2017

Power System
Analysis and Design
Lecture Note
This document is prepared for teaching the course Power System Analysis and
Design at Faculty of International Training, Thai Nguyen University of
Technology, Thai Nguyen, Viet Nam

Tran Thai Trung, MSc.

Nguyen Minh Y, PhD.
Thai Nguyen University of Technology
2/15/2017
 Thai Nguyen University of Technology

Faculty of International Training

Lecture Note
Power System Analysis and Design

(EE-0015)

By

Tran Thai Trung, MSc.

Nguyen Minh Y, PhD.

Dept. of Electrical Engineering taught in Faculty of International Training

English

MSc. Vu Ngoc Huy MSc. Nguyen Tien Hung

Thai Nguyen, 2017

1
 Syllabus
Lecturer

 Nguyen Minh Y

Email: [email protected]

Office room: A2 – 309.

Office hours: Wed. and Fri. 08:00am – 11:00am (or by appointment)

Lecture schedule

TA

Website

 http://www.elearning.tnut.edu.vn

Objectives

Textbook

1) Glover/Sarma, Power system analysis and design, 4rd edition, PWS.

2) Bergen/Vittal, Power systems analysis, 2nd edition, Prentice Hall.

References

Quizzes:

 There will be 15-minute biweekly quizzes starting in the second week.

Midterms:

 There will be 1-hour closed-book midterm test.

Finals:

 There will be two-hour closed-book finals at the end of the semester.

Homework

 Each week we will assign you a few questions and problems as homework. Your
answer and solutions for each homework will be scored.

 We will be posting the solutions to homework on our website.

2
  The above grading policy should make clear that we do not intend the homework as
tests, but as vehicles for learning. You can expect that every quiz will include
problems of the same flavor and difficulty as those encountered in the homework,
but sufficiently modified to test your thinking and understanding, rather than your
ability to “pattern match”.

Course Grade

The final grade in the course will be based on our best assessment of your understanding of the
material and your participation in the course. The approximate relative weighting given to
different components of the course in arriving at a preliminary grade will be:

 Home works: 20%

 Presentation: 10%

 Midterm exams: 30%

 Final exam: 30%

 Attendance and award: 10%

Factors such as your interaction with the staffs and your participation in lectures can also
affect the final grade, particularly if your preliminary grade falls near a borderline.

The process of assigning a final grade involves considerable discussion between the staffs,
and very often involves a careful review of the quizzes to look behind the numbers and
understand better the kinds of mistakes that were made. We know that the final grade is
important to you, and we take the process seriously.

3
TABLE OF CONTENT

4
 CHAPTER I
THE POWER SYSTEM: AN OVERVIEW

In this chapter we introduce the summary of history of power system and a simplified
description of a power system. We also give concepts of the typical components of power
systems such as power sources (or generators), power end users (or customers), and delivery
systems (including transmission and distribution system).

1.1. ELECTRIC ENERGY

Electricity is only one of many forms of energy used in industry, homes, businesses, and
transportation. It has many desirable features such as clean (particularly at the point of use),
convenience, easy to transfer from point of source to point of use and high flexibility. In some
cases, it is an irreplaceable source of energy.
Figure 1.1 is a useful summary of the total World energy consumption by electric energy
sources in 2012. It can be seen that oil, coal, and natural gas comprised 87% of global energy
consumption. Oil accounted for 33% of energy consumption worldwide and remains the world’s
leading energy source. Coal made up 30% while gas comprised 24% of the total energy
consumption. The growth of nuclear capacity has been halted by rising construction costs,
licensing delays, and public opinion. Although there are no emissions associated with nuclear
power generation, the safety and environmental issues, such as the disposal of used nuclear fuel
and the impact of heated cooling – tower water on aquatic habitats are the topics of universal
interest. Renewable sources include conventional hydroelectric (water power), geothermal,
landfill gas, other biomass, solar and wind power. Renewable sources of energy cannot be
ignored, but they are not expected to supply a large percentage of the world’s future energy
needs because of their extremely high construction and operation costs.

Fig. 1.1: World energy consumption by source, 2012.

5
1.2. HISTORY OF POWER SYSTEM
In 1878, Thomas A. Edison began work on the electric light and formulated the concept of
a centrally located power station with distributed lighting serving a surrounding area. 1982
marked the beginning of the electric utility industry (see Figure 1.2).
The first electric network was established in the United States in 1882 at the Pearl Street
Station in New York City by Thomas A. Edison. The power was generated by DC generators
and distributed by underground cables. In the same year, the first water wheel driven generator
was installed Appleton, Wisconsin. Within a few years, many companies were established
producing energy for lighting – all operated under Edison’s patents. Because of the excessive
power loss, RI2 at low voltage, Edison’s companies could deliver energy only a short distance
from their stations.

Fig. 1.2: Milestones of the early utility industry.

With the invention of the transformer (William Stanley, 1885) to raise the level of AC
voltage for transmission and distribution following with the invention of the induction motor
(Nikola Tesla, 1888) to replace the DC motors, the advantages of the AC system became
apparent, and made the AC system prevalent.
The first single-phase AC system was at Oregon City where power was generated by two
300 hp waterwheel turbines and transmitted at 4 kV to Portland. In Germany, the first three-
phase AC system was installed in 1891, while Southern California Edison Company built the

6
first one in United States in 1893. In the beginning, individual companies were operating at
different frequencies anywhere from 25 Hz to 133 Hz. But, as a need for interconnection ad
parallel operation became evident, a standard frequency of 60 Hz was adopted throughout the
U.S and Canada and of 50 Hz was selected in most European countries.

1.3. POWER SYSTEM

The power system of today is a complex interconnected network as shown in Figure 1.3. A
power system can be subdivided into four major parts:
- Generation
- Transmission and Sub-transmission
- Distribution
- Loads

Fig. 1.3: Basic components of a power system.

1.3.1. Generation
Generators – One of the essential components of power systems is the three-phase AC
generator known as synchronous generator or alternator. It is a rotating machine that converts
mechanical power into electrical power by creating relative motion between a magnetic field and
a conductor. The energy source harnessed to turn the generator varies widely. It depends chiefly
on which fuels are easily available, cheap enough and on the types of technology to which the
power company has access. Most power stations in the world burn fossil fuels such as coal, oil,
and natural gases to generate electricity, and some use nuclear power; but there is an increasing
use of cleaner renewable sources such as solar, wind, wave and hydroelectric.

7
Transformer – Another major component of a power system is the transformer. It transfers
power with very high efficiency from one level of voltage to another level. The power
transferred from the secondary side of the transformer is almost the same as the primary, except
for losses in the transformer. Figure 1.4 presents two types of transformer typically used in
power system. The step-up transformers are used at the sending end of the transmission lines to
allow the delivery of power over long distances with lowest losses possible. At the receiving end
of the transmission lines, step-down transformers are used to reduce the voltage to suitable
values for distribution or utilization. In a modern utility system, the power may undergo four or
five transformations between generator and ultimate user.

Fig. 1.4: Step-up and Step-down transformers.

1.3.2. Transmission and Sub-transmission

The purpose of an overhead transmission network is to transfer electric energy from
generating unit at various locations to the distribution system which ultimately supplies the
loads. Standard transmission voltage levels at more than 60 kV are 69 kV, 115 kV, 138 kV, 161
kV, 230 kV, 345 kV, 500 kV, and 765 kV line-to-line. Transmission voltages above 230 kV are
usually referred to as extra-high voltage (EHV).
High voltage transmission lines are terminated in several substations, which are called
high-voltage substations, receiving substations, or primary substations. The function of some
substations is switching circuits in and out of the service; they are referred to as switching
stations. At the primary substations, the voltage is stepped down to a value more suitable for the
next part of the journey toward the load. Very large industrial customers may be served from the
transmission system. The portion of the transmission system that connects the high-voltage
substation through step-down transformers to the distribution substations are called the sub-
transmission network. There is no clear delineation between transmission and sub-transmission
voltage levels. Typically, the sub-transmission voltage level ranges from 69 to 138 kV.
8
1.3.3. Distribution
Distribution system is that part which connects the distribution substations to the
consumers’ service-entrance equipment. The primary distribution lines are usually in the range
of 4 to 34.5 kV and supply the load in a well-defined geographical area. The secondary
distribution network reduces the voltage for utilization by commercial and residential consumers.
The secondary distribution serves most of the customers at levels of 240/120 V, single-phase,
three wire; 208Y/120 V, three-phase, four-wire; or 480Y/277 V, three-phase, four-wire.
Distribution systems are both overhead and underground. The growth of underground
distribution has been extremely rapid and as much as 70% of new residential construction is
served underground.

1.3.4. Loads
Loads of power systems are divided into industrial, commercial, and residential loads.
Very large industrial loads may be served from the transmission system. Large industrial loads
are served directly from the sub-transmission network, and small industrial loads are served from
the primary distribution network. Commercial and residential loads consist largely of lighting,
heating, and cooling appliances.
The real powers of loads are expressed in terms of kW or MW. The magnitude of load
varies throughout the day, and power must be available to consumers on demand. The daily-load
curve of a utility is a composite of demands made by various classes of users (see Figure 1.5).
The greatest value of load during a 24-hr period is called the peak or maximum demand.

Fig. 1.5: Typical daily load demand profile.

9
1.4. MODERN POWER SYSTEM

1.4.1. Microgrid
Microgrid are small-scale, low voltage supply networks designed to supply electrical and
heat loads for a small community, such as a housing estate or a suburban locality, or an academic
or public community such as university or school, a commercial area, an industrial site, a trading
estate or a municipal region. The generators or microsources employed in a Microgrid are
usually renewable/non-conventional energy sources such as wind, solar, tidal energy sources.
A typical Microgrid configuration is shown in Figure 1.6. It consists of electrical/heat
loads and microsources connected through a low-voltage distribution network. The loads
(especially the heat loads) and the sources are placed close together to minimize heat loss during
heat transmission. The microgrids have plug-and-play features. They are provided with power
electronic interfaces (PEIs) to implement the control, metering and protection functions during
stand-alone and grid-connected modes of operation.

