Power Flow Studies of an AC-DC

Transmission System

Ummaiah Bondhala

A Thesis Submitted to
Indian Institute of Technology Hyderabad
In Partial Fulfillment of the Requirements for
The Degree of Master of Technology

Department of Electrical Engineering

June 2011
 Acknowledgements

First I would like express my sincere gratitude to my supervisor Dr. Vaskar Sarkar for his
help, sincere guidance, patience and constant encouragement, without which this thesis
would not be possible to finish. The support i got from him through out my course work
is invaluable.

I would like to express my gratitude to Dr. Ravinder David Koilpillai and faculty at
IIT Hyderabad for their constant supervision as well as for providing necessary informa-
tion regarding the project.

I would also like to thank the Indian Institute of Technology Hyderabad College for
their generous support and providing peaceful environment to complete my work.

My thanks and appreciations also go to my friends and people who have willingly helped
me out with their abilities.

At last but not the least, i wish to thank my parents and sisters for their unconditional
love and support. They inspired me a lot and encouraged me to go my own way, without
whom I would be unable to complete my course and thesis. special thanks to my brother
Haranath.

Ummaiah Bondhala
 To
My Parents and Sisters
 Abstract

High Voltage Direct Current (HVDC) technology has characteristics that make it espe-
cially attractive for certain transmission applications. The same includes long-distance
bulk power delivery, asynchronous interconnections, long submarine cable crossings and
fast power controllability. Then, the HVDC transmission has proved its potential to be
an interesting alternative or complement to the AC transmission.

The thesis presents Newton - Raphson method for the load flow analysis modified to
achieve compatibility for AC - DC systems with the integrated DC link in the AC net-
work. The elements of the Jacobian for the AC network are modified to include DC real
and reactive power at the AC - DC buses, and their dependency on the a.c system vari-
ables. The DC equations such as voltage expressions at rectifier and inverter, network
configuration equation i.e., point-to-point, multiterminal series connection or multitermi-
nal radial connection, real and reactive power demands at converters are represented in a
per unit form which will be compatible with per unit a.c equations. For various control
strategies, simulations are carried out for point-to-point DC link and multiterminal DC
links.

Simulations are also carried out for optimization of total power generation cost with the
help of GAMS (General Algebraic Modeling System) and MATLAB for point-to-point
and multiterminal HVDC transmission system. All equality and inequality constraints
are considered in this optimization so that the entire power system can remain intact
during the real-time operation.
Contents

Abstract i

List of Figures vi

List of Tables vii

Nomenclature viii

1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Literature survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Structure of thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 HVDC Transmission System 6

2.1 Configurations of HVDC . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.1 Monopolar HVDC system . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.2 Bipolar HVDC system . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.3 Homopolar HVDC system . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.4 Back-to-back HVDC system . . . . . . . . . . . . . . . . . . . . . . 8
2.1.5 Multiterminal HVDC system . . . . . . . . . . . . . . . . . . . . . 8
2.2 Components of HVDC System . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Converter Performance Analysis . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.1 Operation of the bridge . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.2 Rectifier operation . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.3 Inverter operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

ii
 2.4 HVDC Transmission Advantages . . . . . . . . . . . . . . . . . . . . . . . 15
2.5 HVDC Transmission Disadvantages . . . . . . . . . . . . . . . . . . . . . . 16
2.6 HVDC Transmission Applications . . . . . . . . . . . . . . . . . . . . . . . 17
2.7 Control of HVDC System . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.7.1 Basic principles of control . . . . . . . . . . . . . . . . . . . . . . . 18
2.7.2 Basis for selection of controls . . . . . . . . . . . . . . . . . . . . . 20

3 Power Flow Studies 21

3.1 Significance of Load Flow Study . . . . . . . . . . . . . . . . . . . . . . . 21
3.2 Types of Buses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2.1 PQ bus-bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2.2 PV bus-bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2.3 Slack or swing bus-bar . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.3 Formulation of Power Flow by Newton-Raphson Method . . . . . . . . . . 23
3.4 AC-DC Power Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.4.1 Point-to-point HVDC . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.4.2 Multiterminal HVDC . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.5 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.5.1 Point-to-point DC transmission Line . . . . . . . . . . . . . . . . . 36
3.5.2 Multiterminal HVDC . . . . . . . . . . . . . . . . . . . . . . . . . . 40

4 Optimization of Power Flow 44

4.1 Conventional Optimal Economic Scheduling . . . . . . . . . . . . . . . . . 44
4.2 Formulation of OPF problem . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.2.1 Objective function . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.2.2 Network equations . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.2.3 Generation limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.2.4 Bus voltage limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.2.5 Limits on reactive power Support . . . . . . . . . . . . . . . . . . . 48
4.2.6 Limits on power Flow . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.2.7 DC terminal voltage equations . . . . . . . . . . . . . . . . . . . . . 48

iii
 4.2.8 Firing angle limits . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.2.9 Transformer tap changer limits . . . . . . . . . . . . . . . . . . . . 49
4.3 Characteristic Features of OPF . . . . . . . . . . . . . . . . . . . . . . . . 49
4.4 OPF Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.5 Results and Discusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.5.1 Point-to-point DC transmission line . . . . . . . . . . . . . . . . . . 51
4.5.2 Multiterminal HVDC . . . . . . . . . . . . . . . . . . . . . . . . . 53

5 Conclusions and Future Scope of Research 57

5.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.2 Future Scope of Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

Bibliography 60

iv
List of Figures

1.1 Schematic diagram of overall system with HVDC system . . . . . . . . . . 2

2.1 Monopolar HVDC link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2.2 Bipolar HVDC link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.3 Homopolar HVDC link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 Back-to-back HVDC link . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.5 Multiterminal series link . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.6 Multiterminal parallel link . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.7 Components of HVDC system . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.8 Three phase, full wave Graetz circuit . . . . . . . . . . . . . . . . . . . . . 12
2.9 Output voltage waveforms with firing angle delay . . . . . . . . . . . . . . 13
2.10 Equivalent circuit of rectifier . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.11 Equivalent circuit of inverter . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.12 Equivalent circuit of an monopolar HVDC link . . . . . . . . . . . . . . . 19

3.1 Commutation failures due to voltage magnitude reduction and phase angle
shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2 Commutation failure due to increased in DC current . . . . . . . . . . . . 33
3.3 HVDC connections of several networks (a) By separate point-to-point HVDC’s
(b) By multiterminal HVDC . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.4 Characteristics of multiterminal parallel DC lines . . . . . . . . . . . . . . 35
3.5 Four bus system for AC-DC load flow . . . . . . . . . . . . . . . . . . . . . 36
3.6 Relation between reactive power at converters, Id versus power at inverter
side in case study case (i) . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

v
3.7 Relation between reactive power at converters, Id versus power at inverter
side in case study (ii) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.8 Relation between reactive power at converters, Id versus power at inverter
side in study case (iii) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.9 Schematic diagram of series connection . . . . . . . . . . . . . . . . . . . . 40
3.10 Relation between transformer tap positions at converters, Id and power at
inverter side in series connection . . . . . . . . . . . . . . . . . . . . . . . . 41
3.11 Schematic diagram of radial connection . . . . . . . . . . . . . . . . . . . 42
3.12 Relation between transformer tap positions at converters, Id and power at
inverter side in radial connection . . . . . . . . . . . . . . . . . . . . . . . 42

vi
List of Tables

3.1 Converter Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

4.1 Line Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

4.2 Bus Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.3 Coefficients for Cost Functions of Generators . . . . . . . . . . . . . . . . . 52
4.4 Point-to-point DC Transmission Result . . . . . . . . . . . . . . . . . . . . 53
4.5 DC Variables in Point-to-point Connection . . . . . . . . . . . . . . . . . 53
4.6 GAMS Status and Objective Value . . . . . . . . . . . . . . . . . . . . . . 53
4.7 Multiterminal Series Connection Result . . . . . . . . . . . . . . . . . . . . 54
4.8 DC variables in Multiterminal Series Connection . . . . . . . . . . . . . . 54
4.9 GAMS Status and Objective Value . . . . . . . . . . . . . . . . . . . . . . 54
4.10 Multiterminal Radial Connection Result . . . . . . . . . . . . . . . . . . . 55
4.11 Firing Angles of Converters in Multiterminal Radial Connection . . . . . . 55
4.12 Transformer Tap positions at Converters in Multiterminal Radial Connec-
tion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.13 Voltages, Currents and Real Powers in Multiterminal Radial Connection . 56
4.14 GAMS Status and Objective Value . . . . . . . . . . . . . . . . . . . . . . 56

vii
Nomenclature

List of Symbols

α Delay firing angle.

γ Extinctison advance angle.

Vdi Inverter output voltage.

Pdi Power at inverter end.

Id Direct currrent.

Y −Y Star-Star connection of transformer.

Y −∆ Star-Delta connection of transformer.

Vdr Rectifier output voltage.

Vdo Rectifier output voltage at α= 0.

Lc Source inductance.

ELL RMS value of line to line voltage.

µ Overlap angle.

δ Extinction delay angle.

φ Phase angle difference between voltage and current.

β Ignition advance angle.

viii
Rdc DC line resistance.

Pdr Power at rectifier end.

Ir Direct current associated at rectifier end.

Ii Direct current associated at inverter end.

trec Transformer tap position associated at rectifier end.

tinv Transformer tap position associated at inverter end.

p.u Per unit.

v Voltage.

i Current.

R Resistance.

X Reactance.

B Suseptance.

PGi Real power generation at ith bus.

QGi Reactive power generation at ith bus.

PLi Real power load at ith bus.

QLi Reactive power laod at ith bus.

Qmin,i Minimum reactive power limit at ith bus.

Qmax,i Maximum reactive power limit at ith bus.

Pi Net real power at it h bus.

Qi Net reactive power at it h bus.

PDC Active power associated with converter.

ix
QDC Reactive power associated with converter.

t1 Transformer tap position associated with rectifier.

t2 Transformer tap position associated with inverter.

$ Dollar.

List of Acronyms

HVDC High Volatge Direct Current.

AC Alternating Current.

DC Direct Current.

GAMS General Algebraic Modeling System.

MATLAB MATrix LABoratory.

CSC Current Source Converter.

VSC Voltage Source Converter.

IGBT Insulated Gate Bipolar Transistor .

