This representation
is equivalent to Pmr, Pmi. The phase of PWM Index dphiu is expressed with
reference to the global reference frame.
D. Pm_in, f0 (50Hz) – Pm _in is the magnitude of the PWM index and f0
permits varying the frequency of the output voltage. The control variable f 0
defines the frequency in p.u., the base value 50Hz. This input pair ―is
especially useful in variable speed-drive applications, in which a PWM
converter is used for driving an induction machine.
6.3.2.2 Generator side controller
The stator voltage of the generator and the voltage of the DC-link with
respect to real power are regulated. The controller is equipped with fast current
loop. The current reference values are derived from the voltage regulators as
indicated in Figure 6.11.
 85

Figure 6.11: Model of generator side converter controller in DIgSILENT

The generator controller blocks are built with block definition to control
the real and reactive power generation. The input signals are Vdc, Vac, Vdc_ref
and Vac_ref, The output signals are id_ref and iq_ref .The parameters are :Tfp,
TfVac, TfVdc, Tp and Kp.The internal variables are: Imax, Imin, deltaV, dVac,
dVdc and id ref as shown in Figure 6.12.The sub blocks of the generator controller
model are MeasFlit blk , PI controller blk, absolute blk as shown in Figure 6.13.
The DC voltage (Vdc) measured at the DC busbar, Vdcref voltage
measured from system damping and Pref measured form look table of MPT block,
the real power (P) is measured at the generator busbar with help of in-built power
measurement blk are connected to the PI controller through the summing point to
track the Vdc. The AC voltage (Vac) is measured at the generator busbar and
Vacref voltage are connected to the PI controller through the summing point to
track the Vac. The outputs of the PI controllers are idref and iqref and are connected
to the converter. The modulation indices Pmd and Pmq are varied and these depend
on the error signals.
 86

Figure 6.12 Definition of variables in Generator side controller

 87

Figure 6.13 Block definition of Generator side controller

 88

6.3.2.3 Grid side controller

The DC voltage and reactive power at the grid terminal are controlled by
grid side controller is shown in Figure 6.14.The block definition are built with the
input signals are Vdc, Qref, Vdc_ref, Vac_ref, P_ref and P. The output signals are
Pmd and Pmq.The parameters are Tfp, TfVac, TfVdc, Tp and Kp, imax, imin.The
internal variables are Imax, Imin, deltaV, dVac, dVdc, id ref, iqref and iq as shown
in Figure 6.15.

Figure 6.14: Model of Grid side converter controller in DIgSILENT

The sub blocks of the grid controller model are MeasFlit blk , PI controller
blk, absolute blk as shown in Figure 6.16.
The reactive power (Q) is measured at the grid bus bar with help of power
measurement .The variables Q and Qref are connected to the PI controller through
the summing point to control the reactive Var. The Vdc is measured at the
capacitance busbar and Vdcref are connected to the PI controller through the
summing point to reduce the steady state error. The output of the PI controllers is
idref and iqref is connected to the rectifier, where the modulation indices Pmd and Pmq
vary with the error signals.
 89

Figure 6.15: Definition of variables in Grid side controller

Figure 6.16: Block definition of Grid side controller

 90

6.3.2.4 Frame
A PWM Frame , connecting all the individual blocks such as Rectifier blk,
grid side controller blk (Vdc-Q), Generator side controller blk (Vac-P) ,MPT blk,
Inverter in-built blk are connected together through the slot in the frame as shown
in Figure 6.17. When a slot in the frame definition is selected, the project block
gets opened where in the number of functional blocks already defined are
displayed. The connection of the individual blk at generator side controlling and
grid side controlling are visualized with the help of PWM frame.

Figure 6.17: PWM Frame definition to form slots to integrate all the blocks

6.3.2.5 Composite Model

In the frame definition, the power electronic in built component and
controller blks are modeled to integrate all the blocks that have been discussed in
the previous section. In order to connect the wind electric system to the grid, the
composite model is used. In the composite model, a frame is selected for linking
 91

the individual slots to the variables of the common model. The various slots and
their linking to the common block variables are shown in Figure 6.18.

Figure 6.18 PWM Composite models

6.4 Grid connection

The DIgSILENT consists of inbuilt power system components like
transformer, cable (power cable, underground, overhead cable), feeder, bus bar,
terminal, current transformer, voltage transformer, power electronics component
(Inverter and Converter) and consumer load. The grid connected wind electric
system is modeled with the help of these components as shown in Figure 6.19. The
wind turbine is connected to the variable speed synchronous generator unit and it is
connected to the grid through the frequency converter. To regulate the variable
power and variable frequency, a shunt capacitor is connected across the converter
and inverter. The filter circuit (LC) is connected at the DC bus bar to remove the
 92

ripple content. The generator power is synchronized to the grid through the step up
transformer and underground cable.