Fig. 1.6: Typical microgrid configuration.

The key differences between a Microgrid and a conventional power plant are as follows:
1) Microsources are of much smaller capacity with respect to the large generators in
conventional power plants.
2) Power generated at distribution voltage can be directly fed to the utility distribution
network.
3) Microsources are normally installed close to the customer side so that the
electrical/heat loads can be efficiently supplied with satisfactory voltage and frequency
profile and negligible line losses.
10
1.4.2. Smart Power Grid
A smart power grid (or smart grid) is an electricity network based on digital technology
that is used to supply electricity to consumers via two-way digital communication. This system
allows for monitoring, analysis, control and communication within the supply chain to help
improve efficiency, reduce energy consumption and cost, and maximize the transparency and
reliability of the energy supply chain. Like the Internet, the smart grid will consists of controls,
computers, automation, and new technologies and equipment working together, but in this case,
these technologies will work with the electrical grid respond digitally to our quickly changing
electric demand.

Fig. 1.7: Typical smart grid configuration.

The smart grid represents an unprecedented opportunity to move the energy industry into a
new era of reliability, availability, and efficiency that will contribute to our economic and
environmental health. The benefits associated with the smart grid include:
- More efficient transmission of electricity.
- Quicker restoration of electricity after power disturbances.
- Reduced operations and management costs for utilities, and ultimately lower power
costs for consumers.
- Reduced peak demand, which will also help lower electricity rates.
- Increased integration of large-scale renewable energy systems.
- Better integration of customer-owner power generation systems, including renewable
energy systems.
- Improved security.

11
1.5. GOAL OF SYSTEM ANALYSIS AND DESIGN

1.5.1. System design

When considering the design of a power system for a given customer and facility, the
electrical engineering must consider alternate design approaches that best fit the following
overall goals.

1.5.1.1. Safety
The first goal is to design a power system that will not present any electrical hazard to the
people who use the facility, and/or the utilization equipment fed from the electrical system. It is
also important to design a system that is inherently safe for the people who are responsible for
electrical equipment maintenance and upkeep.

1.5.1.2. Minimum Initial Investment

The owners overall budget for first cost purchase and installation of the electrical
distribution system and electrical utilization equipment will be a key factor in determining which
of various alternate system designs are to be selected. When trying to minimize initial investment
for electrical equipment, consideration should be given to the cost of installation, floor space
requirements and possible extra cooling requirements as well as the initial purchase price.

1.5.1.3. Maximum Service Continuity

The degree of service continuity and reliability needed will vary depending on the type and
use of the facility as well as the loads or processes being supplied by the electrical distribution
system. For example, for a smaller commercial office building, a power outage of considerable
time, say several hours, may be acceptable, whereas in a larger commercial building or industrial
plant only a few minutes may be acceptable. In other facilities such as hospitals, many critical
loads permit a maximum of 10 seconds outage and certain loads, such as real-time computers,
cannot tolerate a loss of power for even a few cycles.

1.5.1.4. Maximum Flexibility and Expendability

In many industrial manufacturing plants, electrical utilization loads are periodically
relocated or changed requiring changes in the electrical distribution system. Consideration of the
layout and design of the electrical distribution system to accommodate these changes must be
considered. For example, providing many smaller transformers or load centers associated with a
given area or specific groups of machinery may lend more flexibility for future changes than one
large transformer; the use of plug-in bus ways to feed selected equipment in lieu of conduit and
wire may facilitate future revised equipment layouts.

12
 In addition, consideration must be given to future building expansion, and/or increased
load requirements due to added utilization equipment when designing the electrical distribution
system. In many cases considering transformers with increased capacity or fan cooling to serve
unexpected loads as well as including spare additional protective devices and/or provision for
future addition of these devices may be desirable. Also to be considered is increasing appropriate
circuit capacities or quantities for future growth. Power monitoring communication systems
connected to electronic metering can provide the trending and historical data necessary for future
capacity growth.

1.5.1.5. Maximum Electrical Efficiency (Minimum Operating Costs)

Electrical efficiency can generally be maximized by designing systems that minimize the
losses in conductors, transformers and utilization equipment. Proper voltage level selection plays
a key factor in this area and will be discussed later. Selecting equipment, such as transformers,
with lower operating losses, generally means higher first cost and increased floor space
requirements; thus, there is a balance to be considered between the owners utility energy change
for the losses in the transformer or other equipment versus the owners first cost budget and cost
of money.

1.5.1.6. Minimum Maintenance Cost

Usually the simpler the electrical system design and the simpler the electrical equipment,
the less the associated maintenance costs and operator errors. As electrical systems and
equipment become more complicated to provide greater service continuity or flexibility, the
maintenance costs and chance for operator error increase. The systems should be designed with
an alternate power circuit to take electrical equipment (requiring periodic maintenance) out of
service without dropping essential loads. Use of draw out type protective devices such as
breakers and combination starters can also minimize maintenance cost and out-of service time.

1.5.1.7. Maximum Power Quality

The power input requirements of all utilization equipment has to be considered including
the acceptable operating range of the equipment and the electrical distribution system has to be
designed to meet these needs. For example, what is the required input voltage, current, power
factor requirement? Consideration to whether the loads are affected by harmonics (multiples of
the basic 60 Hz sine wave) or generate harmonics must be taken into account as well as transient
voltage phenomena. The above goals are interrelated and in some ways contradictory. As more
redundancy is added to the electrical system design along with the best quality equipment to
maximize service continuity, flexibility and expandability, and power quality, the more initial

13
investment and maintenance are increased. Thus, the designer must weigh each factor based on
the type of facility, the loads to be served, the owners past experience and criteria.

1.5.2. System analysis

1.5.2.1. Short-circuit currents – General

The amount of current available in a short-circuit fault is determined by the capacity of
the system voltage sources and the impedances of the system, including the fault. Voltage
sources include the power supply (utility or on-site generation) plus all rotating machines
connected to the system at the time of the fault. A fault may be either an arcing or bolted
fault. In an arcing fault, part of the circuit voltage is consumed across the fault and the
total fault current is somewhat smaller than for a bolted fault, so the latter is the worst
condition, and therefore is the value sought in the fault calculations.
Basically, the short-circuit current is determined by applying Ohms Law to an equivalent
circuit consisting of a constant voltage source and a time varying impedance. Time-varying
impedance is used in order to account for the changes in the effective voltages of the rotating
machines during the fault. In an AC system, the resulting short-circuit current starts out higher in
magnitude than the final steady state value and asymmetrical (due to the DC offset) about the X-
axis. The current then decays toward a lower symmetrical steady-state value. The time-varying
characteristic of the impedance accounts for the symmetrical decay in current. The ratio of the
reactive and resistive components (X/R ratio) accounts for the DC decay. The fault current
consists of an exponentially decreasing direct current component superimposed upon a decaying
alternating-current. The rate of decay of both the DC and AC components depends upon the ratio
of reactance to resistance (X/R) of the circuit. The greater this ratio, the longer the current
remains higher than the steady-state value that it would eventually reach.
The total fault current is not symmetrical with respect to the time-axis because of the
direct-current component; hence it is called asymmetrical current. The DC component depends
on the point on the voltage wave at which the fault is initiated.

1.5.2.2. Fault current calculations

The calculation of asymmetrical currents is a laborious procedure since the degree of
asymmetry is not the same on all three phases. It is common practice for medium voltage
systems, to calculate the RMS symmetrical fault current, with the assumption being made that
the DC component has decayed to zero, and then apply a multiplying factor to obtain the first
half-cycle RMS asymmetrical current, which is called the momentary current. For medium
voltage systems (defined by IEEE as greater than 1000V up to 69,000V) the multiplying factor is
established by NEMA and ANSI standards depending upon the operating speed of the breaker.

14
For low voltage systems, short-circuit study software usually calculates the symmetrical fault
current and the faulted system X/R ratio using ANSI guidelines. If the X/R ratio is within the
standard, and the breaker interrupting current is under the symmetrical fault value, the breaker is
properly rated. If the X/R ratio is higher than ANSI standards, the study applies a multiplying
factor to the symmetrical calculated value (based on the X/R value of the system fault) and
compares that value to the breaker symmetrical value to assess if it is properly rated. In the past,
especially using manual calculations, a multiplying factor of 1.17 (based on the use of an X/R
ratio of 6.6 representing a source short-circuit power factor of 15%) was used to calculate the
asymmetrical current. These values take into account that medium voltage breakers are rated on
maximum asymmetry and low voltage breakers are rated average asymmetry.
To determine the motor contribution during the first half-cycle fault current, when
individual motor horsepower load is known, the sub-transient reactances found in the IEEE Red
Book should be used in the calculations. When the system motor load is unknown, the following
assumptions generally are made: Induction motors use 4.0 times motor full load current
(impedance value of 25%).

1.5.2.3. Voltage drop considerations

The first consideration for voltage drop is that under the steady-state conditions of normal
load, the voltage at the utilization equipment must be adequate. Fine-print notes in the NEC
recommend sizing feeders and branch circuits so that the maximum voltage drop in either does
not exceed 3%, with the total voltage drop for feeders and branch circuits not to exceed 5%, for
efficiency of operation. (Fine print notes in the NEC are not mandatory.)
In addition to steady-state conditions, voltage drop under transient conditions, with sudden
high-current, short time loads, must be considered. The most common loads of this type are
motor inrush currents during starting. These loads cause a voltage dip on the system as a result of
the voltage drop in conductors, transformers and generators under the high current. This voltage
dip can have numerous adverse effects on equipment in the system, and equipment and
conductors must be designed and sized to minimize these problems. In many cases, reduced-
voltage starting of motors to reduce inrush current will be necessary.