PWM Pulse Width Modulation.

HVDCPLUS HVDC Power Link Universal System.

ELD Economic Load Dispatch.

OPF Optimal Power Flow.

SCR Silicon Controlled Rectifier.

MW Mega Watt.

MVAR Mega Volt Ampere Reactive.

x
Chapter 1

Introduction

1.1 Background

Alternating current (AC) is widely used in industries and residential areas, but for the
long transmission line (more than 600 Km) AC transmission is more expensive than direct
current (DC). Technically, AC transmission line control is more complicated because of
the frequency and dependency of power transfer on angle difference between the voltage
phasors at the two ends. DC transmission does not have these limitations, which has led
to build long High Voltage Direct Current (HVDC) transmission lines [1], [2]. HVDC
technology is a high power electronics technology used in electric power systems to trans-
fer bulk power over long distances. The DC transmission requires conversion at two ends,
from AC to DC at the sending end and DC to AC at the receiving end. This conversion
is done at converter stations. By simple control action, converter can be switched from
rectifier to inverter and vice-versa. Thus facilitating power reversal.

The invention of the high voltage mercury valve shown the way towards the develop-
ment of HVDC transmission. By 1954, the first commercial HVDC connecting two AC
systems came in to operation in form of submarine cable link between the Swedish main-
land and the island of Gotland [2]. Nowadays, the HVDC is being widely used all around
the world. Until recently, HVDC based on thyristors uses the Current Source Converter
(CSC) configuration. Now, a new type of HVDC transmission using more advanced

1
semiconductor technology instead of thyristors is available for power conversion. The
semiconductors used are Insulated Gate Bipolar Transistors (IGBTs) and the converters
are voltage source converters (VSCs) which operate with high switching frequencies (1-
2kHz) utilizing pulse width modulation (PWM). The technology is commercially available
as HVDC Light or HVDCPLUS (Power Link Universal Systems) [3].

1.2 Literature survey

A brief introduction to the HVDC technology and the configurations like monopolar, bipo-
lar, back to back and multiterminal with their operation are presented in [3], [4]. Figure
1.1 shows a simplified schematic picture of an HVDC system with the basic principle of
transferring electric energy from one AC system or node to another in any direction. The

Figure 1.1: Schematic diagram of overall system with HVDC system

system consists of three blocks: two converter stations and a DC line. Each converter
station has several components involved in power conversion. In [5], [6], analysis of Graetz
bridge, which is used for AC/DC or DC/AC conversion, is explained with commutation
and without commutation, and the equivalent circuit to represent HVDC line are pre-
sented. The following assumptions are made during the analysis.
(i) The valves of the converter are ideal and have no arc voltage drop.
(ii) AC source delivers a constant voltage and frequency.
(iii) The DC voltage and current have no ripple.

2
 With the presence of DC link, the operating condition of the AC-DC system is defined
as a.c bus voltage magnitudes, a.c bus voltage angles and DC variables. DC variables
includes firing angles at converters, tap positions of the transformers located at converter
stations, DC voltages at rectifier and inverter, and direct current through the line. Recti-
fier has minimum (α) limit of about 5o to ensure adequate voltage across the valve before
firing. Normally rectifier operates at a value between 15o and 30o so as to leave room
for increasing voltage for power control. In case of inverter, it is necessary to maintain
minimum extinction angle (γ) to avoid commutation failure [7]. Commutation failures
are also caused because of the voltage dips due to a.c system faults, shift in phase angle
of the line to line voltage and increase in the DC current because of faults at inverter side
[8].

The primary functions of the DC control in a typical two terminal HVDC line are
(i) control of power flow between the terminals.
(ii) protect the equipment against the current/voltage.
HVDC control can be accomplished by gate control of valve firing angle or control of the
ac voltage through tap changing of the converter transformer.

Modeling of AC-DC converter and the method used to include the equations for DC
converters and transmission lines in a Newton Raphson AC system power flow are ex-
plained in [9]. The DC equations are presented in a per unit form that is compatible with
the per unit AC equations. As a result of this conversion, the AC and DC equations can
be solved simultaneously, rather than serially. This paper also includes the assumptions
made in the derivation of the equations representing the AC-DC converter.

When the AC system includes an HVDC link, the equations of the power balance at
the AC terminal are modified by including the powers at converter stations. The methods
for load flow calculations of mixed AC-DC systems are grouped into three categories.

(i) H.A Sangavi and S.K Banarjee [10] proposed a sequential approach for carrying out
the 1oad flow analysis of an integrated AC - DC power system. The particular method is
based on solving the DC system and AC system equations separately.

3
(ii) In extended variables/unified method [11], [12], AC system and DC system are com-
bined together and solved simultaneously. Because of its better computing efficiency and
convergence it is more useful than sequential method in AC - DC power system.

(iii) Eliminated variable method [13], converters are treated as voltage dependent loads
and the DC variables are eliminated from the power flow equations.

In [12], [14], more emphasis is placed on HVDC line control modes and suitable mod-
ification of conventional power flow jacobian matrix to include new elements contributed
by the HVDC link. Control modes presented in [12] are, specifying firing angle α, the
extinction angle γ, voltage at inverter side (Vdi ) and power at inverter side (Pdi ) is con-
trol mode A. Converter transformer taps are varied in order to meet these specifications.
In Control mode B, the extinction angle γ is kept at its lower limit at the inverter, the
transformer taps and the power Pdi are specified. Power is controlled by varying the firing
angle α at the rectifier. Firing angle (α) control is preferred to tap transformer, because
of tap changer control is too slow. In control mode C, α is fixed to its minimum value
and γ is regulated to maintain constant power. If the current Id is kept constant instead
of the power Pdi , three other control modes are defined.

Concept of extending two terminal HVDC transmissions to multiterminal systems is

briefly explained in [15], [16], [17]. It comprises classifications of multiterminal HVDC
such as serial, radial and mesh connections, and their operation with their control char-
acteristics. In a multiterminal transmission system, telecommunication is important for
coordination of current level in different stations. A brief overview of DC line faults, AC
faults and commutation failures are also discussed. A method for tapping energy from
an intermediate point on an HVDC transmission line is presented in [18] has two main
parts. The first one is the unit to extract power from the line that contains the conven-
tional bridge and the second part of the station is the AC machine which provides the
commutating voltage for the bridges.

In Economic Load Dispatch (ELD) [19] the demand is considered to be an aggregate

4
parameter for the entire system. While load flow equations are introduced in ELD as a
system of demand supply balance constraints, the optimum solution yields a set of decision
variables satisfying desired objective (of cost minimization, loss minimization ....). Such
a formulation is called as optimal power flow (OPF). OPF problem formulation with
incorporation of HVDC link in a AC transmission system is explained in [20], [21]. This
includes the vector of equality constraints such as power balance equations, inequality
constraints such as upper and lower bounds of transmission line capacity, active and
reactive power generation outputs and operating angles of converters.

1.3 Structure of thesis

Chapter 2 presents the different configurations of HVDC transmission system, main
components of HVDC, rectifier and inverter analysis of Greatz circuit, advantages and
disadvantages, applications of HVDC and principles of HVDC control .

Chapter 3 explains importance of power flow studies, formulation of load flow studies
by Newton-Raphson method and inclusion of Direct current transmission line equations
in power flow for both point-to-point and multiterminal transmission systems. Different
control strategies for operation of AC DC transmission are also explained in this chapter.
MATLAB simulations for both single and multiple HVDC transmission line are discussed.

Chapter 4 presents an Optimal Power Flow (OPF) model considering both AC and DC
lines. Optimization of power generation cost is done with the help of General Algebraic
Modeling System (GAMS) and MATLAB by including point-to-point and multiterminal
HVDC lines in IEEE standard 14 bus system.

Chapter 5 Concludes the thesis and shows the road map for future work.

5
Chapter 2

HVDC Transmission System

2.1 Configurations of HVDC

HVDC converter bridges together with lines or cables can be arranged in a number of
configurations [3], [4] as shown in Figs. 1.1 to 1.6

2.1.1 Monopolar HVDC system

In the monopolar configuration, two converters are connected by a single pole line
and a positive or a negative DC voltage is used. In Fig. 2.1, there is only one insulated
transmission conductor installed and the ground or sea provides the path for the return
current. Alternatively, a metallic return conductor may be used as the return path where
possible interference with underground/ underwater is objectionable.

2.1.2 Bipolar HVDC system

This is the most commonly used configuration of HVDC transmission systems. The
bipolar configuration is shown in Fig. 2.2. It uses two insulated conductors as positive
and negative poles. The two poles can be operated independently if both neutrals are
grounded. The bipolar configuration increases the power transfer capacity. Under normal
operation, the currents flowing in both poles are identical and there is no ground current.
In case of failure of one pole, power transmission can continue in the other pole which

6
 Figure 2.1: Monopolar HVDC link

increases the reliability.

Figure 2.2: Bipolar HVDC link

2.1.3 Homopolar HVDC system

In the homopolar configuration, shown in Fig. 2.3. Here, two or more conductors have
the negative polarity can be operated with ground or a metallic return. With two poles
operated in parallel, the homopolar configuration reduces the insulation costs. However,
the large earth return current is the major disadvantage.

7
 Figure 2.3: Homopolar HVDC link

2.1.4 Back-to-back HVDC system

This is the common configuration for connecting two adjacent asynchronous AC sys-
tems. Two converter stations are located at the same site and transmission line or cable
is not needed. A block diagram of a back-to-back system is shown in Fig. 2.4. The two
AC systems interconnected may have the same or different nominal frequencies.

Figure 2.4: Back-to-back HVDC link

2.1.5 Multiterminal HVDC system

In the multiterminal configuration, three or more HVDC converter stations which are
geographically separated are interconnected through transmission lines or cables. The
system can be either parallel, where all converter stations are connected to the same

8
voltage as shown in Fig. 2.5. or series multiterminal system, where one or more converter
stations are connected in series as shown in Fig. 2.6. A hybrid multiterminal system
contains a combination of parallel and series connections of converter stations.

Figure 2.5: Multiterminal series link

Figure 2.6: Multiterminal parallel link

2.2 Components of HVDC System

A typical HVDC system, shown in Fig. 2.7, comprises AC filters, transformers, con-
verters, smoothing reactors, DC capacitors and DC lines/cables, Electrodes and AC circuit
Breakers [5], [6].
Converters
Perform AC/DC and DC/AC conversion. The valve bridges consists of high voltage valves

9
 Figure 2.7: Components of HVDC system

connected in a 6-pulse or 12-pulse arrangements as shown in Fig. 2.8.