Figure 6.19: Simulation model of variable speed wind electric system with
PMSG
 93

6.5 Conclusion
With a view to getting a better understanding of the variable speed wind
electric system, a generic model was introduced, where the various independent
elements of a wind turbine and their interaction are explained. The individual
blocks such as turbine BlkDef (block), shaft BlkDef, pitch angle control BlkDef,
Grid side controller BlkDef, Generator side controller BlkDef, MPT BlkDef were
designed using the mathematical model in DSL. All the individual blocks were
connected together with help of frame definition block including the built-in model
of SYM BlkDef and Converter BlkDef .The values of the parameter variable in
each block are entered using common model. The composite model (ElmComp)
connects the frame slot and common model. Finally, the simulation model of the
grid connected variable speed wind turbine was built with the help of power system
component such as bus bar, transformer, cable, feeder etc from the software. The
generator side converter controls the stator voltage and optimal power generation
with respect to the speed of the machine. The grid side controller controls the real
power and reactive power at the grid bus bar.
 94

CHAPTER 7
ANALYSIS AND SIMULATION OF E2 WIND FARM FEEDER
IN CHINNAPUTHUR SUBSTATION
7.1 Introduction
The E2 wind farm feeder has eleven individual variable speeds PMSG of
same rating of 0.85MW. The total generated power capacity is 9.45MW. The
generators are located at 8.25KM to 12.25KM away from the substation. The
Figure 7.1 shows the layout of E2 wind farm feeder. The Table 7.1 shows the
distance of generators from the grid 110kV. The E2 feeder layout is modeled with
the help of DIgSILENT software to analyze the various power quality issues. The
generator in the left hand side of the layout is considered as G1 which is far away
from the grid and G11 near to the grid in this simulation. The power quality
analyzer, Dranetz 431 was installed at E2 wind farm feeder. The power quality
events as per EN50160 standard were recorded.
The electromechanical transient (RMS) simulation is performed for 600sec
under steady state and dynamic state conditions. The various power quality events
such as Sag, Swell and Transient are simulated.
Table 7.1 Distance of generators from grid for E2 feeder
Generator G1 G2 G3 G4 G5 G6 G7 G8 G9 G10 G11
Location 12.25 12.5 12.00 11.86 9.68 9.46 9.42 8.9 8.78 8.57 8.23
from the
grid in
KM
 95

Figure 7.1: Substation Layout of E2 feeder

 96

7.2 Load flow studies

The power flow or load flow analysis of the grid connected PMSG helps us
to calculate the Real power generation, Reactive power consumption, Current and
the Voltage values at PCC and different bus bars. This wind farm feeder consists of
eleven individual wind turbines of same rating 0.85MW, 0.44kV connected to a
transformer of rating 0.95MW,0.44/22kV through the converter and inverter
controller. The HT side is connected to 22kV bus bar through the underground
cable of capacity 22kV.The specification of the components of all the wind
turbines are of the same rating. The performance of the individual wind turbine is
shown in Table 7.2.
Table 7.2: Load flow study of E2 feeder

Si. Power system component Load flow measurement

No
1 Generator 0.85MW,0.088Mvar,1.18kA

3 Generators bus bar 0.44kV

4 Generators side converter 0.85MW,0.0Mvar,1.18kA
6 DC bus bar 0.71kV
7 Grid sides converter 0.85MW,0.08Mvar,1.18kA
8 LV bus bar 0.44kV
9 MV bus bar 22kV
The simulation model of the E2 wind farm feeder using DIgSILENT software is show
10 HV bus bar 110kV
 97

Figure 7.2 Simulation model of E2 feeder

7.3 Steady state simulation of theE2 feeder connected to the grid

The steady state analysis of the grid connected wind farm helps us to
analyze the system performance under the constant wind speed and load
conditions. The system behavior is analyzed under the variation in the grid voltage,
frequency and wind speed conditions.
 98

7.3.1 Simulation of E2 feeder system for various bus bar voltages

There are situations where the grid bus bar voltage may increase or decrease
depending upon the load and symmetrical fault. Hence, it becomes significant to
study the system behavior under various bus bar voltages. The grid voltage is
increased by 10% of its nominal value and the performance of G1 which is far
away from the grid bus bar is shown in Table 7.3.The Table 7.4 shows the
performance of the G11 generator which is very near to the grid bus bar HV or
PCC. It is observed that the generators near and away from the grid bus bar are
affected similarly. The generators current of the E2 feeder is decreased when the
grid voltage is increased. The grid voltage is decreased by 10% of its nominal value
and the results are shown in Table 7.5. It is observed that the current increases
when the grid bus bar voltage varies, the generator controller regulates the
generator bus bar voltage and DC bus bar voltage at constant values as per the set
values. The Figures 7.3 and 7.4 show the bar chart of bus bar voltages for the
increase and decrease in the grid voltage.
Table 7.3: Grid bus bar voltage increased from 110kV to 121 kV at wind
velocity of 12m/s
Si. Power system Real power Reactive Voltage Current
No component flow power flow kV kA
MW Mvar
1. Generator G1 0.85 0.08 0.44 1.122
2. Generator side 0.85 0.08 0.44 1.122
converter
3. Grid side converter 0.85 0.12 0.44 1.01
4. Generator bus bar 0.44
5. DC bus bar 0.702
6. LV bus bar 0.44
7. MV bus bar22 24.5
8. HV bus bar 110kV 121
 99