1.5.2.4. Motor starting

Motor inrush on starting must be limited to minimize voltage dips. Where the power is
supplied by a utility network, the motor inrush can be assumed to be small compared to the
system capacity, and voltage at the source can be assumed to be constant during motor starting.
Voltage dip resulting from motor starting can be calculated on the basis of the voltage drop in the
conductors between the power source and the motor resulting from the inrush current. Where the

15
utility system is limited, the utility will often specify the maximum permissible inrush current or
the maximum hp motor they will permit to be started across-the-line.

16
 CHAPTER II
FUNDAMENTALS

The concept of power is of central importance in electrical power systems and is the main
topic of this chapter. Firstly, the flow of energy in an AC circuit is investigated. The average
power P, reactive power Q, and also the volt-ampere S which is a mathematical formulation
based on the phasor forms of voltage and current, are introduced. Next, the transmission of
complex power between two voltage sources is considered, and the dependency of real power on
the voltage phase angle and the dependency of reactive power on voltage magnitude is
established. Finally, the balanced three-phase circuit is examined.

2.1. PHASORS
A sinusoidal voltage or current at constant frequency is characterized by two parameters: a
maximum value and a phase angle. Consider a voltage
v(t )  Vmax cos(t   ) (2.1)

has a maximum value Vmax and a phase angle  when referenced to cos(t). The root-mean-
square (rms) value, also called effective value, of the sinusoidal voltage is
Vmax
V  (2.2)
2
Euler’s identity, e j  cos  j sin  , can be used to express a sinusoid in terms of a
phasor. For the above voltage,
v(t )  Re[Vmax e j (t  ) ] (2.3)
The rms phasor representation of the voltage is given in three forms – exponential, polar,
and rectangular:
Vmax j Vmax V V
V  e    max cos   j max sin  (2.4)
2 2 2 2
A phasor can be easily converted from one form to another. Conversion from polar to
rectangular is shown in the phasor diagram of Figure 2.1.

17
 Fig. 2.1: Phasor diagram for converting from polar to rectangular form.
Example 2.1:
The voltage
v(t )  169.7cos(t  60o ) (volts)
has a maximum value Vmax = 169.7 volts, a phase angle  = 60o when referenced to cos(t), and
an rms phasor representation in polar form is
V  12060o (volts)

2.2. SINGLE-PHASE AC CIRCUIT POWER

Consider a circuit in Figure 2.2, let the instantaneous voltage be
v(t )  Vmax cos(t   ) (2.5)
and the instantaneous current be given by
i(t )  I max cos(t   ) (2.6)

Fig. 2.2: Sinusoidal source supplying a load.

By reviewing some additional results from elementary circuit theory, the instantaneous
power p(t) delivered to the load is the product of voltage v(t) and current i(t), given by
p(t )  v(t )i (t )  Vmax I max cos(t   )cos(t   ) (2.7)
It is informative to write (2.7) in another form using the trigonometric identity
1 1
cos A cos B  cos( A  B)  cos( A  B) (2.8)
2 2
which results in
1
p(t )  Vmax I max [cos(   )  cos(2t     )
2
1
 Vmax I max {cos(   )  cos[2(t   )  (   )]}
2 (2.9)
1
 Vmax I max [cos(   )  cos 2(t   ) cos(   )
2
 sin 2(t   )sin(   )]
The above equation, in terms of the rms values, is reduced to
p(t )  V I cos  [1  cos 2(t   )]  V I sin  sin 2(t   ) (2.10)
pR ( t ) pX ( t )

18
 where  = ( - ) is the angle between voltage and current, or the impedance angle.  is
positive if the load is inductive (i.e., current is lagging the voltage) and is negative if the load is
capacitive (i.e., current is leading the voltage). According to (2.10), the instantaneous power has
two components. The first term pR(t) refers to the power absorbed by the resistive component of
the load and the second term pX(t) accounts for power oscillating into and out of the load because
of its reactive element (inductive or capacitive).

2.2.1. Active power

The resistive component of the load absorbs the power
pR (t )  V I cos  V I cos cos2(t   )] (2.11)

The second term in (2.11), which has a frequency twice that of the source, accounts for the
sinusoidal in the absorption of power by the resistive portion of the load. Since the average value
of this sinusoidal function is zero, the average power delivered to the load is given by
P  V I cos (2.12)

This is the power absorbed by the resistive component of the load and is also referred to as
the active power or real power.

2.2.2. Reactive power

The reactive component of the load absorbs the power
pX (t )  V I sin  sin 2(t   ) (2.13)

This component pulsates with twice the frequency and has an average value of zero. The
amplitude of this pulsating power is called reactive power and is designed by Q.
Q  V I sin  (2.14)

2.2.3. Power factor

Since cos plays a key role in the determination of the average power, it is called power
factor. Unlike the case of DC, the product of voltage and current does not give the power in AC
circuits. To find the power we need to multiply the product of rms voltage and rms current by the
power factor.
For inductive loads, the current lags the voltage, which mean  > 0, the power factor is said
to be lagging. For capacitive loads, the current leads the voltage, which mean  < 0, the power
factor is said to be leading.
Note
- The product of the rms voltage value and the rms current value VI is called the
apparent power and is measured in unit of ampere.

19
 - For the pure resistor, the impedance angle is zero and the power factor is unity (UPF),
so that the apparent and real power are equal. The electric energy is transformed into
thermal energy.
- If the circuit is purely inductive, the current lags the voltage by 900 and the average
power is zero. Therefore, in a purely inductive circuit, there is no transformation of
energy from electrical to non-electrical form. The instantaneous power at the terminal
of a purely inductive circuit oscillates between the circuit and the source. When p(t) is
positive, energy is being stored in the magnetic field associated with the inductive
elements, and when p(t) is negative, energy is being extracted from the magnetic fields
of the inductive elements.
- If the load is purely capacitive, the current leads the voltage by 900, and the average
power is zero, so there is no transformation of energy from electrical to non-electrical
form. In a purely capacitive circuit, the power oscillates between the source and the
electric field associated with the capacitive elements.
Example 2.1
The voltage in Figure 2.3 is given by v(t) = 141.4 cos(t) is applied to a load consisting of
a 10  resistor in parallel with an inductive reactance XL = L = 3.77 . Calculate the
instantaneous power absorbed by the resistor and by the inductor. Also calculate the real and
reactive power absorbed by the load, and the power factor.

Fig. 2.3: Circuit diagram for example 2.1.

Solution
The load voltage is
141.4 o
V 0  1000o (volts)
2
The resistor current is
V 100 o
IR   0  100o (A)
R 10
The inductor current is
V 100
IL   0o  26.53  90o (A)
jX L ( j 3.77)
20
 The total load current is
I  I R  I L  10  j 26.53  28.35  69.34o (A)
The instantaneous power absorbed by the resistor is, from 2.11,
pR (t )  100  10  [1  cos(2t )]  1000[1  cos(2t )] (W)
The instantaneous power absorbed by the inductor is, from 2.13,
pL (t )  100  [26.53sin(2t )]  2653sin(2t )] (W)
The real power absorbed by the load is, from 2.12
P  100  28.53  cos(0o  69.34o )  1000 (W)
(Note: P is also equal to VIR = V2/R)
The reactive power absorbed by the load is, from 2.14,
P  100  28.53  cos(0o  69.34o )  2653 (Var)
(Note: Q is also equal to VIL = V2/XL)
The power factor is
p. f  cos(0o  69.34o )  0.3528 (lagging)

2.3. COMPLEX POWER

Let the voltage across a circuit element be V  V  v , and the current into the element be

I  I  i . The product of the voltage and the conjugate of current results in

VI *  [ V  v ][ I  i ]*  V I ( v   i )
(2.15)
= V I cos( v   i )  j V I sin( v   i )
The above equation defines a complex quantity where its real part is the active power P
and its imaginary part is the reactive power Q. Thus, the complex power designated by S is given
by
S  VI *  P  jQ (2.16)

The magnitude of S, S  P 2  Q 2 , is called the apparent power; its unit is volt-amperes

(VA) and the larger unit are kVA or MVA.

Note
- The reactive power Q is positive when the phase angle  between voltage and current
(impedance angle) is positive (i.e., when the load impedance is inductive, and I lags V).
Q is negative when  is negative (i.e., when the load impedance is capacitive and I
leads V).
- In working with Equation (2.16) it is convenient to think of P, Q and S as forming the
sides of a triangle as shown in Figure 2.4.

21
 V

I S

Q
v
i 
P

(a) For an inductive load (lagging PF).

I V

 P

i Q
S
v

(b) For a capacitive load (leading PF).

Fig. 2.4: Phasor diagram and power triangle.
- Let Z is the complex impedance of the circuit then V = ZI, substituting for V into 2.16
yields
2
VV * V
S  VI  *  *
*
(2.17)
Z Z
From (2.17), the complex impedance of the complex power S is given by
2
V
Z * (2.18)
S

Fig. 2.5: The complex impedance.

Example 2.2
The circuit in Figure 2.6 has three loads. The voltage applied to the circuit is V  12000o
; three loads are Z1  60  j0 Ω, Z 2  6  j12 Ω, Z3  30  j30 Ω. Find the power absorbed by
each load and the total complex power.

22
 Fig. 2.6: The circuit with three loads in parallel.
Solution:
The current into each load is calculated as follows:
12000o
I1   20  j 0 (A)
600
12000o
I2   40  j80 (A)
6  j12
12000o
I3   20  j 20 (A)
30  j 30
The power absorbed by each load is given by:
S1  VI1*  12000o (20  j 0)  24,000 W  j 0 Var
S2  VI 2*  12000o (40  j80)  48,000 W  j 96,000 Var
S3  VI 3*  12000o (20  j 20)  24,000 W  j 24,000 Var
The total load complex power adds up to
S  S1  S2  S3  96,000 W  j72,000 Var
Alternatively, the sum of complex power delivered to the load can be obtained by first
finding the total current.
I  I1  I 2  I 3  80  j60  100  36.87o ( A)
and
S  VI *  (12000o )(100  36.87o )  120,00036.87o (VA)
 96,000 W  j72,000 Var
Example 2.3
Two loads Z1  100  j0 Ω and Z2  10  j 20 Ω are connected across a 200-V rms, 60 Hz
source as shown in Figure 2.7.
(a) Find the total real and reactive power, the power factor at the source, and the total
current.
(b) Find the capacitance of the capacitor connected across the loads to improve the overall
power factor to 0.8 lagging.