Transformers
Normally, the converters are connected to the AC system via transformers. The most
important function of the transformers is to transform the voltage of the AC system to a
level suitable for the converter
Smoothing reactors
These are large reactors having inductance as high as 1.0 H connected in series with each
pole of each converter station which serves following purposes:
1) Decrease harmonic voltages and currents in the DC line.
2) Prevent current from being discontinuous at light load.
3) Limit the crest current in the rectifier during short-circuit on the DC line.
AC Filters
The AC voltage output contains harmonic components, caused by the switching of the
Thyristirs/IGBTs. The harmonics emitted into the AC system have to be limited to
prevent them from causing malfunction of AC system equipment or radio and telecom-
munication disturbances.

10
Electrodes
Most DC lines are designed to use earth as a neutral conductor for some time. The con-
nection to the earth requires a large-surface-area conductor to minimize current densities
and surface voltage gradients. This conductor is referred to as an electrode.
AC Circuit Breakers
For clearing faults in the transformer and for taking the DC line out of service, circuit-
breakers are used on the AC side.
DC Lines
These may be overhead lines or cables.

2.3 Converter Performance Analysis

HVDC converter also known as Graetz bridge is shown in Fig. 2.8. Normally two such
six pulse converters, one connected Y-Y and the other Y-∆ transformers are used. This
helps in eliminating multiples of sixth harmonics on the dc side which reduces harmonic
filters significantly.

The following assumptions are made during analysis [5].

a) All phases of the supply voltage are identical and are displaced by exactly 120o ;
b) The direct current (Id ) is constant and ripple free;
c) The transformer leakage reactance is unchanged;
d) The valves are ideal switches.

2.3.1 Operation of the bridge

Fig. 2.8 shows the bridge which consists of six thyristors (SCR) 1 to 6 that are supplied
from a transformer for converting AC to DC and vice-versa. The valves fails to ignite if
the delay angle exceeds 180o . Referring to the Fig. 2.9, average direct voltage when delay
angle is equal to α is given by
Vd = Vdo cos α (2.1)

11
 Figure 2.8: Three phase, full wave Graetz circuit

√
3 2
where Vdo = π
ELL

Desired output voltage can be obtained by controlling firing angles i.e ignition delay an-
gle for rectifier, extinction advance angle for inverter or by varying converter transformer
tapings.

Due to the presence of AC source inductance, the transfer of current from one phase
to another phase requires some time is called commutation time. let us consider the
commutation from valve 1 to valve 3. Commutation begins at ωt=α and ends when
ωt=α+µ=δ during the period of commutation

di3 di3
eb − ea = Lc − Lc (2.2)
dt dt

since i1 = Id − i3
di1 di3
=0− (2.3)
dt dt
12
 Figure 2.9: Output voltage waveforms with firing angle delay

Then √
3Em
i3 = (cos α − cos ωt) (2.4)
2ωLc
The corresponding average voltage drop is

Vdo
∆Vd = (cos α − cos δ) (2.5)
2

from Eqn’s 2.4 and 2.5

Vd = Vdo cos α − Rc Id (2.6)

The RMS value of the fundamental frequency component of the alternating line current
is [6] √
6
I1 = Id (2.7)
π
If we neglect losses in the converter, ac power must be equal to dc power then power
factor of the fundamental wave is

cos φ = cos α (2.8)

So, when the converter acts as a rectifier or a inverter draws reactive power from the
system.

13
2.3.2 Rectifier operation

The equivalent circuit of the bridge rectifier based on the above analysis is shown in
figure 2.10.

Figure 2.10: Equivalent circuit of rectifier

2.3.3 Inverter operation

As valves conducts only in one direction reversal of power takes place with reversal of
direct voltage. An alternating voltage must exist on the primary side of the transformer
for inverter operation. The direct voltage of the inverter opposes the current, as in a dc
motor, and is called as counter voltage. The inverter operation also be described in terms
of α and δ as
β = π-α = ignition advance angle
γ = π-δ = extinction advance angle
µ = δ-α = β-γ = overlap
since cos α = -cos β and cos γ=-cos δ
The general converter equations can be written as

Vd = Vdo cos γ − Rc Id (2.9)

Figure 2.11 shows the inverter equivalent circuit

14
 Figure 2.11: Equivalent circuit of inverter

2.4 HVDC Transmission Advantages

Following are the main advantages of HVDC transmission compared to AC transmission

system [1], [2].
(1) A bipolar HVDC overhead line only requires two conductors with positive and nega-
tive polarities, thereby providing simple tower structure, low DC-line investment and less
power loss. Compared to a double circuit HVAC line with six conductor bundles, one
bipolar HVDC line with two conductor bundles takes much less the width of transmission
routine. Under the effect of direct voltage, the capacitance of transmission line is never
taken into account. Since capacitive current does not exist, direct voltage maintains the
same along the transmission line.

(2) For the AC and DC cables with the same insulation thickness and cross section,
the transmission capability for DC cable is considerably higher than that for AC cable.
DC cable lines only require one cable for monopolar link or two cables for bipolar link
and AC cable lines need three cables, due to three phase AC transmission. Therefore, the
price for DC cable lines is substantially lower than AC cable lines. Since there is no the
cable capacitance in a DC cable transmission, the transmission distance for DC cable is
unlimited theoretically.

(3) HVDC links can be used to interconnect asynchronous AC systems and the short
circuit current level for each AC system interconnected will not increase. The intercon-

15
nected AC systems can be operated with different nominal frequencies and the exchange
power between interconnected AC systems can be controlled rapidly and accurately.

(4) Due to the rapid and controllable features, HVDC systems can be used to improve the
performance of AC system, e.g. the stability of frequency and voltage, the power quality
and reliability of interconnected AC systems. For the DC/AC hybrid transmission sys-
tem, the rapid and controllable features of HVDC system can also be used to dampen the
power oscillations in AC systems, so as to increase AC lines transmission capacity.

(5) For an HVDC system, earth can be used as the return path with lower resistance, loss
and operational cost. For a bipolar link, earth is normally used as a backup conductor.
If faults occur on one pole, the bipolar link can be changed into the monopolar link au-
tomatically, thereby improving the reliability of HVDC system.

(6) An HVDC transmission system can also be used to link renewable energy sources,
such as wind power, when it is located far away from the consumer.

2.5 HVDC Transmission Disadvantages

Some of HVDC transmission disadvantages are,

(1) in a converter station, except for converter transformers and circuit breakers, there
are converter valves, smoothing reactors, AC filters, DC filters and reactive power com-
pensators. For the same rating, the investment for a converter station is several times
higher than the investment for an AC substation.

(2) a converter acts as not only a load or a source, but also a source of harmonic currents
and voltages, thereby distorting current and voltage waveforms.

(3) In a conventional converter station, the reactive power demand is approximately 60

percent of the power transmitted at full load. Since, reactive power must balance in-

16
stantaneously. Reactive power compensators must be installed in the converter station in
order to improve the stability of commutation and dynamic voltage.

(4) without current zero crossing point, DC circuit breakers are difficult to manufacture,
thereby developing multiterminal HVDC systems very slowly. With developing power
semiconductors with high switching frequency, DC circuit breaker can be innovated.

2.6 HVDC Transmission Applications

The first application for HVDC converters was to provide point-to-point electrical power
interconnections between asynchronous AC power networks. There are other applications
which can be met by HVDC converter transmission which include [1]

Long Distance and Bulk Capacity Transmission

For the same transmission capacity, above a certain distance, an HVDC transmission
offers more economic benefits than HVAC transmission. As the transmission distance
increases, the transmission capacity for HVAC line is restricted by stability limitation,
thereby necessarily increasing additional investment for short-circuit limitation, voltage
support.

Power System Interconnection

In order to optimize the resource utilization, several AC systems intend to be inter-

connected with the development of power industry, but it will give rise to the problems in
the super system. For example, the interconnection for AC systems always increases the
short-circuit levels, thereby exceeding the capacity of the existing circuit breakers. AC
systems can also be interconnected by HVDC transmission and thereby, not only obtains
the interconnection benefits but also avoids the serious consequences.

17
 DC Cable Transmission

For DC cable, without capacitance current, the transmission capacity is not restricted
by transmission distance. Except for the purpose of long distance and bulk capacity, DC
cables are also widely used across strait in the world. Due to environmental issue, large
capacity power stations are not allowed to build in the vicinity of city. Moreover, it is
very difficult to select appropriate the over headline routine, owing to high population
and load density. Therefore, using HVDC underground/submarine cables is an attractive
solution to deliver power from remote power station to urban load center.

Increasing the capacity of power transmission

It is some what difficult for new transmission rights of way. But, if we upgrade the
existing overhead AC transmission lines with DC transmission can substantially increase
the power transfer capability on the existing right of way.

2.7 Control of HVDC System

2.7.1 Basic principles of control

An HVDC transmission system is highly controllable. Its effective use depends on ap-
propriate utilization of controllability to ensure desired performance of the power sys-
tem.consider a monopolar HVDC link whose equivalent circuit is shown in Fig. 2.12.
Where subscript r and i refer to rectifier and inverter respectively.
The following issues are considered
(1) The transformers have variable transforming ratios. The effect of leakage reactance
on DC voltage was included through the commutation resistance Rcr on rectifier side and
Rci on inverter side.
(2) The DC overhead line is represented through resistance and its reactance is neglected.
A terminal which supplies power to DC link is termed as rectifier terminal and the

18
 Figure 2.12: Equivalent circuit of an monopolar HVDC link

terminal which takes power from DC link is termed as inverter terminal. The direction
of power flow can be changed by changing firing angle control.
The direct current flowing from the rectifier to the inverter in a monopolar HVDC link
can written as

Vdr cos α − Vdi cos γ

Id = (2.10)
Rdc
The power at the rectifier terminal is

Pdr = Vdr Id (2.11)

and at the inverter terminal

Pdi = Vdi Id = Pdr − RL Id2 (2.12)

The direct voltage at any point on the line and the current can be controlled by
controlling the internal voltages (Vdor cos α) and (Vdoi cos γ). This is accomplished by gate
control of valve ignition angle or control of the ac voltage through tap changing of the
converter transformer.