Table 7.4: Grid bus bar voltage increased from 110kV to 121 kV at wind
velocity of 12m/s
Si. Power system Real power Reactive Voltage Current
No component flow power flow kV kA
MW Mvar
1. Generator G11 0.85 0.08 0.44 1.122
2. Generator side 0.85 0.08 0.44 1.122
converter
3. Grid side converter 0.85 0.12 0.44 1.01
4. Generator bus bar 0.44
5. DC bus bar 0.702
6. LV bus bar 0.44
7. MV busbar22 24.5
8. HV bus bar 110kV 121

Figure 7.3: Bar chart for rise in bus bar voltage for the Generator G11 far
away from the grid
 100

Table 7.5: Grid bus bar voltage decreased from 110kV to 99kV
at wind speed of 12m/s

Si. Power system Real power Reactive Voltage Current

No component flow power flow kV kA
MW Mvar
1. Generator G1 and G11 0.85 0.088 0.44 1.122
2. Generator side 0.85 0.088 0.44 1.122
converter
3. Grid side converter 0.85 0.12 0.44 1.32
4. Generator bus bar 0.44
5. DC bus bar 0.702
6. LV bus bar 0.44
7. MV bus bar 22 18.36
8. HV bus bar 110kV 99
9. Grid power 0.85 0.02 1.32

Figure 6.4: Decrease in grid busbar voltage

7.3.2 Frequency variations
The variations in the frequency occur due to the variation in the speed of
the generator. But the speed of generator depends upon the real power demand and
wind speed variation. The generator side converter controller controls the stator
voltage and DC voltage under variable wind speeds. The generator frequency
 101

remains constant for wind speeds above the rated value and decreases gradually
below the rated wind speed from 11m/s to cut-in speed of 5m/s. The grid side
converter controller controls the real and reactive power flow to the grid under the
variable load conditions. Therefore, the system frequency is almost maintained
constant under all the load conditions.
The Table 7.6 shows the constant frequency for the wind speed condition
above the rated value. The Table 7.7 shows the variation in the frequency for the
speeds below the rated wind speed. The generator and reactance bus bar
frequencies are varied but the grid side converter controller maintaining the
constant grid bus bar frequency. The Figure 7.5 shows the bar chart of generator
and grid frequency under change in wind speed.
Table 7.6: Constant frequency above the rated wind speed at 19m/s
 102

Table 7.7: Variation of the frequency below the rated wind speed at 8m/s

Figure 7.5: Variation in generator and grid busbar frequency

7.3.3 Simulation of E2 feeder system for various wind speeds

The Tables 7.8 shows the performance of G1 in the E2 feeder for the wind
speed of 19m/sec. Above the rated wind speed condition, the pitch control
mechanism reduce the power extraction from the wind and maintains the constant
power generation. It is observed that the generator side converter controls
maintains the stator voltage of the generator and the capacitor DC voltage at the
fixed reference value. The performance of the generators G1 and G11 in the E2
feeder are the same.
 103

Table 7.8: Performance for wind velocity, v=19 m/s

Si. Power system Real Reactive Voltage Current
No component power power flow kV kA
flow Mvar
MW
G1 G11 G1 G11 G1 G11 G1 G11
1. Generator 0.85 0.85 0.088 0.086 0.44 0.44 1.122 1.12
2. Generator side 0.85 0.85 0.0 0.0 0.44 0.44 1.122 1.12
converter
3. Grid side 0.85 0.85 0.12 0.12 0.44 0.44 1.122 1.13
converter
4. Generator bus bar 0.44 0.44
5. DC bus bar 0.702 0.702
6. LV bus bar 0.44 0.44
7. MV bus bar 22 22 22
8. HV bus bar 110kV 110 110

The Table 7.9 shows the value of G1 in the E2 feeder for the wind speed of
8m/s. The generator real power gradually decreases with respect to change in the
wind speed but the generator stator voltage remains constant. The Figure 7.6 shows
the bar chart of generator real power and bus bar voltage for wind speed below and
rated speed.
 104

Table 7.9: Performance of G1 for wind velocity, v=8 m/s

Si. Power system Real Reactive Voltage Current
No component power power kV kA
flow flow
MW Mvar
1. Generator G1&G11 0.20 0.20 0.47 0.47 0.44 0.44 0.66 0.66
2. PWM Generator side 0.20 0.20 0.0 0.0 0.44 0.44 0.66 0.66
3. PWM Grid side 0.20 0.20 0.56 0.56 0.44 0.44 0.6 0.6
4. Generator bus bar 0.44
5. DC bus bar 1
6. LV bus bar 0.44
7. MV bus bar 22 22
8. HV bus bar 110kV 110
9. Grid power 0.20 0.02 0.6