23
 Fig. 2.6: The circuit with three loads in parallel.

2.4. BALANCED THREE-PHASE CIRCUITS

The generator, transmission and distribution of electric power is accomplished by means of
three-phase circuits. At the generating station, three sinusoidal voltages are generated having the
same amplitude but displaced in phase by 120o. This is called a balanced source. If the generated
voltages reach their peak values in the sequential order ABC, the generator is said to have a
positive phase sequence, as shown in Figure 2.7(a). If the phase order is ACB, the generator is
said to have a negative phase sequence, as shown in Figure 2.7(b).

(a) (b)
Fig. 2.7: (a) Positive phase sequence, (b) Negative phase sequence.

2.4.1. Y-connected loads

Figure 2.8 shows a three-phase Y-connected (or “wye-connected”) voltage source feeding
a balanced Y-connected load. For a Y connection, the neutrals of each phase are connected. In
Figure 2.8 the neutral connection is labeled bus n.

24
 Fig. 2.8: A Y-connected generator supplying a Y-connected load.
Line-to-line voltage
To find the relationship between the line voltages (line-to-line voltages) and the phase
voltages (line-to-neutral voltages), we assume a positive sequence. We arbitrarily choose the
line-to-neutral voltage of a-phase as the reference, thus
Van  V p 0o
Vbn  V p   120o (2.19)
Vcn  V p   240o

Where V p represents the magnitude of the phase voltage (line-to-neutral voltage).

The line voltages at the load terminals in terms of the phase voltages are found by the
application of Kirchhoff’s voltage law

Vab  Van  Vbn  V p (10o  1  120o )  3 V p 30o

Vbc  Vbn  Vcn  V p (1  120o  1  240o )  3 V p   90o (2.20)
Vca  Vcn  Van  V p (1  240o  10o )  3 V p 150o

The voltage phasor diagram of the Y-connected loads of Figure 2.8 is shown in Figure 2.9.
The relationship between the line voltages and phase voltages is demonstrated graphically.

Fig. 2.9: Phasor diagram showing phase and line voltages (Y-connection).

25
 In case of the Y-connected loads, the magnitude of the line voltage is 3 times the
magnitude of the phase voltage, and for a positive phase sequence, the set of line voltages leads
the set of phase voltages by 30o. If the rms value of any of the line voltages is denoted by V L,
then
VL  3 Vp 30o (2.21)

Line current
The three-phase currents in Figure 2.8 also possess three-phase symmetry and are given by
Van
Ia   Ip  
Zp
Vbn
Ib   I p   120o   (2.22)
Zp
Vcn
Ic   I p   240o  
Zp

where is the impedance phase angle.

The line currents are also the phase currents. Thus
IL  I p (2.23)

Note that the currents form a balanced set of phasors and the sum of the currents is zero
Ia  Ib  Ic  0 (2.24)

2.4.2. -connected loads

A balanced -connected load (with equal phase impedances) is shown in Figure 2.10.

Fig. 2.10: A balanced -connected load.

Line-to-line voltage

It is clear from the inspection of the circuit that the line voltages are the same as the phase
voltages. Thus,

VL  Vp (2.25)

Phase current
26
 Consider the phasor diagram shown in Figure 2.11, where the phase current Iab is
arbitrarily chosen as reference, we have

I ab  I p 0o
I bc  I p   120o (2.26)
I ca  I p   240o

Where I p represents the magnitude of the phase current.

Fig. 2.11: Phasor diagram showing phase and line voltages (-connection).

The relationship between phase and line currents can be obtained by applying Kirchhoff’s
current law at the corners of .

I a  I ab  I ca  I p (10o  1  240o )  3 I p   30o

I b  I bc  I ab  I p (1  120o  10o )  3 I p   150o (2.27)
I c  I ca  I bc  I p (1  240o  1  120o )  3 I p 90o

If the rms of any the line currents is denoted by IL, then one of the important characteristics
of the -connected three-phase load may be expressed as

I L  3 I p   30o (2.28)

Thus in the case of -connected loads, the magnitude of the line current is 3 times the
magnitude of the phase current, and with positive phase sequence, the set of line currents lags the
set of phase currents by 30o.

2.4.3. -Y transformation

For analyzing network problems, it is convenient to replace the -connected circuit with an
equivalent Y-connected circuit, or vice versa. Consider the Y-connected circuit of ZY Ω/phase
which is equivalent to a balanced -connected circuit of Z Ω/phase, as shown in Figure 2.12.

27
 Fig. 2.12: (a)  to (b) Y-connection.
For the  load,

3Vab  30o
I a  3 I ab   30o  (2.29)
Z
and for the Y load,
Van Vab  30o
Ia   (2.30)
ZY 3ZY

Comparison of (2.29) and (2.30) indicates that IA will be the same for both the  and Y
loads when
Z
ZY  (2.31)
3
Thus a balanced -connected load can be converted to an equivalent balanced Y-connected
load by dividing the  load impedance by 3. The angles of these - and equivalent Y- load
impedances are the same. Similarly, a balanced Y load can be converted to an equivalent
balanced  load using Z = 3ZY.

2.4.4. Equivalent line-to-neutral diagrams

When working with balanced three-phase, only one phase need to be analyzed.  loads can
be converted to Y loads, and all source and load neutrals can be connected with a zero-Ω neutral
wire without changing the solution. Then one phase of the circuit can be solved. The voltages
and currents in the other two phases are equal in magnitude to and ±120o out of phase with those
of the solved phase. Figure 2.13 shows an equivalent line-to-neutral diagram for one phase of the
circuit in Figure 2.8.

28
 Fig. 2.13: Single-phase circuit for per-phase analysis.

2.4.5. Balanced three-phase power

Instantaneous power
Consider a balanced three-phase source applying a balanced Y- or -connected load with
the following instantaneous voltages

van  2 V p cos( t   )
vbn  2 V p cos( t    120o ) (2.32)
vcn  2 V p cos( t    240o )

For a balanced load, the phase currents are

ia  2 I p cos( t   )
ib  2 I p cos( t    120o ) (2.33)
ic  2 I p cos( t    240o )

where V p and I p are the magnitudes of the rms phase voltage and current, respectively. The

total instantaneous power is the sum of the instantaneous power of each phase, given by
p3  vania  vbnib  vcnic (2.34)

Substituting for the instantaneous voltages and currents from (2.32) and (2.33) into (2.34),
using the trigonometric identity we have
p3  V p I p [cos(   )  cos(2t     )]
 V p I p [cos(   )  cos(2t      240o )] (2.35)
 V p I p [cos(   )  cos(2t      480o )]

The three double frequency cosine terms in (2.35) are out of phase with each other by 120o
and add up to zero (the easiest way to check that is to add the phasor representations). Therefore
the three-phase instantaneous power is

p3  P3  3 Vp I p cos(   )  3 Vp I p cos (2.36)

29
 Note that although the power in each phase is pulsating, the total instantaneous power is
constant and equal to three times the real power in each phase.
Complex power
In order to obtain the three-phase complex or apparent power (S), the three-phase reactive
power is also defined as

Q3  3 Vp I p sin  (2.37)

Thus, the complex three-phase power is

S3  P3  jQ3  3 Vp I p [cos  jsin  ] (2.38)

or
S3  3Vp I *p (2.39)

Alternatively, the real and reactive powers can be presented in term of line-to-line voltage
and line current as follow:
P3  3 VL I L cos  (2.40)

Q3  3 VL I L sin  (2.41)

Example 2.4
A three-phase line has an impedance of (2 + j4) Ω as shown in Figure 2.14.

Fig. 2.14: Three-phase circuit diagram for example 2.4.

The line feeds two balanced three-phase loads that are connected in parallel. The first load is Y-
connected and has an impedance of (30 + j40) Ω per phase. The second load is -connected and
has an impedance of (60 – j45) Ω. The line is energized at the sending end from a three-phase
balanced supply of line voltage 207.85 V. Taking the phase voltage Va as reference, determine:
(a) The current, real power, and reactive power drawn from the supply.
(b) The line voltage at the combined loads.
(c) The current per phase in each load.
(d) The total real and reactive powers in each load and the line.

30
2.5. PER-UNIT ANALYSIS

2.5.1. The per-unit system

In the analysis of power networks, instead of using actual values of quantities it is usual to
express them as fractions of reference quantities, such as rated or full-load values. By using this
expression, the different voltage levels disappear, and a power network involving generators,
transformers, and lines reduces to a system of simple impedances. These fractions are called per
unit (denoted by p.u) and the p.u value of any quantity is defined as
actual quantity
Quantity in p.u = (2.42)
base value of quantity
For example,
S V I Z
S pu  ; Vpu  ; I pu  ; Z pu  (2.43)
SB VB IB ZB
Where the numerators (actual values) are phasor quantities or complex values and the
denominators (base values) are always real numbers.
Usually, the three-phase base volt-ampere SB and the line-to-line base voltage VB are
selected. Base current and base impedance are then calculated. These are given by
SB
IB  (2.44)
3VB
and

VB / 3
ZB  (2.45)
IB
Substituting for IB from (2.44), the base impedance becomes
(VB )2 (kVB )2
ZB   (2.46)
SB MVAB

2.5.2. Change of base

Let Z old old old
pu be the per-unit impedance on the power base S pu and the voltage base V pu ,

which is expressed by
Z SBold
Z old
pu  old  Z  old 2 (2.47)
ZB (VB )
Expressing ZΩ to a new power base and a new voltage base, results in the new per-unit
impedance
Z SBnew
new
Z pu   Z  (2.48)
Z Bnew (VBnew )2
From (2.47) and (2.48), the relationship between the old and the new per-unit values is
31
 2
SBnew  VBold 
Z new
pu Z old
pu   (2.49)
SBold  VBnew 

2.6. ASSIGNMENT

32
 CHAPTER III
POWER TRANSFORMER MODELING

The power transformer is a major power system component that permits economical power
transmission with high efficiency and low series-voltage drops. In this chapter, we review basic
transformer theory and develop simple equivalent models of power transformers for steady-state
balanced operation. We look at models of single-phase two-winding, three-phase two-winding,
and three-phase three-winding transformers, as well as autotransformers and regulating
transformers.