19
2.7.2 Basis for selection of controls

The following considerations influence the selection of control

1) Prevention of large fluctuations in direct current due to variations in ac system voltage.
2) Maintaining direct voltage near rated value
3) Maintaining power factors at the sending and receiving end that are as high as possible.
There are several reasons for maintaining the power factor high
(a) To reduce the stresses in the valves.
(b) To minimize losses and current rating of equipment in the ac system to which the
converter is connected.
(c) To minimize voltage drops st the ac terminals as loading increases and
(d) To minimize the cost of reactive power supply to the converters.

Rapid control of the converters to prevent large fluctuations in direct current is an

important requirement for satisfactory operation of the HVDC link. Referring to equation
(2.10), the line resistance is small. Hence, a small change in Vdr or Vdi causes a large change
in Id . It means even 25 % change in voltage at either the rectifier or inverter could cause
direct current to change by as much as 100%. This implies that, if both α and γ are kept
constant, the direct current can vary over a wide range for small changes in the alternating
voltage magnitude at either end. Such variations are not unacceptable for satisfactory
performance of the power system. In, addition to this the resulting current may be high
enough to damage the valves and other equipment. Therefore, rapid converter control
to prevent fluctuations of direct current is essential for proper operation of the system,
without such a control the HVDC system would be impractical.
For a given power transmitted, the direct voltage profile along the line should be close
to the rated value. This minimizes the direct current and thereby the line losses.

20
Chapter 3

Power Flow Studies

Power flow analysis aims at determination of system parameters like voltage, current,
power factor and power (real and reactive) flow at various points in the electric system
under existing conditions of normal operation. This analysis helps in determining the
scope of future expansion of the system. Power flows studies, commonly referred to as
load flow, are the backbone of power system analysis and design. They are necessary for
planning, operation, economic scheduling and exchange of power between utilities. In ad-
dition, power flow analysis is required for many other analyses such as transient stability
and contingency studies.

3.1 Significance of Load Flow Study

1. Determination of current, voltage, voltage angle, active power, reactive power etc. at
various buses in power system operating under normal steady state or static condition.

2. To plan best operation and control of existing system.

3. To plan future expansion to keep pace with load growth.

4. Help in ascertaining the effect of new load, new generating stations, new lines and

21
new interconnections before they are installed.

5. Due to this information system losses are minimized and also check is provided on
system stability.

6. Provides the proper prefault power system analysis to avoid system outage due to
fault.

3.2 Types of Buses

Four variables are associated with each node
a. Bus voltage magnitude (V )
b. Voltage angle (δ)
c. Real power (P)
d. Reactive power (Q)
Each node introduces two equations, namely the real and reactive power balance equa-
tions. To obtain solutions for a set of simultaneous equations, it is necessary to have the
same number of equations as unknowns. Therefore two of the variables associated with
each bus must be specified. The other two variables are free to vary during the solution
process. The traditional way of specifying bus-bar quantities allows buses to be identified
as follows

3.2.1 PQ bus-bar

At which the net active and reactive powers are specified. The net power entering a
bus-bar is the power supplied to the system from a generating source minus the power
consumed by a load at that bus-bar.

3.2.2 PV bus-bar

At which the net active power is specified, and the voltage magnitude is specified. The
net reactive power is an unknown which is determined as part of the power flow solution.

22
This type of bus-bar typically represents a node in the system at which a synchronous
source (generator or compensator) is connected, where the sources reactive power output
is varied to control the voltage magnitude to a scheduled value.

3.2.3 Slack or swing bus-bar

Where the voltage magnitude and angle are specified. Generally the angle is set to
zero. Unlike the other two bus types, which represent physical system conditions, this
bus-bar type is more a mathematical requirement. It is needed to provide a reference angle
to which all other angles are referred. Also, this bus absorbs any real power mismatch
across the system. Normally there can only be one slack bus-bar in the system. It is
generally chosen from among the voltage controlled bus-bars.

3.3 Formulation of Power Flow by Newton-Raphson

Method
Consider an n -bus power system contains a total np number of P-Q (load) buses
while the number of P-V (generator) buses be ng such that n = np + ng + 1. Bus-1 is
assumed to be the slack bus. The approach to Newton-Raphson load flow is similar to
that of solving a system of nonlinear equations using the Newton-Raphson method. At
each iteration formation of Jacobian matrix and solve for the corrections are given in Eq’n
3.1. For the load flow problem, this equation is of the form
   
∆δ ∆P2
 2  
 .   . 
   
   
 .   . 
   
   
h i  
∆δn   ∆Pn 

 ∆V2  = 
J  (3.1)
  

 |V |   ∆Q2 
 2   
   
 .   . 
   
   
 .   . 
   
∆Vn
|Vn |
∆Qnp

23
 Where
Real and Reactive power injections at each bus is given by

n
X
Pi = |Vi |2 Gii + |Yik Vi Vk | cos(θik + δk − δi ) (3.2)
k=1,k6=i

n
X
2
Qi = −|Vi | Bii − |Yik Vi Vk | sin(θik + δk − δi ) (3.3)
k=1,k6=i

and Jacobian matrix is divided into submatrices as

 
h i J11 J12
J =  (3.4)
J21 J22

The size of the jacobian matrix will be (n+np -1) × (n+np -1) and the submatrices are
 
∂P2 ∂P2
. .
 ∂δ2 ∂δn 
 
 . . . 
J11 =


 (3.5)
 . . . 
 
∂Pn ∂Pn
∂δ2
. . ∂δn

 
|V | ∂P2 . . |V1+np | |V∂P 2
 2 |V2 | 1+np | 
 
 . . . 
J12 =


 (3.6)
 . . . 
 
∂Pn ∂Pn
|V2 | |V2 | . . |V1+np | |V1+np |

 
|V2 | ∂Q 2
|V2 |
. . |V1+np | |V∂Q 2
1+np |
 
 
 . . . 
J21 =
 
 (3.7)
 . . . 
 
∂Qn ∂Qn
|V2 | |V2 | . . |V1+np | |V1+np |

24
  
∂Q2 ∂Q2
∂δ2
. . ∂δn
 
 
 . . . 
J22 =


 (3.8)
 . . . 
 
∂Q1+np ∂Q1+np
∂δ2
. . ∂δn

and Newton-Raphson procedure is as follows

Step-1: Choose the initial values of the voltage magnitudes |V |(0) of all np load buses
and n-1 angles δ (0) of the voltages of all the buses except the slack bus.

Step-2: Use the estimated |V |(0) and δ (0) to calculate a total n-1 number of injected
real power P calc(0) and equal number of real power mismatch ∆P (0) .

Step-3: Use the estimated |V |(0) and δ (0) to calculate a total np number of injected
reactive power Qcalc(0) and equal number of reactive power mismatch ∆Q(0) .

Step-3: Use the estimated |V |(0) and δ (0) to formulate the Jacobian matrix J (0) .

∆V (0)
Step-4: Solve (4.30) for δ (0) and |V (0) |
.

Step-5: Update the voltage magnitudes and voltage angles from

δ (1) = δ (0) + ∆δ (0)

∆|V |(0)
|V |(1) = |V |(0) [1 + ]
|V |(0)

25
Step-6: Check if all the mismatches are below a small number. Terminate the process
if yes. Otherwise go back to step-1 to start the next iteration with the updated values
obtained from step-5.

3.4 AC-DC Power Flow

Most of the solution techniques for AC/DC power flow are divided into two differ-
ent approaches [10], [11], [12]. The sequential and the unified (or simultaneous) solution
methods. The sequential solution methods, AC and DC system equations are solved sep-
arately in each iteration until the terminal conditions of converters are satisfied. Because
of modular programming, the sequential methods are generally easy to implement and
simple to incorporate various control specifications. In the unified methods, the AC as
well as DC system equations are combined together with the residual equations, describ-
ing the rectifier terminal behaviors, in one set of equations to be solved simultaneously.
The unified methods, with their better computing efficiency and convergence, seem more
suitable than the sequential methods for use in industrial AC/DC power systems, even if
they might be more complicated to program.

The Gauss-Seidel (G-S), the Newton-Raphson (N-R), and the fast decoupled N-R
methods may be used to solve the power flow problems in AC/DC systems as they do in
the pure AC systems. The G-S method generally needs accelerating factors to improve
the iteration process because of its slow rate of convergence. The N-R method, with
its powerful convergence characteristics, appears to be the most attractive technique in
solving the AC/DC power flows.

3.4.1 Point-to-point HVDC

When the AC system includes a HVDC link, say rectifier is connected at bus fr and
inverter is connected at bus to with transformer taps of 1:t1 and 1:t2 . Then the operating

26
condition of a an AC-DC power system is defined by the vector

T
[V̄ , θ̄, x̄] (3.9)

Where

V̄ is a vector of the voltages magnitudes at all AC system busbars

θ̄ is a vector of the angles at all AC system busbars (except the reference bus which
is assigned θ = 0 )

x̄ is a vector of DC variables

The equations of power balance at the terminal buses of the AC system are modified
by including the powers at the converter stations i.e., at rectifier end both active and
reactive powers are consumed and at the inverter end real power is injected and reactive
power is consumed [11], [12].

Then power mismatches at AC-DC buses are

∆Pf r = PG,f r − PL,f r − Pdc,f r (3.10)

∆Pto = PG,to − PL,to + Pdc,to (3.11)

∆Qf r = QG,f r − QL,f r − Qdc,f r (3.12)

∆Qto = QG,to − QL,to − Qdc,to (3.13)

where

PG , QG are function of AC system variables

PL , QL are usual AC system load at bus

Pdc , Qdc are function of DC system variables as

Pdc = f(Vdr , Vdi , x̄) (3.14)

Qdc = f(Vdr , Vdi , x̄) (3.15)

27
where

Vdr ,Vdi are the voltage magnitudes at the rectifier and inverter buses;

x̄ is the vector of independent variables of the HVDC system.