Figure 7.6: Variation in Generator real power and bus voltage for wind speeds
below and above the rated speed
 105

7.4 Dynamic simulation of the grid connected wind electric system

Sudden large disturbances in the power system such as change in wind
speed, symmetrical fault are considered for the dynamic simulation studies.
As the wind speed is above its rated value, vw = 12m/s, the pitch mechanism
is active and limits the power and the speed to their rated values. For lower wind
speeds below 12 m/s the pitch mechanism is passive and the pitch angle is kept to
its optimal value of zero .Under the variable speed operation, the MPP -tracking
assures optimal energy capture.
7.4.1 System performance under deterministic wind speeds
The typical variable of the turbine system such as pitch angle, generator
speed and active power are considered for the simulation period for varying wind
speed. Under this condition, the response of two generators, G11 (8.29KM to the
Grid) and G1 (12.35KM to the grid) are considered. The Figure 6.7 shows the wind
speed variation, beta, speed of the generators and powers. The green line represents
the performance variations of G11 generator and the red line to that of G1
generator. It is seen clearly that when the wind speed decreases below the rated
wind speed 12m/s, the pitch angle mechanism becomes inactive and the beta value
is maintained zero. The speed and the power of the generator oscillate at every step
change in wind speed due to the system damping. The Figure 7.7 shows clearly
that the performance of both the generators which are near and far away from the
grid are the same as the red line of G1 overlaps the green line of G11 generator.
The Figure 7.8 shows that the DC link voltage and the stator voltage are maintained
constant at the reference value with the help of generator side converter control.
The damping of the system causes oscillations in the DC-link voltage around its
reference value of 0.705kV. The stator voltage is controlled at its rated value.
 106

Figure 7.7: Variation of Pitch angle, Generator speed and Active power of the
G1 and G11 for wind speed varying from 12 m/s down to 5 m/s
 107

Figure 7.8: DC-link voltage and stator voltage for the variation of wind speed
from 12 m/s down to 5 m/s
Another case study has been carried out in which the wind speed is increased
in steps from 12 m/s to 20 m/s. The response of pitch angle, speed and power is
depicted in Figure 6.9. The step variation in the wind speed changes both the pitch
angle and the generator speed. The pitch mechanism reacts slowly compared to the
power controller. It is noticed that the variations in the generator speed occur at
each step change in the wind speed. The response of the pitch angle and the
generator speed has oscillations and the generator power is kept at its rated value of
0.85 MW. The speed responses of generator 1(green line) and generator 11 (red
line) are similar with a small deviation of 0.01p.u as shown Figure 7.9.
The Figure 7.10 shows the response of DC bus voltage with respect to simulation
time. The DC-link voltage is maintained constant at the set value with the help of
generator side controller.
 108

Figure 7.9: Variation in Pitch angle, Generator speed and Active power for
change in wind speed
 109

Figure 7.10: Variation of DC-link voltage for wind speed from 12 m/s to 20
m/s

7.4.2 Dynamic behavior of the wind turbine driven PMSG under grid faults
In order to evaluate the dynamic behavior of the wind turbines driven PMSG
under grid faults, a three-phase grid fault is created at the high voltage grid
terminal. The turbine system and its grid connection are illustrated in Figure 7.11.
 110

Figure 7.11: Schematic diagram G1 and G11 of E2 feeder

A symmetrical three-phase fault is applied at the high voltage terminal at
time 10.00 and it lasts for 10 millisecs. The Figure 7.12 shows the variation in
voltage, real power and reactive power. It is inferred that the simulated fault causes
30 % voltage drop of the rated voltage at the MV bus bar of G1 and G11 with 0.01
p.u deviation. The grid side converter continues to restore the reactive power to its
reference value and the active power drops.
The Figure 7.13 shows the Constant DC bus bar voltage of G1 and G11.
During the fault, the generator side converter control keeps the DC-link voltage
constant. This causes a decrease in the absolute value of the actual d-component of
the converter current as illustrated. Although the DC voltage is controlled to a
constant reference value, a switching transient of the DC-link voltage is observed at
the fault incident. It is also noticed that the inner control loop response is fast to
 111

settles down the DC voltage to its reference value with in 4 milliseconds during the
fault period. After clearing the fault, the DC-link voltage was found to be
oscillating and settles down to the reference value with in millisecond.

Figure 7.12 Variation in voltage, Active Power and Reactive power of G1and
G11 generators
 112

Figure 7.13: Variation in DC bus bar voltage of G1 and G11

7.4.3 Flickering
With the help of harmonic load flow in the DIgSILENT software the long
term and the short term flickering due to continuous and switching operations can
calculated for the constant wind speed and for the various grid impedance angles
.The Table 7.10 shows the flicker value appearing at various bus bars for 50˚
impedance angle.
 113

Table 7.10: Flickering value at the bus bars for 50˚ impedance angle

Pst Plt Pst Plt

Si. Continuous Continuous Switching Switching
no Element Operation Operation Operation Operation
LV bus bar
1. 0.005683 0.006567 0.005126 0.02364
2. MV bus bar 0.005687 0.002565 0.004532 0.04538
3. HV bus bar 0.004754 0.006984 0.002526 0.01235

The simulated flicker values for Pst and Plt are very less when compared to
the threshold value of unity. The RMS simulation is executed for minimum of
600sec and the values of Pst =0.004563 and Plt= 0.025468 are measured with the
help of flicker meter.