3.1. SINGLE-PHASE TRANSFORMER

3.1.1. The ideal transformer

Figure 3.1 shows a basic single-phase two-winding transformer. The diagram is intended to
illustrate the working principle of a transformer.

Fig. 3.1: Basic single-phase two-winding transformer.

Shown in the figure are the phasor voltage E1 and E2 across the windings, the phasor
current I1 entering winding 1, which has N1 turns, and the phasor current I2 leaving winding 2,
which has N2 turns. The phasor flux c set up in the core and a magnetic field intensity phasor
Hc are also shown. The core has a cross-sectional area denoted Ac, a mean length of the magnetic
circuit lc, and a magnetic permeability µc, assumed constant.
An ideal transformer which is schematically represented in Figure 3.2 has the following
physical properties:
1. The windings have zero resistance; therefore, the losses in the winding are zero.
2. The core permeability µc is infinite, which corresponds to zero core reluctance.
3. There is no leakage flux; that is, the entire flux c is confined to the core and links both
windings.
4. There are no core losses (hysteresis and eddy current losses are negligible).
33
 Fig. 3.2: Schematic representation of a single-phase two winding transformer.
While the assumption of no losses implies that there is no resistance in the circuit model,
the assumption of no leakage fluxes mean we have 1  N1 c and 2  N 2 c as the flux linkages
of the primary and secondary circuits, respectively. The terminal voltages are then given by
d 1 dc
E1   N1
dt dt (3.1)
d d c
E2  2  N 2
dt dt
and the voltage gain is
E2 N 2
 n (3.2)
E1 N1
The assumption that the magnetic circuit of the ideal transformer is lossless implies that the
magnetomotive forces (mmfs) produced by the windings balance, that is, primary mmf equals
secondary mmf. In term of the winding currents, this may be expressed as
N1I1  N 2 I 2 (3.3)
Equation (3.3) shows that the winding currents are in phase and that their magnitudes are
related by
I1 N1 1
  (3.4)
I2 N2 a
Two additional relations concerning complex power and impedance can be derived from
(3.2) and (3.4) as follows. The complex power entering winding 1 in Figure 3.2 is
S1  E1I1* (3.5)
Using (3.2) and (3.4),
*
I 
S1  E I  (n E2 )  2   E2 I 2*  S2
*
1 1 (3.6)
n
As shown by (3.6), the complex power S1 entering winding 1 equals the complex power S2
leaving winding 2. That is, an ideal transformer has no real or reactive power loss.
If an impedance Z2 is connected across winding 2 of the ideal transformer in Figure 3.2,
then

34
 E2
Z2  (3.7)
I2
This impedance, when measured from winding 1, is
2
E nE2 N 
Z2  1 
'
 n2Z2   1  Z2 (3.8)
I1 I 2 / n  N2 
Thus, the impedance Z2 connected to winding 2 is referred to winding 1 by multiplying Z2
by n2, the square of the turn ratio.

3.1.2. The nonideal or actual transformer

Figure 3.3 shown an equivalent circuit for a practical single-phase two winding
transformer. For an actual transformer, the assumptions of an ideal transformer are no longer
capable. An actual transformer is characterized by
- Having resistances in its windings.
- Not all of the flux produced by one winding will link the other winding because of flux
leakage.
- The core has a finite permeability.
- Core losses are not neglected.

Fig. 3.3: Equivalent circuit of a practical single-phase two winding transformer.

The resistance R1 is included in series with winding 1 to account for I2R losses in this
winding. A reactance X1, called the leakage reactance of winding 1, is also included in series
with winding 1 to account for the leakage flux of winding 1. There is also a reactive power loss
I12 X 1 associated with this leakage reactance. Similarly, there is a resistance R2 and a leakage
reactance X2 in series with winding 2.
For the purpose of modeling, we assume a sinusoidal no-load current with the rms value of
I0, known as the no-load current. This current has a component Im, in phase with flux, known as
the magnetizing current, to set up the core flux. Since flux is lagging the induced voltage E1 by
90o, Im is also lagging the induced voltage E1 by 90o. Thus, this component can be represented in
the circuit by the magnetizing reactance jXm1. The other component of I0 is Ic, which supplies the
eddy-current and hysteresis losses in the core. Since this is a power component, it is in phase
with E1 and is presented by the resistance Rc1 as shown in Figure 3.3.
35
 A phasor diagram for a lagging power factor (inductive) load is shown in Figure 3.4.

Fig. 3.4: Phasor diagram for a lagging power factor.

To obtain the performance characteristics of a transformer, it is convenient to use an
equivalent circuit model referred to one side of the transformer. From Kirchhoff’s voltage law
(KVL), the voltage equation of the secondary side is

E2  V2  Z 2 I 2 (3.9)
Or
2
N N 
E1  1 V2   1  Z 2 I 2'  V2'  Z 2' I 2' (3.10)
N2  N2 
where
2 2
N  N 
Z  R  jX   1  R2 
'
2
'
2
'
2 j  1  X2 (3.11)
 N2   N2 
Relation (3.10) is the KVL equation of the secondary side referred to the primary, and the
equivalent circuit of Figure 3.3 can be redrawn as shown in Figure 3.5, so the same effects are
produced in the primary as would by produced in the secondary.

Fig. 3.5: Exact Equivalent circuit referred to the primary side.

On no-load, the primary voltage drop is very small, and V1 can be used in place of E1 for
computing the no-load current I0. Thus, the shunt branch can be moved to the left of the primary
series impedance with very little loss of accuracy. In this manner, the primary quantities R 1 and

36
X1 can be combined with the referred secondary quantities R2' and X 2' to obtain the equivalent
primary quantities Re1 and Xe1. The equivalent circuit is shown in Figure 3.6 where we have
dispensed with the coils of the ideal transformer.

Fig. 3.6: Equivalent circuit referred to the primary.

From Figure 3.6,
V1  V2'  (R e1  jXe1 ) I 2' (3.12)

where
2 2
 N1   N1  S L*
R e1  R1    R2 ; X e1  X 1    X 2 ; and I 2  '*
'
(3.13)
 N2   N2  3V2
The equivalent circuit referred to the secondary is also shown in Figure 3.7. From Figure
3.7 the referred primary voltage V1' is given by

V1'  V2  (R e 2  jXe 2 ) I 2 (3.14)

Fig. 3.7: Equivalent circuit referred to the secondary.

The parameters of the equivalent circuit are readily obtained from open-circuit and short-
circuit tests. In the open-circuit test, rated voltage is applied at the terminals of one winding
while the other winding terminals are open-circuited, as shown in Figure 3.8.

Fig. 3.8: Equivalent circuit for the open-circuit test.

37
 Since the secondary winding copper loss (resistive power loss) is zero and the primary
copper loss R1 I 02 is negligible, the no-load input power P0 represents the transformer core loss
commonly referred to as iron loss. The shunt elements Rc and Xm may then be determined from
the relations
V12
Rc1  (3.15)
P0
The two components of the no-load current are
V1
Ic  (3.16)
Rc1
and

I m  I 02  I c2 (3.17)

Therefore, the magnetizing reactance is

V1
X m1  (3.18)
Im
In the short-circuit test, a reduced voltage Vsc is applied at the terminals of one winding
while the other winding terminals are short-circuited. The applied voltage is adjusted until rated
currents are flowing in the windings. The transformer appears as a short when viewed from the
primary with the equivalent leakage impedance Ze1 consisting of the primary leakage impedance
and the referred secondary leakage impedance as shown in Figure 3.9.

Fig. 3.9: Equivalent circuit for the short-circuit test.

The series elements Re1 and Xe1 may then be determined from the relations:
Vsc
Z e1  (3.19)
I sc
and
Psc
Re1  (3.20)
( I sc )2
Therefore, the equivalent leakage reactance is

X e1  Se21  Re21 (3.21)

38
3.1.3. Approximate equivalent circuit
Power transformer are generally designed with very high permeability core and very small
core loss. Consequently, a further approximation of the equivalent circuit can be made by
omitting the shunt branch, as shown in Figure 3.10.

Fig. 3.10: Simplified circuits referred to one side.

3.2. THREE-PHASE TRANSFORMER

The three-phase transformers are used quite extensively in power systems to transform a
balanced set of three-phase voltages at a particular level into a balanced set of voltages at another
level.

Three-phase transformer are formed in either of two ways:

- Connect three single-phase transformers to form a three-phase bank.

- Manufacture a three-phase transformer bank with all three phases located on a common
multi-legged core.

In the first way, the primary and secondary windings can be connected in either wye (Y) or
delta () configurations. This results in four possible combinations of connections: Y-Y, -, Y-
 and -Y shown by the simple schematic in Figure 3.11.

Fig. 3.11: Three-phase transformer connections.

39
3.3. AUTOTRANSFORMERS

A single –phase two-winding transformer with the two windings connected in series is
called autotransformer, as shown in Figure 3.12.

Fig. 3.12: An autotransformer.

The winding from X1 to X2 is called the series winding, and the winding from H1 to H2 is
called the common winding. From an inspection of this figure it follows that an autotransformer
can operate as a step-up as well as a step-down transformer. The performance of an
autotransformer is governed by the fundamental considerations already discussed for
transformers having two separate windings.