Because of introduction of the vector of unknowns x̄, supplementary equations de-

scribing the DC link operation and control strategy should be added as

R(Vdr , Vdi , x̄) = 0 (3.16)

So, the non linear system of equations use in the load flow calculation of mixed AC DC
system by using the extended variables method are
 
∆P (V̄ , θ̄)
 
 ∆P (V̄ , θ̄, [x̄])
 

 
=0 (3.17)
 
 ∆Q(V̄ , θ̄)
 
 
 ∆Q(V̄ , θ̄, [x̄]) 
 
R(Vdr , Vdi , x̄)

(A) At rectifier side

1) The DC voltage in terms of the AC source voltages referred to the transformer secondary
is √
3 2
Vdr = t1 Vf r cos α (3.18)
π
where Vf r is the busbar voltage on the system side of the converter transformer and α is
the gating delay for rectifier operation
2) Current flowing through the line is same as Eq’n 2.6.
3)The primary current drawn by the transformer at rectifier end is given by
√
6
I1 = t1 Id (3.19)
π

4)If we assume converter transformer is lossless and the magnetising admittance is ignored,

28
then the primary real power consumed by rectifier can be written as

Pdr = Vd .Id = Vf r I1 cos α (3.20)

5) Reactive power consumed consumed can be written as

Qdr = Vf r t1 I1 sin α (3.21)

The DC variables are

[x̄] = [Vdr , Id , t1 , cos α]t (3.22)

cos α is used as a variable than α is to linearize the equations.

(B) At inverter side

The equations presented above are equally applicable to inverter operation. But, during
inversion extinction advance angle γ controls the action instead of ignition delay angle α.
Therefore, DC voltage after inversion is
√
3 2
Vdi = t2 Vto cos γ (3.23)
π

(C) DC per unit base

Computational simplicity can be achieved by using common power and voltage base pa-
rameters on both sides of the converter. But to maintain consistency of power in per unit
√
the direct current base is 3 times more than the AC current base as
√
6
I1 = Id (3.24)
π

Translating above equation to per unit yields

√
6√
I1 (p.u.) = 3Id (p.u.) (3.25)
π

29
(D) Incorporation of control equations
In a AC - DC hybrid system, DC link operating condition is determined by means of the
values of the electrical quantities associated to the converter station which are grouped
into a vector
[X] = [Vdr , t1 , cos α, Vdi , t2 , cos γ, Id ]t (3.26)

As we have three equations like at the rectifier, at the inverter, the DC line configuration
and seven independent variables. So, four more equations are to be added for modeling the
control strategy of the DC link by specifying the four independent variables and also the
limit values (minimum and maximum limits ) and thus the operating mode is obtained.
Examples of valid control specifications are
(i) Specified converter transformer tap

t − tsp = 0 (3.27)

(ii) Specified DC voltage

Vd − V sp = 0 (3.28)

(iii) Specified minimum firing angle

cos α − cos αsp = 0 (3.29)

(iv) Specified DC power transmission

sp
Vd Id − PDC =0 (3.30)

Above control equations are simple and are easily incorporated into the solution algo-
rithm. During the iterative solution procedure the uncontrolled converter variables may
go outside (minimum or maximum limit ) pre-specified limits. When this occurs the of-
fending variable is usually held to its value equal to the exceeding limit and then next
operating mode is considered. Thus, Some of the operating modes are [12], [14]
(1) cos α Pdi cos γ Vdi

30
(2) cos α Pdi t2 Vdi
(3) t1 Pdi t2 Vdi
(4) t1 Pdi cos γ Vdi
(5) t1 Pdi cos γ t2
(6) t1 Pdi cos α cos γ
(7) t1 Vdr cos γ Pdi
Some more control modes can be obtained,
(A) if the current Id is kept constant instead power Pdi ;
(B) by keeping constant power Pdr instead of Pdi ;
(C) if the voltage at rectifier side i.e., Vdr in place of Vdi

The operating mode one corresponds to operation with constant angles α and γ,
ensured by transformer tap changers in order to maintain constant power and voltage
desired through DC line. Suppose, in this process if the transformer tap positions hits the
maximum or minimum limits then the tap changer variable will be fixed to the exceeded
limit and the operating mode of DC link will now shift to mode two or four depending
on the variable which is exceeded. Thus continuous operation of DC can be obtained.

Rectifier, has a minimum α limit of about 5o to ensure adequate voltage across the
valve before firing and it is normally operates at a value of α within the range of 15o to
30o so as to leave some room for increasing rectifier voltage to control DC power flow [7].

In case of an inverter, it is necessary to maintain a certain minimum extinction angle

[7] to avoid commutation failure. It is important to ensure that commutation is completed
with sufficient margin to allow for de-ionization before commutating voltage reverse at α
= 180o or γ = 0o . The extinction angle γ is equal to β − µ with overlap µ depending on Id
and the commutating voltage. Since a converter transformer has inductance, the trans-
former current cannot change instantly. The finite rate of change of current means that
the transfer of current from one valve to another requires a finite commutation time. The
volttime area A, which is shown in Figures 3.1 and 3.2. is required for the commutation.
The volt-time area A is related to the commutating current. The higher the commutating
current, the larger the volt-time area A will be. Typical full load values of µ are in the

31
range 200 to 500 under normal steady-state operation.
Commutation failures in HVDC systems are mainly caused by voltage dips due to AC
system faults. As indicated in Fig. 3.1 and Fig 3.2, voltage dips may cause both voltage
magnitude reduction and phase-angle shift [8]. Voltage dips may affect the commutation
in three ways:

1) Voltage magnitude reduction

Commutating ac line-to-line voltage decreases because of a voltage dip, as shown in Fig.(b)
in Fig. 3.1. Since the voltage magnitude has decreased, but the commutation area still
remain the same, so the end of commutation will be delayed and the extinction angle will
change from γ to γ́.

2) Phase-angle shift
The classification of three-phase voltage dips is shown in Fig.(a) in Fig. 3.1. Figure shows
that the phase-angle of the line-to-line voltage may shift either backward or forward during
voltage dips. The backward phase-angle shift affects the commutating process negatively.
If we assume that the firing instant does not change, although the volt-time area remains
the same, the final extinction is reduced from γ to γ́.

3) Increased dc current.
The dc current increases on the initiation of the fault at the inverter. Since the volt-time
area increases with the increased dc current, a relatively larger overlap µ will be needed
to complete the commutation. This will in the end reduce γ to γ́ in Fig.(c) in Fig. 3.2.

3.4.2 Multiterminal HVDC

It is increasingly being realized that multiterminal DC (MT HVDC) [15], [16] systems may
be more attractive in many cases to fully exploit the economic and technical advantages of
HVDC technology. When the power is to be transmitted from a large hydroelectric power
station to a distant load center, if the the overhead line passes through areas whose rising

32
Figure 3.1: Commutation failures due to voltage magnitude reduction and phase angle
shift

power demands could also be met by the new power station, then it can be solved by
arranging an HVDC converter station in series or parallel in path of the dc line. In some
cases combination of both series and parallel is used. It is also flexible to connect more
than two power stations. Consider an example of connecting four ac networks A through
D In comparison with separate HVDC point-to-point system, the HVDC multiterminal
network has the following basic advantages.
(1) The number of HVDC converter stations is lower.

Figure 3.2: Commutation failure due to increased in DC current

33
ȩnter

Figure 3.3: HVDC connections of several networks (a) By separate point-to-point HVDC’s
(b) By multiterminal HVDC
enter

(2) Transmission losses are lower since energy transport over more than two HVDC sta-
tions in series is avoided i.e., In case (a) in Fig. 3.3, the energy flow would take place
from A to D through six stations.
(3) In a mesh network (additional line from A to D), a current distribution is automati-
cally realized in the HVDC network which corresponds to minimal line losses.

In series connected multiterminal HVDC scheme, the converters are connected in se-
ries with a common direct current through all terminals. An auxiliary set up is required
during start-up. One disadvantage of this is that full I 2 R loss even at light load and
disadvantage can be somewhat overcome by decreasing the current instead of the voltage
in periods when all stations are at light load. Moreover, sometimes some converters may
be light loaded and others heavily loaded, the lightly loaded converters operate at low
voltage, which can be obtained either by low transformer tap or by operating at a large
control angle (α or γ). The first method requires a wide tap range, which increases the
cost of the converter transformer. The second method subjects the valves to large voltage
jumps, which increase the probability of converter faults, increase the losses in the valve
damping circuits, and increase the consumption of reactive power. This scheme is con-

34
fined to applications with small power taps where it may be more economical to operate
at higher current and lower voltage than for a full voltage tap at full voltage and reduced
current.
On the other side parallel connected multiterminal HVDC scheme, converters are con-
nected in parallel and operate at a common voltage is shown in Fig. 3.4 [16]. having three
terminals - one rectifier and two inverters is assumed. Rectifier currents are taken as pos-
itive and inverter currents are taken as negative. The stations having the lowest ceiling
voltage , i.e., the lowest Vdi cos γo controls the line voltage. This station is normally one
of the inverters operating at constant extinction angle. The other two stations operate
on constant current and at a voltages lower than the respective ceiling voltages.

If any of the converter is to be changed from a rectifier to an inverter or vice versa.

Its current must be brought back to zero, a reverse switching must operate. Finally, the
current must be increased to the desired value. This scheme is widely accepted compared
to series connected scheme.

Figure 3.4: Characteristics of multiterminal parallel DC lines

(A) Power flow equations

The formulation of power flow equations developed in point-to-point connection can be
easily extended to multiterminal also by simply adding network configuration equation,
rectifier and inverter equations.

35
3.5 Results and Discussions

3.5.1 Point-to-point DC transmission Line

A four bus AC system shown in Fig. 3.5 is modified to include a DC line between buses 2
and 4. The converters at bus 2 and bus 4 operate in the rectifier mode and in the inverter
mode respectively. The DC link data considered for simulation is given in Table 3.1.
∗ = intial value for power flow iteration.

Figure 3.5: Four bus system for AC-DC load flow

Table 3.1: Converter Data

Rectifier Inverter
Bus Number 2 4
Minimum control angle αmin = 5o γmin = 5o
Resistance of the DC line 0.3
Rated DC volatge(pu) 1.00∗ 1.00∗
Tap of transformer 0.5∗ 0.5∗

The number of possible modes in which HVDC can be operated are presented in Chapter
3.