7.5 Comparison of simulation result with measurement

7.5.1 Introduction
The comparison between the results of the simulations and measurements
was performed to validate the model used. The power quality instrument
manufactured by Dranetz was connected at the common breaker of the wind farm.
During this study, power quality issues such as voltage sag, Interruption, swell and
oscillatory transient were recorded.
7.5.2 Simulation of voltage sag
Voltage sag is defined as the decrease (between 10% and 90%) in RMS
voltage at the power frequency for the duration of 0.5 cycles to 1 minute. In the
network design considered, the three phase fault is simulated at 110kV bus bar for
10 ms duration. The sags are momentary in nature as shown in the Figure 7.14. The
voltage sag measured at the wind feeder is depicted in Figure 7.15. The dip in the
 114

voltage has occurred due to the fault alone at the grid bus bar but the generator bus
bar voltage and DC bus bar voltage are not affected because of the DC link.

Figure 7.14: Simulated Voltage Sag at 110kV bus bar at Chinnaputhur

Substation

Figure 7.15: Measured voltage sag at chinnaputhur substation

7.5.3 Tripping of the wind generator

The simulation assumes that the generator 11 is switched out from the
network during the concerned interval. The scenario is to switch off the feeder and
 115

then restoring it back. The simulation results of such a scenario show that there is a
variation in phase voltage as given Figures 7.16.
The noteworthy observations are: The line-to-line voltage of generator 11
suddenly starts decaying and becomes zero at some point in time. It is restored
back to the normal operation after the clearance of fault. The Figure 7.17 shows
that the measured voltage interruption wave is very similar to the simulated wave
form in the DIgSILENT.

Figure 7.16: Simulated voltage waveform of generators during tripping

 116

Figure 7.17: Measured Voltage waveform during tripping

7.5.4 Swell
Voltage swell is an increase (between 10% and 90%) in RMS voltage at a
power frequency for duration from 0.5 cycles to 1 min. In the designed network,
the load connected to HV bus bar is disconnected. The voltage thus rises from its
nominal value of 1 p.u to1.01 p.u shown in Figure 7.18. The Figure 7.19 shows the
measured waveform which is similar to the simulated one.

Figure 7.18: Waveform of simulated Voltage swells at generator bus bar

 117

Figure 7.19: Waveform of measured Voltage swells at 110kV

7.6 Conclusion
The E2 wind feeder of Chinnaputhur substation is modelled as per the
layout. It consists of eleven wind turbine of same rating 0.85MW, 0.44kV and total
capacity is 9.35MW.As the gird busbar voltage is increased 10%, current flow is
decreased and grid reactive power controlled to its reference value. The grid
frequency is maintained at 50Hz by the grid side converter control. But the
generator frequency varies to change in the wind speed below its rated speed
12m/s. The pitch mechanism is active above the rated wind speed and the generator
power of G1 and G11 under consideration is kept at its rated value of 0.85 MW.
The speed responses of both generators are similar except a small deviation of
0.01p.u. This deviation is due to the servo mechanism slow response in the pitch
control block. The DC-link voltages of G1 and G11 are maintained constant at the
set value with help of generator side controller. The symmetrical three-phase fault
is applied at the high voltage terminal at the time10.00s and it lasts for 10 millisecs.
The fault causes a voltage drop of 30 % of the rated voltage at the MV bus bars of
G1 and G11. The fault causes drop in the active power.
The wind farm performance´ and their contribution to the voltage control
is assessed and evaluated by means of simulations with the use of a generic
and realistic power system model of Chinnaputhur substation at Dharapuram
District. The simulation results of PMSG wind farm equipped with voltage control
 118

can participate to re-establish the voltage properly during and after grid fault. The
conclusion is that a variable speed wind generator equipped with converter
control can help the grid to maintain the voltage stability. It also provides a good
dynamic performance for change in wind speed and grid voltage.
 119

CHAPTER 8
REMEDIAL MEASURES AND SUGGESTIONS FOR
IMPROVING POWER QUALITY

8.1 Improvement techniques

Many techniques can be utilized for eliminating voltage drop in electric
power system. Few of them are:
1. Capacitor application
2. Re-Conductoring
3. Bifurcation
4. Load balancing
5. Reconfiguration
6. Fact devices