An autotransformer has the following advantages:

- Smaller per-unit leakage impedances than the usual transformer.

- Smaller series voltage drops.
- Lower per-unit losses (higher efficiency).
- Lower exciting current, and
- Lower cost if the turns ratio is not too large.

From the Figure 3.12, the two-winding voltages and currents are related by

V1 N1
 n (3.22)
V2 N 2

and

I 2 N1
 n (3.23)
I1 N 2

where n is the turns ratio of the two-winding transformer. We also have

VH  V2  V1 (3.24)

40
Substituting for V1 from (3.22) into (3.58) yields

N1
VH  V2  V2  (1  n) V2 (3.25)
N2

Since V2 = VL, the voltage relationship between the two sides of an autotransformer becomes

VH
VH  (1  n)VL or  1 n (3.26)
VL

The current relationship between the two sides of an autotransformer is also determined the
same as voltage relationship, given as

IL
 1 n (3.27)
IH

The ratio of the apparent power rating of an autotransformer to a two-winding transformer,

known as the power rating advantage, is found from

Sauto (V1  V2 ) I1 N 1
  1 2  1 (3.28)
S2  w V1I1 N1 n

From (3.28), we can see that the higher the number of turns of the common winding (N2),
the higher power rating is obtained. Compared with a two-winding transformer of the same
rating, auto transformers are smaller, more efficient, and have lower internal impedance. Three-
phase autotransformers are used extensively in power systems where the voltages of the two
systems coupled by the transformers do not differ by a factor greater than about three.

Example 3.1:

A two-winding transformer is rated at 60 kVA, 240/1200 V, 60 Hz. When operated as a

conventional two-winding transformer at rated, 0.8 power factor, its efficiency is 0.96. This
transformer is to be used as a 1440/1200-V step-down autotransformer in a power distribution
system.

(a) Assuming ideal transformer, find the transformer kVA rating when used as an
autotransformer.
(b) Find the efficiency with the kVA loading of part (a) and 0.8 power factor.

3.4. THREE-WINDING TRANSFORMER

Transformers having three windings are often used to interconnect three circuits which
may have different voltages. These winding are called primary, secondary and tertiary windings,

41
as shown in Figure 3.13. Typical applications of three-winding transformers in power systems
are for the supply of two independent loads at different voltages from the same source and
interconnection of two transmission systems of different voltages. Usually the tertiary windings
are used to provide voltage for auxiliary power purposes in the substation or to supply a local
distribution system.

Fig. 3.13: Single-phase three-winding transformer.

If the exciting current of a three-winding transformer is neglected, it is possible to draw a

simple single-phase equivalent T-circuit as shown in Figure 3.14.

Fig. 3.13: Equivalent circuit of three-winding transformer.

The parameters Zp, Zs and Zt are the impedances of the three separate windings referred to
the primary side. They are calculated via three tests which are similar in that in each case one
winding is open, one shorted, and reduced voltage is applied to the remaining winding. Then

42
 1
Z p  ( Z ps  Z pt  Z st )
2
1
Z s  ( Z ps  Z st  Z pt ) (3.28)
2
1
Z t  ( Z pt  Z st  Z ps )
2

where

Zps – impedance measured in the primary circuit with the secondary short-circuited and the
tertiary open.

Zpt – impedance measured in the primary circuit with the tertiary short-circuited and the
secondary open.

Zst – impedance measured in the secondary circuit with the tertiary short-circuited and the
primary open.

3.5. VOLTAGE CONTROL OF TRANSFORMERS

Voltage control in transformers are required to compensate for varying voltage drops in the
system and to control reactive power flow over transmission lines. Transformers may also be
used to control phase angle and, therefore, active power flow. The two commonly used methods
are tap changing transformers and regulating transformers.

3.5.1. Tap changing transformers

Practically all power transformers and many distribution transformers have taps in one or
more windings for changing the turns ratio. This method is the most popular since it can be used
for controlling voltages at all levels. There are two types of tap changing transformers

(i) Off-load tap changing transformers.

The off-load tap changing transformer requires the disconnection of the transformer when
the tap setting is to be changed. Off-load tap changers are used when it is expected that the ratio
will need to be changed only infrequently, because of load growth or some seasonal change. A
typical transformer might have four taps in addition to the nominal setting, with spacing of 2.5%
of full-load voltage between them. Such an arrangement provides for adjustments of up to 5%
above or below the nominal voltage rating of the transformer.

(ii) Tap changing under load (TCUL) transformers.

43
 TCUL is used when changes in ratio may by frequent or when it is undesirable to de-
energize the transformer to change a tap (an on-load changing type). It is used on transformers
and autotransformers for transmission tie, for bulk distribution units, and at other points of load
service. Basically, a TCUL transformer is a transformer with the ability to change taps while
power is connected. The range of ±10% is usually used in such special transformers.

Tapping on both ends of a radial transmission line can be adjusted to compensate for the
voltage drop in the line. Consider one phase of a three-phase transmission line with a step-up
transformer at the sending end and a step-down transformer at the receiving end of the line. A
single-line representation is shown in Figure 3.14, where tS and tR are the tap setting in per-unit.
In this diagram, V1' is the supply phase voltage referred to the high voltage side, and V2' is the
load phase voltage, also referred to the high voltage side. The impedance shown includes the line
impedance plus the referred impedances of the sending end and the receiving end transformers to
the high voltage side. If VS and VR are the phase voltages at both ends of the lines, we have

VR  VS  ( R  jX ) I (3.29)

The phasor diagram for the above equation is shown in Figure 3.14.

Fig. 3.14: Voltage phasor diagram.

The phase shift  between the two ends of the line is usually small, and we can neglect the
vertical components of VS. Approximating VS by its horizontal component results in

VS  VR  ab  de  VR  I R cos  I X sin  (3.29)

Substituting for I from P  VR I cos and Q  VR I sin  will result in

RP  XW
VS  VR  (3.30)
VR

Since VS  tSV1' and VR  tRV2' , the above relation in term of V1' and V2' becomes

44
 RP  XW
tS V1'  tR V2'  (3.31)
tR V2'

or

1  RP  XW 
tS   tR V2'   '   (3.32)
V1'  tR V2 
 

Assuming the product of tS and tR is unity, i.e., tStR = 1, and substituting for tR in (3.32), the
following expression is found for tS.

V2'
V1'
tS  (3.33)
RP  XW
1
V1' V2'

Example 3.2:

A three-phase transmission line is feeding from a 23/230 kV transformer at its sending end.
The line is supplying a 150 MVA, 0.8 power factor load through a step-down transformer of
230/23 kV. The impedance of the line and transformer at 230 kV is 18+j60 Ω. The sending end
transformer is energized from a 23 kV supply. Determine the tap setting for each transformer to
maintain the voltage at the load at 23 kV.

3.5.2. Regulating transformers or Boosters

Regulating transformer, also known as Boosters, are used to change the voltage magnitude
or phase angle at a certain point in the system by a small amount. A booster consists of an
exciting transformer and a series transformer.

Advantage

- The main transformers are free from tappings.

- The regulating transformers can be used at any intermediate point in the system.
- The regulating transformers and the tap changing gears can be taken out of service for
maintenance without affecting the system.

Voltage magnitude control

Figure 3.15 shows the connection of a regulating transformer for phase a of a three-phase
system for voltage magnitude control. Other phases have identical arrangement. The secondary
of the exciting transformer is tapped, and the voltage obtained from it is applied to the primary of
45
the series transformer. The corresponding voltage on the secondary of the series transformer is
added to the input voltage. Thus the output voltage is

Van'  Van  Van (3.34)

Since the voltages are in phase, a booster of this type is called an in-phase booster. The
output voltage can be adjusted by changing the excitation transformer taps. By changing the
switch from position 1 to 2, the polarity of the voltage across the series transformer is reversed,
so that the output voltage is now less than the input voltage.

Fig. 3.15: Regulating transformer for voltage magnitude control.

Phase angle control

Regulating transformers are also used to control the voltage phase angle. If the injected
voltage is out of phase with the input voltage, the resultant voltage will have a phase shift with
respect to the input voltage. Phase shifting is used to control active power flow at major intertie
buses. A typical arrangement for phase a of a three-phase system is shown in Figure 3.16.

Fig. 3.16: Regulating transformer for voltage phase angle control.

The series transformer of phase a is supplied from the secondary of the exciting
transformer bc. The injected voltage Vbc is in quadrature with the voltage Van, thus the resultant
voltage Van' goes through a phase shift , as shown in Figure 3.17. The output voltage is

Van'  Van  Vbc (3.35)

46
 Similar connections are made for the remaining phases, resulting in a balanced three phase
output voltage. The amount of phase shift can be adjusted by changing the excitation transformer
taps. By changing the switch from position 1 to 2, the output can be made to lag or lead the input
voltage.

Fig. 3.17: Voltage phasor diagram showing phase shifting effect for phase a.

Example 3.3:

The one-line diagram of a three-phase power system is shown in Figure 3.18. Select a
common base of 100 MVA and 22 kV on the generator side. Draw an impedance diagram with
all impedances including the load impedance marked in per-unit. The manufacturer’s data for
each device is given as follow:

Fig. 3.18: One-line diagram for example 3.3.

G: 90 MVA 22 kV X = 18 %

T1: 50 MVA 22/220 kV X = 18 %

T2: 40 MVA 220/11 kV X=6%

T3: 40 MVA 22/110 kV X = 6.4 %

47
 T4: 40 MVA 110/11 kV X = 8.0 %

M: 66.5 MVA 10.45 kV X = 18.5 %

The three-phase load at bus 4 absorbs 57 MVA, 0.6 power factor lagging at 10.45 kV. Line
1 and line 2 have reactances of 48.4 and 65.43 Ω, respectively.