36
The MATLAB simulation are carried out for Newton-Raphson AC-DC load flow, as ex-
plained in Section 3.4 for the case of single DC line and multiple DC lines. Lower limit
on firing angles of converters is taken as 50 . The lower and upper limits of transformer
tap positions located at converters are taken as 0.7 p.u and 1.2 p.u for point-to-point.
MATLAB code is written in such a way that it takes the priority of the transition modes.
Transition from one mode to another mode takes place when any unspecified variable
exceeds its limits. In the case of, unspecified variable exceeds its boundary limits it is
fixed to the exceeded limit and one of the specified variable is freed to vary. Hence, HVDC
link operating mode will be shifted to new mode provided which is specified in the priority.

Study Case (i):

Generally, for the sake of minimum reactive power requirement by the converters, con-
verters operate at minimum firing angles. with starting mode of specifying α at rectifier,
γ at inverter, for a fixed D.C voltage (Vdi ) and specified power (Pdi ) at inverter side. The
simulation results, as shown in Fig. 3.6 indicates requirement of reactive power at rectifier
side, inverter side and direct current flowing through D.C line. From the Fig 3.6. it is
clear that the power at inverter side varying 0.8 p.u to 2.3 p.u.
It is observed from simulations results that the tap position at rectifier side exceeds its
maximum limit when D.C line power is reached to 1.6 p.u. Here, the transition taken
place by keeping the rectifier tap changer position at its maximum limit and the variable
α is freed. The variables, t1max , γ, Vdi , Pdi , that are going to be fixed in new operating
mode after transition. In this mode firing angle, α is varied to deliver the required power
at inverter side. The second transition takes place when α reached its minimum value
for a power of 1.8 p.u. As depicted in Fig. 3.6, current through the DC line and reactive
power is abruptly changed to large value when the second transition occurs at a power of
1.8 p.u. Hence, it could be better to operate the DC link in either first mode or second
mode for specified power of the inverter.

Study Case (ii):

Instead of Pdi in case (i), simulations are carried out by specifying power at rectifier end.

37
Figure 3.6: Relation between reactive power at converters, Id versus power at inverter
side in case study case (i)

From Fig.3.7, it is inferred that reactive powers consumed at the converter stations, Id
are less in this case. Same thing is also observed for the following operating modes
i) α, γ, t1 , Pdi ;
ii) α, γ, t2 , Pdi ;
iii) α, γ, t1 , Pdr ;
iv) α, γ, t2 , Pdr ;

But the specified tap position in the above modes influence the peak value of the direct
current. Fig. 3.8 shows the reactive power values and line current for mode specifying α,
γ, t2= 0.8, Pdi . From this Fig. 3.8 it can be observed that peak value of the current is
more than the current in Fig. 3.7.

Study Case (iii):

There are some combinations of variables or modes for which the operation of the DC
link is difficult. For example, In mode α, t1, Vi and Pdi , guessing a correct initial specified

38
Figure 3.7: Relation between reactive power at converters, Id versus power at inverter
side in case study (ii)

Figure 3.8: Relation between reactive power at converters, Id versus power at inverter
side in study case (iii)

39
values of alpha (α) and t1 is difficult to satisfy both Eqn’s 3.31 and 3.32 simultaneously.

Vr − Vi
Id = (3.31)
Rdc

Pdi
Id = (3.32)
Vi
examples of such combinations are
i) t1 , Vr , Vi , Pdi /Pdr ;
ii) α, t1 , Vi , Pdi /Pdr ;
iii) γ, Vr , Vi , Pdi /Pdr ;
iv) γ, t2 , Vr , Pdi /Pdr ;
v) t2 , Vr , Vi , Pdi /Pdr ;

3.5.2 Multiterminal HVDC

Multiterminal series connection

The schematic diagram for series connection of DC lines is shown in Fig. 3.9. Fig. 3.9
shows that converters Rec, Inv, Inv1, Inv2 constitutes a series connection of multiterminal
HVDC. Simulations for series connection of multiterminal HVDC are shown in Fig. 3.10.

Figure 3.9: Schematic diagram of series connection

It shows the relation between tap positions of converter transformers and direct current
through the line for varying power at inverter one. From the Fig. 3.10, it is found that,

40
Figure 3.10: Relation between transformer tap positions at converters, Id and power at
inverter side in series connection

1) The current flowing through line is constant.

2) As the demand at inverters one is varying, tap position of transformer is also ad-
justed accordingly to meet the demand and the tap positions of inverter and inverter two
are constant.

3)As the rectifier has to supply the sum of loads at inverters, large change in tap po-
sitions are needed which in turn increase the cost of the transformers. Because of this,
series type multiterminal connection is preferred only for tapping of low power and this
problem can be eliminated using radial type connection.

Multiterminal radial connection

Converters Rec1, Rec2, Inv1, Inv2 in Fig. 3.11 constitutes a radial connection of multi-
terminal HVDC. Converters Rec1 and Rec2 are included at buses 2 and 4 respectively.

In the simulations two multiterminal HVDC configurations i.e., series and radial con-

41
 Figure 3.11: Schematic diagram of radial connection

Figure 3.12: Relation between transformer tap positions at converters, Id and power at
inverter side in radial connection

nections are studied. From the simulation shown in Fig. 3.12, point out that the affect
of the configurations on the taps of the transformer and in turn the effect of on the cost
of the transformer.

From Fig. 3.12, it clearly indicate that for the radial configuration the range of the
taps required on the transformer is less than in case of the series configuration. In both

42
cases the load variation are kept same. So, this indicates that the case discussed, the cost
of transformer is more for series configuration than the radial configuration.

43
Chapter 4

Optimization of Power Flow

The optimal Power Flow (OPF) problem is a powerful tool for power system operation
and planning. In general, OPF problem is a nonlinear programming (NLP) problem that is
used to determine the ”optimal” control parameter settings to minimize a desired objective
function, subject to certain system constraints. The OPF problem represents a variety of
optimization problems include fuel or active power cost optimization, active power loss
minimization, maximum transfer capability, MVAR cost minimization, minimum of total
emission.

4.1 Conventional Optimal Economic Scheduling

Conventional optimal economic scheduling (Economic Load Dispatch activity) [19] is exe-
cuted in the dispatch stage. Its primary function involves allocating the total load between
the available generating units in such a way that the total cost of operation is kept at
minimum. Total cost may be approximated by a variety of expressions such as linear or
quadratic functions of the active power generation of the unit. The total active power
generation in the system must equal the load plus the active transmission losses.
The objective function is augmented by the constraints using a Lagrange-type multi-
plier lambda, λ.
∂Fi h
∂PL
i
=λ 1− i = 1, ..., N
∂Pi ∂Pi

When the generator’s limits are not binding, The optimal solution to the ELD problem

44
is obtained when all units operate at the same value of λ i.e., equal incrementing cost
operation yields the optimal dispatch. The optimality conditions along with the physi-
cal constraints are a set of nonlinear equations that requires iterative methods to solve.
Newtons method has been widely accepted to solve problems such as the load flow and
optimal load flow.
A solution can usually be obtained within a few iterations, provided that a reasonably
good initial estimate of the solution is available.

4.2 Formulation of OPF problem

In the Economic Load Dispatch(ELD), The demand is considered to be an aggregate
parameter for the entire system. When load flow equations are introduced in ELD as
a system of demand supply balance constraints, the optimum solution yields a set of
decision variables satisfying the physical laws of flow of electricity while achieving a desired
objective (of cost minimization, loss minimization ....) such a formulation is called optimal
power flow (OPF). It is a static, constrained, nonlinear, optimization problem. Due
to the presence of the detailed network configuration instead of lumped formulation of
ELD, The demand is now disaggregated and is available at all buses individually. As the
load flow equations include a reactive power balance at each node. The OPF has the
additional advantage of considering reactive power as a decision variable, this leads to
reactive power planning problem. The optimal power flow is a constrained optimization
problem requiring the minimization of [20], [21]

f = (x, u) (4.1)

subject to
g(x, u) = 0 (4.2)

h(x, u) ≤ 0 (4.3)

umin ≤ u ≤ umax (4.4)

45
 xmin ≤ x ≤ xmax (4.5)

Here, f (x, u) is the scalar objective function, g(x, u) represents nonlinear equality con-
straints (power flow equations), and h(x, u) is the nonlinear inequality constraint of vector
arguments x and u. The vector x contains dependent variables consisting of bus voltage
magnitudes and phase angles, as well as the MVAr output of generators designated for
bus voltage control and fixed parameters such as the reference bus angle, noncontrolled
generator MW and MVAr outputs, noncontrolled MW and MVAr loads, fixed bus volt-
ages, line parameters, etc. The vector u consists of control variables including real and
reactive power generation, phase shifter angles, DC transmission line flows, control volt-
age settings.
The power system consists of a total of N buses, NG of which are generator buses. M buses
are voltage controlled, including both generator buses and buses at which the voltages
are to be held constant. The voltages at the remaining (N - M) buses (load buses), must
be found. The network equality constraints are represented by the load flow equations

4.2.1 Objective function

A common objective function used in OPF studies is the minimization of generation costs.
The objective function based on generation operation cost can be expressed as

NG
X
Z= Ci (Pi ) (4.6)
i=1

where Ci is generator’s fuel cost function, which is usually quadratic and can be written
as
Ci = ai Pi2 + bi Pi + ci

and NG is the set of all generating units including the generator at the slack bus.

4.2.2 Network equations

The network equations are obtained from the basic kirchoff ’s laws governing the loop
flows and nodal power balance as follows

46
 X
term
Pi −P Di ±PDC = |Vi ||Vj |(Gi,j cos(δi −δj )+Bi,j sin(δi −δj )) ∀i = 1, ..N ; j 6= slack
j
(4.7)
X
Qi − QDi − Qterm
DC = |Vi ||Vj |(Gi,j sin(δi − δj ) − Bi,j cos(δi − δj )) ∀i = 1, ..N L (4.8)
j

Where V is the bus voltage, δ is the angle associated with V, Gi,j and Bi,k are conduc-
tance and susceptance elements of bus admittance matrix, P and Q are real and reactive
power generation respectively, PD and QD are real and reactive power demand respec-
term
tively and NL is the number of P-Q buses, PDC and Qterm
DC are the real and reactive

power associated with the converters and the sign ’+’ is used when the converter at
the corresponding bus is operating as inverter and ’-’ is used when the converter at the
corresponding bus is operating as rectifier.