8.2 Capacitor application

In ac supply, if the load is of purely resistive nature then the apparent power
will be equal to the active power. But in case of inductive loads, the active power is
less than apparent power. It means that inductive load draws two types of power.
One is called active power and other is reactive power. If reactive power is
eliminated, then apparent power becomes equal to active power. The total
elimination of reactive power from the system is difficult; rather it can be
compensated partially by the application of capacitor. Lagging current usually
causes greater voltage drop. This is primarily due to the fact that the reactance of
the line is generally greater than resistance. Shunt capacitors are utilized to
counteract this voltage drop by producing a ―voltage rise‖ across the line. The
installation of capacitor also improves the power factor which in turn reduces the
line losses. The application of capacitor in the distribution system provides many
benefits The Reduction in voltage drop, enhancement of overall system voltage,
improvement in system voltage regulation, release of system capacity, reduction in
 120

the distribution system overloading, increase in system efficiency, reduction in line

current, and power factor improvement are the most distinguishable advantages of
capacitor utilization in the distribution system.
The other benefits of capacitor application include reduction of power loss in
the distribution system, elimination of capital expenditure involved in system
rehabilitation and reactive power compensation.

8.3 Re-Conductoring
Re-Conductoring of distribution feeder is another effective method of
voltage improvement and voltage drop reduction. However, this method is costly as
it involves the replacement of high impedance small capacity conductor by low
impedance high capacity conductor. This technique is recommended only when the
benefit cost ratio of re-conductoring is equal or greater than unity. Cost is reduced
on the installation of new and dismantling of existing conductor. The benefits
include reduction of voltage drop and saving in energy losses for a period of about
five years and return value of the dismantled conductor.

8.4 Bifurcation
Bifurcation is applied to distribution feeder when it is highly overloaded and
simple capacitor application and other techniques do not provide the desired
results. The existing feeder is replaced by two or more feeders as per requirement
of the locality to supply a portion of the existing area. In this method the re-
allocation of loads is carried out. Additional feeder(s) are constructed either from
the same grid or from another nearby grid station depending on the economic
justification and achievement of maximum benefits. This method is costly as it
involves the construction of new feeder(s).
 121

8.5 Load balancing

This technique is economically feasible and can be accomplished quickly.
Unbalance of load on the distribution feeder occurs when single phase loads and
single phase transformers are not effectively divided among the three phases. If
these loads are symmetrically distributed among the three phases of the system,
considerable reduction in voltage drop can be achieved. For optimal operation of
the distribution feeder, load on transformers should be distributed equally on the
basis of their rated power.

8.6 Feeder reconfiguration

Network or feeder reconfiguration is the process of altering the topological
structure of distribution feeders by changing the open/close status of sectionalizing
and tie switches. Reconfiguration of distribution feeder can be used as a planning
as well as time control tool. Modifying the redials of distribution feeder from time
to time, by changing the open/closed status of these switches can improve the path
of power flow. Transfer of load from heavily loaded feeder to lightly loaded feeder
improves the operating condition of overall distribution feeder. Feeder
reconfiguration allows the transfer of load from heavily loaded feeder to relatively
lightly loaded feeder. Such transfers are effective not only in altering the level of
loads on the feeders being switched but also in improving the voltage profile along
the feeders.

8.7 Fact devices

A shunt and series device can be connected at the suitable location in the
wind farm to improve the power quality issue. The statcom is the shunt device
which can inject or absorbs the reactive power from the system to maintain the
system voltage at the reference value. The statcom is suitable fact device to reduce
the voltage sag under the load variation and fault. The DVR is the series voltage
device in the line to control the voltage to the reference value. Transient due to
 122

switching of isolator and capacitance switching are controlled by the dynamic

voltage reactor. The UPFC is the combination of shunt and series device. The
power flow at the power system is controlled either series or shunt injection or
both.

8.8 Implication of poor power quality

Poor power quality in electric power distribution system has many
implications. It increases in the line and equipment current leading to additional
ohmic losses. The excess line and equipment currents enhance the capital
investment. In most of the cases, this increase of current also brings significant
changes in the operating temperature of distribution network and electric
appliances which not only reduce the life of the equipments but also deteriorate the
power quality of the system. The losses in the system increases which in turn
minimizes the efficiency of the distribution system. The numbers of outages in the
system increases and severely deteriorates the quality of production in the industry.
The frequent malfunction of the equipments may also reduce the production
completely. The distribution engineers try utmost to eliminate the implications of
the poor power quality by implementing the proper and effective system designs.

8.9 Custom power solutions

In the deregulated environment, the manufacturers are less willing to put up
with products loss and defects caused by power supply problems. Small outages
play significant role among the utilities and customers. Under such circumstances
the utilities are bound to mitigate the new and sterner concept of power quality
.Custom power solution is implemented on the utility side of the meter and
integrates the measures on the customer‘s side as well. The Majority of the power
conditioning devices is custom built to meet certain customer technical
performance requirement and target cost. The technical performance is driven by
statutory and regulatory requirements, compliance to standard etc. In many cases,
 123

the major driver is the financial incentive arising out of the power conditioning. To
ensure the optimality of the solution, it is essential to capture all the performance
requirements, cost and benefit elements and arrive at an optimal solution.