3.6. ASSIGNMENTS

Lecture #3

CHAPTER IV
TRANSMISSION LINE PARAMETERS AND MODEL

The purpose of a transmission network is to transfer electric energy from generating units
at various locations to the distribution system which ultimately supplies the load. All
transmission lines in a power system exhibit the electrical properties of resistance, inductance,
capacitance, and conductance. The first part of this chapter deals with the determination of
inductance and capacitance of overhead lines. Then the effects of electromagnetic and
electrostatic induction are discussed.

48
4.1. OVERHEAD TRANSMISSION LINES

4.1.1. Transmission circuit components

A transmission circuit consists of conductors, insulators, support structures, and usually
shield wires.

Conductors

Aluminum has replaced copper as the most common conductor metal for overhead
transmission. One of the most common conductor types is aluminum conductor, steel-reinforced
(ACSR), which consists of layers of aluminum strands surrounding a central core of steel
strands, as shown in Figure 4.1. Stranded conductors are easier to manufacture, since larger
conductor sizes can be obtained by simply adding successive layers of strands. Stranded
conductors are also easier to handle and more flexible than solid conductors, especially in larger
sizes. For purposes of heat dissipation, overhead transmission-line conductors are bare (no
insulating cover).

Fig. 4.1: A typical 54/7 ACSR conductor.

Other conductor types include the all-aluminum conductor (AAC), all-aluminum-alloy

conductor (AAAC), aluminum conductor alloy-reinforced (ACAR), and aluminum-clad steel
conductor (Alumoweld). Aluminum conductor steel supported (ACSS) and the gap-type ZT-
aluminum conductor (GTZACSR) are used in transmission lines where the higher temperature
conductors are needed.

Extra high voltage (EHV) lines have more than one conductor per phase; these conductor
are called a bundle. Bundle conductors have a lower electric field strength at the conductor
surfaces, thereby controlling corona. They also have a smaller series reactance.

Insulator

Insulators for transmission lines above 69 kV are typically suspension-type insulators,

which consist of a string of discs constructed porcelain, toughened glass, or polymer. The
number of insulator discs in a strings increases with line voltage. Other types of discs include
larger units with higher mechanical strength and fog insulators for use in contaminated areas.
49
 Fig. 4.2: Cut-away view of a standard porcelain insulator disc for suspension insulator strings.

Support structures

Transmission lines employ a variety of support structures. Wood frame configurations are
commonly used for voltages of 345-kV and below. For 500- and 765-kV lines, the self-
supporting, lattice steel tower are typically used. Double-circuit 345-kV lines usually have self-
supporting steel towers with the phases arranged either in a triangular configuration to reduce
tower height or in a vertical configuration to reduce tower width.

Shield wires

Shield wires located above the phase conductors protect the phase conductors against
lightning. They are usually high- or extra-high-strength steel, Alumoweld, or ACSR with much
smaller cross section than the phase conductors. The number and location of the shield wires are
selected so that almost all lightning strokes terminate on the shield wires rather than on the phase
conductors.

4.1.2. Design factors

The design of a new transmission line has to take into account:

- The future system requirements of load growth and new generation.

- The points of interconnection of each line to the system, as well as the power and
voltage rating of each.

Electrical factors

Electrical design dictates the type, size, and number of bundle conductors per phase. Phase
conductors are selected to have sufficient thermal capacity to meet continuous, emergency
overload, and short-circuit current ratings.

50
 Electrical design also dictates the number of insulator discs, vertical or V-shaped string
arrangement, phase-to-phase clearance, and phase-to-tower clearance, all selected to provide
adequate line insulation. Line insulation must withstand transient overvoltages due to lightning
and switching surges. Reduced clearances due to conductor swings during winds must also be
accounted for.

The number, type, and location of shield wires are selected to intercept lightning strokes
that would otherwise hit the phase conductors. Also power footing resistance can be reduced by
using driven ground rods or a buried conductor running parallel to the line. Line height is
selected to satisfy prescribed conductor-to-ground clearances and to control ground-level electric
field and its potential shock hazard.

Conductor spacings, types, and sizes also determine the series impedance and shunt
admittance.

Mechanical factors

Mechanical design focuses on the strength of the conductors, insulator strings, and support
structures. Conductors must be strong enough to support a specified thickness of ice and a
specified wind in addition to their own weight. Suspension insulator strings must be strong
enough to support the phase conductors with ice and wind loading from tower to tower.
Conductor vibrations, which can cause conductor fatigue failure and damage to towers, are also
of concern. Vibrations are controlled by adjustment of conductor tensions, use of vibration
dampers, and – for bundle conductors – large bundle spacing and frequent use of bundle spacers.

Environmental factors

Environmental factors include land usage and visual impact. When a line route is selected,
the effect on local communities and population centers, land values, access to property, wildlife,
and use of public parks and facilities must all be considered. Also the biological effects of
prolonged exposure to electric and magnetic fields near transmission lines is of concern.

Economic factors

The optimum line design meets all the technical design criteria at lowest overall cost,
which includes the total installed cost of the line as well as the cost of line losses over the
operating life of the line.

51
4.2. LINE PARAMETERS

4.2.1. Resistance
The resistance of the conductor is very important in transmission efficiency evaluation and
economic studies. The DC resistance of a solid round conductor at a specified temperature is
given by

l
Rdc  (4.1)
A

Where:  - conductor resistivity

l - conductor length

A - conductor cross-sectional area

Two sets of units commonly used for calculating resistance, SI and English units, are
summarized in Table 4.1. In English units, conductor cross sectional area is expressed in circular
mils (cmil).

Table 4.1: Comparison of SI and English conductor resistance.

Resistivity depends on the conductor or metal. Annealed copper is the international

standard for measuring resistivity  (or conductivity , where  = 1/). Resistivity of conductor
metals is listed in Table 4.2.

52
 Table 4.2: % Conductivity, resistivity, and temperature constant of conductor metals.

The conductor resistance is affected by three factors: frequency, spiraling, and temperature.

When AC flows in a conductor, the current distribution is not uniform over the conductor
cross-sectional area and the current density is greatest at the surface of the conductor. This cause
the AC resistance to be somewhat higher than the DC resistance. This behavior is known as skin
effect. At 60 Hz, the AC resistance is about 2% higher than the DC resistance.

Since a stranded conductor is spiraled, each strand is longer than the finished conductor.
This results in a slightly higher resistance than the value calculated from (4.1).

The conductor resistance increases as temperature increases. This change can be

considered linear over the range of temperature normally encountered and may be calculated
from

T  t2
R2  R1 (4.2)
T  t1

where R2 and R1 are conductor resistances at t2 and t1 –Co, respectively. T is a temperature

constant that depends on the conductor material.

4.2.2. Conductance
Conductance accounts for real power loss between conductors or between conductors and
ground. For overhead lines, this power loss is due to leakage currents at insulators and to corona.

Insulator leakage current depends on the amount of dirt, salt, and other contaminants that
have accumulated on insulators, as well as on meteorological factors, particularly the presence of
moisture. Corona occurs when a high value of electric field strength at a conductor surface
53
causes the air to become electrically ionized and to conduct. The real power loss due to corona,
called corona loss, depends on meteorological conditions, particularly rain, and on conductor
surface irregularities.

Losses due to insulator leakage and corona are usually small compared to conductor I2R
loss. Conductance is usually neglected in power system studies because it is a very small
component of the shunt admittance.

4.2.3. Inductance

The inductance of a magnetic circuit that has a constant permeability µ can be obtained by
determining the following:

1. Magnetic field intensity H, from Ampere’s law

2. Magnetic flux density B (B = µH)
3. Flux linkage 
4. Inductance from flux linkage per ampere (L = /I)

Inductance of a single conductor

A current-carrying conductor produces a magnetic field around the conductor. The

magnetic flux lines are concentric closed circles with direction given by the right-hand rule. By
definition, for nonmagnetic material, the inductance L is the ratio of its total magnetic flux
linkage to the current I, given by


L (4.3)
I

where  is flux linkages, in Weber turns.

Consider a long round conductor with radius r, carrying a current I as shown in Figure 4.3.

54
 Fig. 4.3: Flux linkage of a long round conductor.

The magnetic field intensity Hx, around a circle of radius x, is constant and tangent to the
circle. The Ampere’s law relating Hx to the current Ix is given by
2 x

0
H x  dl  I x (4.4)

or

Ix
Hx  (4.5)
2 x

where Ix is the current enclosed at radius x. As shown in Figure 4.3, Equation (4.5) is all that is
required for evaluating the flux linkage  of a conductor. The inductance of the conductor can be
defined as the sum of contributions from flux linkages internal and external to the conductor.

Internal inductance

A simple expression can be obtained for the internal flux linkage by neglecting the skin
effect and assuming uniform current density throughout the conductor cross section, i.e.,

I I
 x2 (4.6)
r  x
2

Substituting for Ix in (4.5) yields

I
Hx  x (4.7)
2 r 2

55
For a nonmagnetic conductor with constant permeability µ0, the magnetic flux density is given
by Bx = µ0Hx, or

0 I
Bx  x (4.8)
2 r 2

where µ0 is the permeability of free space (or air) and is equal to 4 x 10-7 H/m. The total flux
linkage is found as follow

0 I r 3 I
4 0
int  x dx  0 (Wb / m) (4.9)
2 r 8

From (4.3), the inductance due to the internal flux linkage is

0 1
Lint    107 H /m (4.10)
8 2

Note that Lint is independent of the conductor radius r.

External inductance

Consider Hx external to the conductor at radius x > r as shown in Figure 4.4. Since the
circle at radius x encloses the entire current, Ix = I and in (4.5) Ix is replaced by I and the flux
density at radius x becomes

0 I
Bx  0 H x  (4.11)
2 x

Fig. 4.3: Flux linkage between D1 and D2.