4.2.3 Generation limits

Generator limits are

PiM in ≤ Pi ≤ PiM ax ∀i ∈ N G (4.9)

QM
i
in
≤ Qi ≤ QM
i
ax
∀i ∈ N G (4.10)

P M in and P M ax are the upper and lower limits on real power generation and QM in and
QM ax upper and lower limits on reactive power generation from a unit.

4.2.4 Bus voltage limits

This constraint ensures that the voltages at different busses in the system are maintained
at specified levels. The generator bus (or PV bus) voltages are maintained at a fixed level.
Voltage level at a load bus is maintained within a specified upper limit V M ax and a lower
limit V M in , determined by the operator.

|Vi | = constant, ∀i = 1, ...., N G (4.11)

47
 ViM in ≤ |Vi | ≤ Vimax , ∀i = 1, ...., N L (4.12)

4.2.5 Limits on reactive power Support

This constraint may be required in case the system operator has to include decisions on
optimal reactive switching at load buses.

QGM
i
in
≤ QGi ≤ QGM
i
ax
∀i ∈ N L (4.13)

QGM in and QGM ax are the limits on bus reactive power support.

4.2.6 Limits on power Flow

Transmission lines are limited by their power carrying capability, which is determined by
the
|Vi ||Vj |
Pi,j = sin(δi − δj ) (4.14)
Xi,j
M ax
Pi,j ≤ Pi,j (4.15)

M ax
Pi,j is the power flow over a line i-j and Pi,j is the maximum limit on power flow over
the line.

4.2.7 DC terminal voltage equations

Using the per unit system, the average value of the DC voltage of a converter connected
to bus fr is

√
3 2
Vdr = t1 Vf r cos α (4.16)
π

and incase if the converter is operating as inverter then the DC voltage can be written as
√
3 2
Vdi = t2 Vto cos γ (4.17)
π

48
4.2.8 Firing angle limits

These constraints ensure that the limits on the ignition delay angle and the extinction
advance angle of the rectifier and the inverter are maintained within specified limits.

cos α ≤ cos αmax ; cos α ≥ cos αmin

As mentioned earlier adequate α is required so as to provide sufficient voltage across

thyristor at the time of firing. similarly at inverter side

cos γ ≤ cos γ max ; cos γ ≥ cos γ min

4.2.9 Transformer tap changer limits

There will be a minimum and maximum limit on the transformer tap changer settings
above which may leads to infeasible solution or not possible because of some constraints.

t1 ≤ tmax
1 ; t1 ≥ t1min

t2 ≤ tmax
2 ; t2 ≥ t2min

4.3 Characteristic Features of OPF

The main features of an OPF is its ability to include the detailed network configuration
and bus-wise demand balance for both active and reactive power. The OPF can include
many operating constraints and model other issues also. Specifically limits on reactive
power generation in addition to real power generation, power flow limits on the transmis-
sion lines and limits on the bus voltages ensure that the system is operated in a secure
manner.

49
4.4 OPF Applications
One of the main feature of the OPF is that it is a flexible analytical tool, which allows the
use of different objective functions to solve different problems. The objective functions
commonly used for operation and planning studies are as follows
• Minimize cost of operation
• Minimize the deviation or minimize control shift
• Loss minimization
• Minimize the cost of installation of new capacitors and reactors and/or cost of MVAr
supplied
• Maximum transfer capacity
• Minimize the emissions, and so on.

4.5 Results and Discusions

A 14 bus system is considered for AC-DC OPF simulations. The simulation study involves
the use of MATLAB and GAMS for optimizing the power flow. GAMS is mainly used
for optimization of an objective function with respect to a set of constraints. It models
and solves complex linear, nonlinear and quadratic problems. Here point-to-point and
multiterminal configurations have been considered. Table 4.1 to Table 4.3 indicates the
general data for all HVDC configurations.

Line data and bus data of the system considered are shown in Table 4.1 and Table
4.2 respectively. The generator’s fuel cost function coefficients is shown in Table 4.3. The
lower limit on generator output is considered as 15 MW. Lower limit on firing angles of
converters is taken as 15o . The lower and upper limits of transformer tap positions lo-
cated at converters are taken as 0.7 p.u and 1.3 p.u for point-to-point, multiterminal radial
and 0.6 p.u and 2.0 p.u for multiterminal series connection. The input to the GAMS is
passed from a MATLAB program through the MATGAMS interface. After optimizing the
given objective function w.r.t constraints in GAMS, the result can be obtained MATLAB.

50
 Programs have been written in GAMS software for the optimization of generation cost
of power producing for the case of point-to-point DC transmission line and multiterminal
HVDC line. Here equality and inequality constraints have been considered as discussed
in Section 4.2. The written GAMS programs have been interfaced with MATLAB

Table 4.1: Line Data

From Bus To Bus R (p.u) X (p.u) B/2 (p.u) Flow Limit (p.u)
1 2 0.01938 0.05917 0.0264 3
1 5 0.05403 0.22304 0.0246 3
2 3 0.04699 0.19797 0.0219 3
2 4 0.05811 0.17632 0.0170 3
2 5 0.05695 0.17388 0.0173 3
3 4 0.06701 0.17103 0.0064 3
4 5 0.01335 0.04211 0.0 3
4 7 0.0 0.20912 0.0 3
4 9 0.0 0.55618 0.0 3
5 6 0.0 0.25202 0.0 3
6 11 0.09498 0.19890 0.0 3
6 12 0.12291 0.25581 0.0 3
6 13 0.06615 0.13027 0.0 3
7 8 0.0 0.17615 0.0 3
7 9 0.0 0.11001 0.0 3
9 10 0.03181 0.08450 0.0 3
9 14 0.12711 0.27038 0.0 3
10 11 0.08205 0.19207 0.0 3
12 13 0.22092 0.19988 0.0 3
13 14 0.17093 0.34802 0.0 3

In Table. 4.2, bus type 1,2 and 3 indicates slack bus, PV (generator) and PQ (load
bus) respectively.

4.5.1 Point-to-point DC transmission line

An HVDC line is included between buses 7 and 10. The converters at bus 7 and at bus 10
operate in rectifier and inverter mode respectively. Simulations result for point-to-point
DC line are shown in Table 4.4 to Table 4.6.
Table 4.4 indicates optimized generator outputs, bus voltage magnitudes and angles.

51
 Table 4.2: Bus Data

Bus Type Vsp δ PGi QGi PLi QLi Qmin Qmax

(p.u) (radians) (MW) (MVAR) (MW) (MVAR) (MVAR) (MVAR)
1 1 1.020 0 0 0 14 0 -10 20
2 2 1.025 0 100 42.4 21.7 12.7 -50 40
3 2 1.010 0 100 23.4 94.2 19.0 -60 100
4 2 1.045 0 90 12.2 11.2 7.5 0 100
5 2 1.010 0 60 17.4 0.0 0.0 -60 10
6 3 1.0 0 0 0 47.8 -3.9 0 0
7 3 1.0 0 0 0 7.6 1.6 -50 300
8 3 1.0 0 0 0 0.0 0.0 0 0
9 3 1.0 0 0 0 29.5 16.6 0 0
10 3 1.0 0 0 0 9.0 5.8 -300 300
11 3 1.0 0 0 0 3.5 1.8 0 0
12 3 1.0 0 0 0 6.1 1.6 0 0
13 3 1.0 0 0 0 13.5 5.8 0 0
14 3 1.0 0 0 0 14.9 5.0 0 0

Table 4.3: Coefficients for Cost Functions of Generators

Bus No. a ($/M W 2 ) b ($/MW) c ($)

1 0.0095 10 150
2 0.0095 15 220
3 0.0095 17 230
4 0.0095 14 200
5 0.01 18 200

This table shows that all bus voltage values are within permissable limits of ± 5% ac-
cording to the standards and generators real power are within limits.
The optimized values of firing angles, transformer tap positions, current through the
DC line are shown in Table 4.5 for the specified values of voltage (Vi ) and real power (Pdi )
at inverter side.
Table 4.6 is presented to show the status of the optimization. In this table model
status 2 indicates the objective is locally optimal and solver status indicates the code has
been compiled normally. It also indicates the optimized cost for all generator is 1030.01
$/MW.
Hence, from the above discussion it is clear that the written code for the optimization

52
 Table 4.4: Point-to-point DC Transmission Result

Generator output Bus voltage magnitude Bus voltage angles

(p.u) (p.u) (radians)
2.0368 1.0200 0.0843
0.1500 1.0250 0.0000
0.1500 1.0100 -0.1108
0.1500 1.0450 -0.0854
0.1500 1.0100 -0.0520
0 1.0354 -0.1887
0 0.9500 -0.2140
0 0.9500 -0.2140
0 1.0326 -0.1768
0 1.0500 -0.1309
0 1.0423 -0.1624
0 1.0270 -0.2053
0 1.0252 -0.2065
0 1.0191 -0.2116

Table 4.5: DC Variables in Point-to-point Connection

Alpha Gamma t1 t2 Id Vr Vi Pdi

(Degrees) (Degrees) (p.u) (p.u) (p.u) (p.u) (p.u) (p.u)
14.9943 14.9943 0.9858 0.8764 0.7083 1.2212 1.2000 0.8500

Table 4.6: GAMS Status and Objective Value

Modelstatus Solvestatus Objective

2 1 1030.01

has been achieved successfully.

4.5.2 Multiterminal HVDC

The optimization code has been written for the two configurations of multiterminal HVDC
as discussed in Chapter 2.
1) Multiterminal series connection
Four converters are considered in multiterminal series connection which is discussed in
Section 3.5. Rectifier (Rec) and inverter (Inv) are connected at buses 7 and 10 respectively.