8.10 Advantages of power conditioning

The advantages of the power conditioning can be classified into technical
and nontechnical types. Technical merits are those that accrue due to improvements
in operating efficiency and are directly related to basic laws of physics and are
invariant with policies and guidelines to greater extent. The reduction of current in
line and equipment minimizes the ohmic losses due to reactive power
compensation and harmonic filtering is one example out of many. Non technical
benefits are those which can be obtained because of fiscal income due to regulatory
norms and compliance and are largely dependent upon the existing policies and
norms. Although some of the non technical advantages have their roots in technical
aspects of power conditioning, largely depend upon utility policies.

The general merits of power quality improvement include;

1. Minimization of line and equipment currents, losses and thus lowered energy
bill.
2. Release of blocked capacity and consequent avoided cost of capital
investment.
3. Power factor improvement and avoided penalty for low power factor.
4. Reduction in maximum demand and reduction in demand charges.
5. Benefits in taxes such as accelerated depreciation benefits for installation of
power conditioning or energy saving devices.
6. Voltage profile improvement and consequent effective operation of electrical
devices.
7. Elimination of harmonic distortion and consequent reduction in copper loss,
core loss and stray loss.
 124

8. Prevention of malfunction of equipment and avoided loss of production.

9. Elimination of unplanned outages and reduction in loss of production and
revenue.
10.Reduction in failure of equipment due to reduced electrical and thermal
stresses.
11.Increased life and reliability of electrical devices due to lower operating
temperatures and lower losses.

8.11 Fault ride through

As a result of the increased wind power development, focus has come to the
influence of wind power on the security of the system. An important security issue
is the ability of wind turbines to stay connected, or ―ride through‖ when the voltage
dips due to a fault in the system. The main fault-ride-through concern is on the
possible effect that a fault in the transmission system will have, because such a
fault will influence the voltage significantly in a large area. Thus, if many wind
turbines are subject to a voltage dip caused by a transmission system fault, it is
important that these wind turbines are able to stay connected. Otherwise,
significant generation capacity can be lost, causing power and frequency control
problems in the system after the fault.

The fault-ride-through demand is also a challenge to the voltage recovery

after a grid fault. Before fault-ride-through requirements became relevant to avoid
loss of significant generation, the wind turbines had to disconnect to avoid large
inrush currents when the voltage recovers, in order to avoid that the inrush current
trips a system protection relay. Thus, the fault-ride-through requirements also raise
challenges on how to recover the voltage after a dip, especially for wind turbines
directly connected to the induction generators.
 125

8.12 Suggestions

1. All new wind generation plants must meet WECC LVRT requirements.
2. All new wind plants should be Type 3 or Type 4 generators that are capable
of providing dynamic reactive support to help the transmission grid to meet
applicable WECC transient stability performance standards and to prevent
the potential tripping due to low voltages.
3. In the event that some of the new wind plants are of Type 1 or 2 with no
dynamic reactive capability, the generator owner must provide sufficient
reactive resources to meet the Low Voltage Ride Through standards and
voltage control standard. Additional studies may be required to verify that
the generator has been provided appropriate additional external dynamic
reactive support to meet the interconnection standards.
4. Re-evaluate the optimal location and size of the dynamic reactive support.
5. Analyze the best solution for improving the nose point of the Q-V analysis
for critical 250 kV buses under critical contingency conditions. Potential
solutions include the use of series compensation and reduction of proposed
shunt compensation.

Type 1 – conventional induction generator

Type 2 – wound rotor induction generator with variable rotor resistance
Type 3 – doubly-fed induction generator
Type 4 – full converter interface

Type 1 machines operate in a very narrow speed range, and always consume
reactive power during operation. The reactive power consumption is a function of
active power production and grid conditions, and it cannot be controlled.
Consequently, both the reactive power consumption of the generator and the
 126

reactive power requirements of the grid must be supplied by additional equipment

— usually switched shunt capacitors.
Type 2 machines have wider speed variation and tend to exhibit slower
active power fluctuations than Type 1 machines, but have similar reactive power
characteristics. Under load, the machines consume reactive power equal to
approximately half of the MW output.
Type 3 and Type 4 machines use substantial power electronics to provide
wider speed range and fine control of active power production. The power
electronics also inherently provide the ability to produce or consume reactive
power. It is largely controllable and independent of the active power production. In
this regard, these machines resemble conventional synchronous generators with
excitation systems and automatic voltage regulators (AVR). The details of
performance are different between manufacturers. Generally, wind plants with
Type 3 or Type 4 generators have the ability to provide relatively fast
voltage or power factor control. The ways in which each manufacturer controls and
coordinates the reactive power production and balance differs