53
 Table 4.7: Multiterminal Series Connection Result

Generator output Bus voltage magnitude Bus voltage angles

(p.u) (p.u) (radians)
2.3000 1.0200 0.0930
0.2725 1.0250 0.0000
0.1500 1.0100 -0.1208
0.9000 1.0450 -0.1043
0.1500 1.0100 -0.0703
0 1.0473 -0.2522
0 1.0500 -0.3614
0 0.9500 -0.3614
0 1.0316 -0.2939
0 1.0500 -0.2419
0 1.0474 -0.2501
0 1.0388 -0.2723
0 1.0350 -0.2773
0 1.0226 -0.3084

Table 4.8: DC variables in Multiterminal Series Connection

Alpha Gamma trec tinv tinv1 tinv2 Id Vr Vi Pdi

(Degrees) (Degrees) (p.u) (p.u) (p.u) (p.u) (p.u) (p.u) (p.u) (p.u)
14.9943 14.9943 1.9294 0.8034 0.6000 0.6000 0.7273 2.6418 1.1000 0.8000

Table 4.9: GAMS Status and Objective Value

Modelstatus Solvestatus Objective

2 1 1045.00

Tables 4.7, Table 4.8 and Table 4.9 show the simulation results for the OPF for mul-
titerminal series connection. Tables 4.7 indicates that all bus voltages and generators
outputs are in well-preserved limits. The optimized values of firing angles, transformer
tap positions and current through the DC line are shown for the specified voltage and
power at inverter are shown in Table 4.8. Table 4.9 indicates the written program in
GAMS executed successfully and the optimized cost is found to be 1045.00 $/MW.

54
2) Multiterminal radial connection
Four converters are considered in multiterminal radial connection which is discussed in
Section 3.5. Rectifiers, Rec1 and Rec2 are connected at buses 7 and 10 respectively.

Table 4.10: Multiterminal Radial Connection Result

Generator output Bus voltage magnitude Bus voltage angles

(p.u) (p.u) (radians)
2.3000 1.0200 0.0872
0.7275 1.0250 0.0000
0.1500 1.0100 -0.1345
0.9000 1.0450 -0.1304
0.1500 1.0100 -0.0972
0 1.0500 -0.3393
0 1.0500 -0.3941
0 0.9500 -0.3941
0 1.0324 -0.3978
0 1.0245 -0.4145
0 1.0350 -0.3809
0 1.0417 -0.3607
0 1.0374 -0.3668
0 1.0241 -0.4059

Table 4.11: Firing Angles of Converters in Multiterminal Radial Connection

Alpha1 Alpha2 Gamma1 Gamma2

(Degrees) (Degrees) (Degrees) (Degrees)
14.9943 14.9943 14.9943 14.9943

Table 4.12: Transformer Tap positions at Converters in Multiterminal Radial Connection

trec1 trec2 tinv1 tinv2

(p.u) (p.u) (p.u) (p.u)
0.8332 0.8540 0.9373 0.9373

Tables from 4.10 to 4.14 indicate the simulation results for the OPF for multiterminal
radial connection. From these tables, it is observed that the voltages and output of

55
 Table 4.13: Voltages, Currents and Real Powers in Multiterminal Radial Connection

Ir1 Ir2 Ii1 Ii2 Vr Vi Pdi1 Pdi2

(p.u) (p.u) (p.u) (p.u) (p.u) (p.u) (p.u) (p.u)
1.1007 0.2629 0.7273 0.6364 1.1409 1.1000 0.8000 0.7000

Table 4.14: GAMS Status and Objective Value

Modelstatus Solvestatus Objective

2 1 1051.83

generators are within boundary limits. Hence, optimization is achieved successfully with
optimization cost of 1051.83 $/MW.

56
Chapter 5

Conclusions and Future Scope of

Research

5.1 Conclusions

There has been a growing interest in HVDC technology since the first HVDC installation
took place in Gotland, sweeden, in 1954. The high flexibility of control available with
HVDC transmission results in a number of new advantages and applications which have
been presented in Chapter 2. The main objective of the thesis is to perform load flow and
optimal power flow analysis of an AC-DC system. Here, load flow analysis for different
configurations of single and multiple DC transmission line are discussed.

MATLAB simulations are carried out for point-to-point DC transmission for various
operating modes. These modes are defined by considering any three out of the seven DC
variables specified for a particular power transfer through the DC line. The respective
modes of operation are discussed in detail in Section 3.4. Transition from one mode to
another mode takes place if any unspecified variable exceeds its boundary limits. Then,
HVDC operating mode shifts to a new mode. Under the transition between modes, the
variable which exceeds its operating limit is fixed at that limit and any one of the fixed
variables in the transiting mode is made free to be varied.
The following results are observed from the case studies carried out.

57
(1) The reactive power consumed by converters is minimum when HVDC is operating in
modes where αmin , γmin , Vdi /Vdr /t1 /t2 are specified. Also, current flowing through the
DC line depends on the specified value of tap position of the converter transformer.

(2) HVDC operation is practically not feasible for some modes. This is explained in
detail in Section 3.5.

(3) Simulations also involve the transition of operating modes under the case of prior-
ities of them given as one of the inputs. The transition between operating modes took
place successfully.
Simulations for series and radial configurations of multiterminal HVDC line are also
carried out. In case of the series configuration, the range of transformer tap positions
required is significantly greater than that in case of radial configuration. This directly
effects the cost of transformer and it is explained in detail in Section 3.5. In case of any
converter failure, partial power can be transferred using a radial configuration, whereas
it leads to complete interruption of power in case series configuration. Therefore, radial
configuration is much superior and is preferred in practical systems.
Apart from simple load flow studies, an optimal power flow study is also carried
out for the AC-DC system. The OPF model is developed by using a generalized matrix
representation of the HVDC link. The OPF calculation is performed by using GAMS. The
input to the GAMS program is passed from a MATLAB program through the MATGAMS
interface. Simulation results validated the satisfactory operation of an AC-DC system such
as, maintaining voltages at all buses within 5 % tolerance level, transferring line power
within acceptable limits. Moreover, it also ensures the correct direction of converter
currents.

5.2 Future Scope of Research

In the present work, steady state analysis of AC-DC system is done. Based on the work
done, it can be concluded that a number of applications and advancements in HVDC

58
transmission system are required, which are mentioned below.

1. Analysis of transient behavior of interconnected AC-DC system under switching and

fault conditions can be done.

2. A generic concept of modeling of all types of HVDC systems considering all the pos-
sible configurations is needed. This can reduces the complexity involved in analyzing
the power system.

3. Effective utilization of VSC in an interconnected AC-DC system can be sought for

the integration of DC sources with the utility grid.

59
Bibliography

[1] K. Meah and Sadrul Ula, “Comparative Evaluation of HVDC and HVAC Transmis-
sion Systems,” in Power Engineering Society General Meeting, pp. 1-5, 2007.

[2] Prabha Kundur, Power System Stability and Control. New York: Tata Mcgraw-Hill
Edition, pp. 463-579. 1994.

[3] J. Arrillaga, Y. H. Liu, and N. R. Watson, Flexible Power Transmission: The HVDC
Options. England: John wiley and sons Ltd, pp. 21-95.

[4] J. Setreus and L. Bertling, “Introduction to HVDC Technology for Reliable Electrical
Power Systems,” in Probabilistic Methods Applied to Power Systems, pp. 1-8, 2008.

[5] Y. Tzeng, Nanming Chen, and Ruay-Nan Wu, “A Detailed R-L Fed Bridge Con-
verter Model for Power Flow Studies in Industrial AC/DC Power Systems,” IEEE
Transactions on Industrial Electronics, vol. 42, no. 5, pp. 531-538, October 1995.

[6] E. W. Kimbark, Direct Current Transmission. New York: wiley interscience, vol. 1,
pp. 21-95, 1971.

[7] P. G. Engstrom, “Operation and Control of HVDC Transmission,” IEEE Transac-

tions on Power Apparatus and Systems, vol. 83, no. 1, pp. 71-77, 1964.

[8] Lidong Zhang and L. Dofnas, “A novel method to mitigate commutation failures in
HVDC systems,” in Power System Technology, vol. 1, pp. 51-56, 2002.

[9] D. A. Braunagel, L.A. kraft, and J.L. Whysong, “Inclusion of DC converter and
transmission equations directly in a Newton power flow,” IEEE Transactions on
Power Apparatus and Systems, vol. 95, no. 1, pp. 76-88, 1976.

60
[10] H.A. Sanghavi and S. K. Banerjee, “Load Flow Analysis of AC - DC Power Systems,”
in TENCON’ 89, pp. 746-751, 1989.

[11] J. Arrillaga and B. Smith, AC-DC power system analysis. London, United Kingdom:
IET, pp. 61-107, 1998.

[12] F. Coffele, R. J. Garcia-Valle, and E. Acha, “The Inclusion of HVDC Control Modes
in a Three-Phase Newton-Raphson Power Flow Algorithm,” in Power Tech, pp. 419-
424, 2007.

[13] T. Smed, G. Andersson, G.B. Sheble, and L. L. Grigsby, “A new approach to AC/DC
power flow,” IEEE Transactions on Power Systems, vol. 6, no. 3, pp. 1238-1244, 1991.

[14] A. Ekstrom, “A Refined HVDC Control System,” IEEE Transactions on Power

Apparatus and Systems, vol. 89, no. 5, pp. 723-732, 1970.

[15] J. Reeve, “Multiterminal HVDC Power Systems,” IEEE Transactions on Power Ap-
paratus and Systems, vol. 89, no. 2, pp. 729-737, 1980.

[16] R. Foerst, G. Heyner, K. W. Kanngiesser, and H. Waldamnn “Multiterminal Oper-

ation of HVDC Converter Stations,” IEEE Transactions on Power Apparatus and
Systems, vol. 88, no. 7, pp. 1042-1052, 1969.

[17] V. F. Lescale, A. Kumar, L. E. Juhlin, H. Bjorklund, and K. Nyberg, “Chal-

lenges with Multi-Terminal UHVDC Transmissions,” in Power System Technology
and IEEE Power India Conference, pp. 1-7, 2008.

[18] J. P. Bowles, H. L. Nakra, and A. B. Turner, “A Small Series Tap on an HVDC Line,”
IEEE Transactions on Power Apparatus and Systems, vol. 100, no. 2, pp. 857-862,
1981.

[19] K. Bhattacharya, M. H. J. Bollen, and J. Daalder, Operation of Restructured Power

Systems. New York: Kluwer Academic Publication, pp. 30-40, 2001.

61
[20] B. Warkad, M. K. Khedkar, and G. M. Dhole “A genetic algorithm approach for
solving AC-DC optimal power flow problem,” Journal of Theoretical and Applied
Information Technology, vol. 16, no. 1, pp. 027-039, 1981.

[21] C. N. Lu, S. S. Chen, and M. Ong, “The Incorporation of HVDC equations in

optimal power flow methods using sequential quadratic programming techniques,”
IEEE Transactions on Power Systems, vol. 3, no. 3, pp. 1005-1011, August 1988.

62