Summary
Electric power quality is an electric power problem manifested in voltage,
current or the frequency deviations and results in failure of load side electric
equipments. From electric utility point of view, the power quality is the supply of
electrical power as per specified standards, whereas from the end user sight, it is
the smooth functioning of electrical equipments without any disruption. In the
deregulated and competitive environment, both the electric utilities as well as the
customers are becoming increasingly concerned about the quality of electric power.
The major reason for increased concerns is the availability of sophisticated
technology that has developed extremely sensitive electrical/electronic equipment.
Any sort of variation in electrical parameters greatly changes the characteristics of
such delicate equipments. Widespread use of electronics in the system from home
 127

electronics to the control of massive and costly industrial processes has increased
the awareness of power quality. The study of power quality and way to control it is
a concern for electric utilities and electricity consumers. Equipments have become
more sensitive to even minute changes in the power supply. Different indices are
used by distribution engineers for the quantification of electric power quality.
These indices have general properties that they are relatively easy to calculate and
are calculated using standardized procedures. Majority of these indices are
interpreted and applied by distribution engineers. The proliferation of micro-
electronic processors in a wide range of equipment, from domestic appliances to
automated industrial and hospital diagnostic systems, has increased the
vulnerability of such equipment to power quality problems. These problems
include a variety of electrical disturbances, which may originate in several ways
and have very different effects on various kinds of sensitive loads.
Distributed Generation (DG) is said to be power generation paradigm of the
new era because of its ability to resolve many customer problems, especially from
power quality point of view. DG is used to provide electricity service at a high
level of reliability and power quality than conventional grid power system. DG is
capable of protecting sensitive loads from momentary voltage variations. It can
provide uninterruptible power supply to ride through any sort of outage until
primary or secondary power is restored. It is environmental friendly, promotes
renewable energy resources, improves system power factor, and power quality in
terms of node voltage drop and power loss reduction. A DG system provides
protection from long term outages. Power quality system needs to include
important design criteria that relates to system hardening. It is necessary to
understand the business mission and the tolerance for outage. To eliminate the
power quality and reliability disruptions, a facility may seek to minimize its energy
prices by introducing DG.
 128

REFERENCES
1. Akhmatov V., Knudsen H., Nielsen A.H., Pedersen J.K. and Poulsen N.K.,
‗Modelling and Transient Stability of Large Wind Farms‘, Electric Power
and Energy Systems, Vol. 25, 2003, pp. 123-144.
2. Ana I. Estanqueiro ‗A Dynamic Wind Generation Model for Power Systems
Studies‘, IEEE Transactions on Power Systems, Vol. 22, No. 3, August
2007, pp. 920-928.
3. Boulaxis N.G., Papathanassiou S.A. and Papadopoulos M.P., ‗Wind Turbine
Effect on the Voltage Profile of Distribution Networks‘, Renewable Energy,
Vol. 25, 2002, pp. 401-415.
4. Chan T.F. and Lai L.L, ‗An Axial-Flux Permanent-Magnet Synchronous
Generator for a Direct-Coupled Wind-Turbine System‘, IEEE Transactions
on Energy Conversion, Vol. 22, No. 1, March 2007, pp. 86-94.
5. Chompoo-inwai C., Wei-Jen Lee, Pradit Fuangfoo, Mitch Williams and
James R. Liao, ‗System Impact Study for the Interconnection of Wind
Generation and Utility System‖, IEEE Trans. on Industry Applications, Vol.
41, No. 1, Jan./Feb 2005, pp. 163-168.
6. Chompoo-inwai C., Yingvivatanapong C., Methaprayoon K. and Wei-Jen
Lee, ‗Reactive Compensation Techniques to Improve the Ride-Through of
Induction Generators during Disturbance‘, IEEE Trans. on Industry
Applications, Vol. 41, No. 3, May-June 2005, pp. 666-672.
7. Divya K.C. and Nagendra Rao P.S., ‗Models for Wind Turbine Generating
Systems and their Application in Load Flow Studies‘, Electric Power
Systems Research, Vol. 76, 2005, pp. 844-856.
8. Eisenhut C., Krug F., Schram C. and Klockl B., ‗Wind-Turbine Model for
System Simulations Near Cut-In Wind Speed‘, IEEE Transactions on Energy
Conversion, Vol. 22, No. 2, June 2007, pp. 414-420.
9. Julio Usaola, Pablo Ledesma, Juan M. Rodriguez, Jose L. Fernandez,
Domingo Beato, Ramon Iturbe and Jose R. Wilhelmi, ‗Transient Stability
 129

Studies in Grids with Great Wind Power Penetration - Modelling Issues and
Operation Requirements‘, IEEE Power Engineering Society General
Meeting, Vol. 3, 2003, pp. 1534-1541.
10.Muljadi E., Wan Y., Butterfield C.P. and Parsons B., ‗Study of a Wind Farm
Power System‘, NREL Report, 2002.
11.Oscar Alexis M.A., ‗Issues Regarding the Integration of Induction Wind
Turbines in Weak Electrical Networks‘, Nordic Wind Power Conference,
Chalmers University of Technology, March 2004, .-23.