56
 Since the entire current I is linked by the flux outside the conductor, the flux linkage dx is
numerically equal to the flux dx. the differential flux dx for a small region of thickness dx and
one meter length of the conductor is then given by

0 I
d x  dx  Bx dx  dx (4.12)
2 x

The external flux linkage between two points D1 and D2 is found by integrating dx from
D1 to D2.

0 I D2 1 D
ext 
2 
D1 x
dx  2  107 I ln 2
D1
(Wb/m) (4.13)

The inductance between two points external to a conductor is then

D2
Lext  2  107 ln (H/m) (4.14)
D1

Inductance of single-phase two-wire lines

Figure 4.4(a) shows a single-phase two-wire line consisting of two conduct x and y.
Conductor x with radius rx carries phasor current Ix = I referenced out of the page. Conductor y
with radius ry carries return current Iy = -I. These currents set up magnetic field lines that links
between the conductors as shown in Figure 4.4(b).

(a) (b)
Fig. 4.4: Single-phase two-wire line.

Inductance of conductor 1 due to internal flux is given by (4.10). The flux beyond D links
a net current of zero and does not contribute to the net magnetic flux linkages in the circuit.

57
Thus, to obtain the inductance of conductor 1 due to the net external flux linkage, it is necessary
to evaluate (4.14) from D1 = r1 to D2 = D.

D
L1( ext )  2  107 ln (H/m) (4.15)
r1

The total inductance of conductor 1 is then

1 D
L1   107  2  107 ln (H/m) (4.16)
2 r1

Equation (4.16) is often rearranged as follows:

1  7  
1
D 1
L1  2  107   ln   2  10  ln e 4
 ln  ln D 
4 r1   r1 
  (4.17)
7 
 ln D 
1
 2  10 ln
 
1

 r1e 
4

1

Let r  r1e
1
' 4
, the inductance of conductor 1 becomes

 1 
L1  2  107  ln '  ln D  (H/m) (4.18)
 r 
 1 

Similarly, the inductance of conductor 2 is

 1 
L2  2  107  ln '  ln D  (H/m) (4.19)
 r 
 2 

If the two conductors are identical, r1 = r2 = r and L1 = L2 = L, and the inductance per
phase per meter length of the line is given by

 1 
L  4  107  ln '  ln D 
 r 
(4.20)
1
 4  107 ln '  4  107 ln D (H/m)
r

The first term is considered as the inductance due to both the internal flux and that external
to conductor 1 to a radius of 1 m. The second term is dependent only upon conductor spacing.
This term is known as the inductance spacing factor.

The term r’ is known mathematically as the self-geometric mean distance of a circle with
radius r and is abbreviated by GMR. It is commonly referred to as geometric radius and will be
designated by Ds. Thus, the inductance per phase in mH/km becomes
58
 D
L  0.2ln (mH/km) (4.21)
Ds

Inductance of three-phase transmission lines with Symmetrical Spacing

Consider one meter length of a three-phase line with three conductors, each with radius r,
symmetrically spaced in a triangular configuration as shown in Figure 4.5.

Fig. 4.5: Three-phase line with symmetrical spacing.

Assuming balanced three-phase currents, we have

Ia  Ib  Ic  0 (4.22)

The total flux linkage of phase a conductor is

 1 1 1
a  2  107  I a ln  I b ln  I c ln  (4.23)
 D
'
r D

Substituting for Ib + Ic = -Ia

 1 1 D
a  2  107  I a ln  I a ln   2  107 I a ln ' (4.24)
 D
'
r r

The inductance of phase a is then

a D
La   2  107 ln (H/m) (4.25)
Ia r'

Due to symmetry, the same result is obtained for Lb and Lc, and the three inductances are
identical. Therefore, the inductance per phase per km length is

D
L  0.2ln (mH/km) (4.26)
Ds

Comparison of (4.26) and (4.21) shows that inductance per phase for a three-phase circuit
with equilateral spacing is the same as for one conductor of a single-phase circuit.
59
Inductance of three-phase transmission lines with asymmetrical Spacing

Practical transmission lines cannot maintain symmetrical spacing of conductors because of

construction considerations. With asymmetrical spacing, even with balanced currents, the
voltage drop due to line inductance will be unbalanced. Consider one meter length of a three-
phase line with three conductors, each with radius r. The conductor are asymmetrically spaced
with distances shown in Figure 4.6.

Fig. 4.6: Three-phase line with asymmetrical spacing.

The flux linkage of each phase is as follows:

 1 1 1 
a  2  107  I a ln '  I b ln  I c ln 
 r D12 D13 
 1 1 1 
b  2  107  I a ln  I b ln '  I c ln  (4.27)
 D12 r D23 
 1 1 1
c  2  107  I a ln  I b ln  I c ln ' 
 D13 D23 r 

Or in matrix form

  LI (4.28)

where the symmetrical inductance matrix L is given by

 1 1 1 
ln r ' ln
D12
ln
D13 

 1 1 1 
L  2  107 ln
D23 
ln ln (4.29)
 D12 r'
 1 1 1 
ln ln ln ' 
 D13 D23 r 

For balanced three-phase currents with Ia as reference, we have

60
 I b  I a 240o  a 2 I a
(4.30)
I c  I a 120o  aI a

Substituting (4.30) to (4.27) results in

a  1 1 1 
La   2  107  ln '  a 2 ln  a ln 
Ia  r D12 D13 
b  1 1 1 
Lb   2  107  a ln  ln '  a 2 ln  (4.31)
Ib  D12 r D23 
c  1 1 1
Lc   2  107  a 2 ln  a ln  ln ' 
Ic  D13 D23 r 

Equation (4.31) shows that the phase inductances are not equal and they contain an
imaginary term due to the mutual inductance.

Inductance of three-phase transposed transmission lines

A per-phase model of the transmission line is required in most power system analysis. One
way to regain symmetry in good measure and obtain a per-phase model is to consider
transposition. This consists of interchanging the phase configuration every one-third the length
so that each conductor is moved to occupy the next physical position in a regular sequence. Such
a transposition arrangement is shown in Figure 4.7.

Fig. 4.7: A transposed three-phase line.

Since in a transposed line each phase takes all three positions, the inductance per phase can
be obtained by finding the average value of (4.31)

La  Lb  Lc
L (4.32)
3

Noting a + a2 = 1120o + 1240o = -1, the average of (4.32) becomes

61
 2  107  1 1 1 1 
L  3ln '  ln  ln  ln  (4.33)
3  r D12 D23 D13 

or

 
   2  107 ln D12 D23 D13
3
7 1 1
L  2  10 ln '  ln (4.34)
 r 1
 r'
3 
 ( D D D
12 23 13 ) 

In modern transmission lines, transposition is not generally used. However, for the purpose
of modeling, it is most practical to treat the circuit as transposed. The error introduced as a result
of this assumption is very small.

Inductance of three-phase composited transmission lines

Consider a single-phase line consisting of two composite conductors x and y as shown in

Figure 4.8. The current in x is I referenced into the page, and the return current in y is –I.
Conductor x consists of n identical strands or subconductors, each with radius rx. Conductor y
consists of m identical strands or subconductors, each with radius ry. The current is assumed to
be equally divided among the subconductors. The current per strand is I/n in x and I/m in y. The
total flux linkage of conductor a is expressed as follows:

Fig. 4.8: Single-phase line with two composite conductors.

I 1 1 1 1 
a  2  107  ln '  ln  ln  ...  ln 
n r Dab Dac Dan 
(4.35)
I  1 1 1 1 
- 2  107  ln  ln  ln  ...  ln 
m  Daa Dab
' Dac ' Dam 
'

or

m Daa' Dab' Dac' ...Dam

a  2  10 I ln
7
(4.36)
n rx' Dab Dac ...Dan

62
The inductance of subconductor a is

a m D D D ... D
La   2n  107 ln
' 'am '
aa ab ac
(4.37)
I /n n r ' D D ... D
x ab ac an

The inductance of other subconductor in x are similarly obtained. For example, the
inductor of the nth subconductor is

n m D D D ... D
Ln   2n  107 ln
' 'nm '
na nb nc
(4.38)
I /n n r ' D D ... D
x na nb nc

The average inductance of any one subconductor in group x is

La  Lb  Lc  ...  Ln
Lav  (4.39)
n

Since all the subconductors of conductor x are electrically parallel, the inductance of x will be

Lav La  Lb  Lc  ...  Ln
Lx   (4.40)
n n2

Substituting the values of La, Lb, Lc,…, Ln in (4.40) results in

mn (Daa' Dab' ... Dam )...(Dna' Dna' ... Dam )

Lx  2  107 ln 2
(4.41)
n (Daa Dab ... Dan )...(Dna Dnb ... Dnm )

where Daa  Dbb  ...  Dnn  rx'

4.2.4. Capacitance
Transmission line conductors exhibit capacitance with respect to each other due to the
potential difference between them.

Capacitance of single-phase lines

Consider one meter length of a single-phase line consisting of two long solid round
conductors each having a radius r as shown in Figure 4.9. The two conductors are separated by a
distance D. Conductor 1 carries a charge of q1 coulombs/meter and conductor 2 carries a charge
of q2 coulombs/meter.

63
 Fig. 4.9: Single-phase two-wire line.

Assuming conductor 1 alone to have a charge of q1, the voltage between conductor 1 and 2
is

q1 D
V12(q1 )  ln (4.42)
2 0 r

Now assuming only conductor 2 have a charge of q2, the voltage between conductors 2 and
1 is

q2 D
V21(q2 )  ln (4.43)
2 0 r

Since V12(q2 )  V21(q2 ) , we have

q2 r
V12(q2 )  ln (4.44)
2 0 D

From the principle of superposition, the potential difference due to presence of both charges is

q1 D q r q D
V12  V12(q1 )  V12(q2 )  ln  2 ln  ln (F/m) (4.45)
2 0 r 2 0 D  0 r

By definition, the capacitance C is the ratio of charge q to the voltage V, given by

q  0
C12   (F/m) (4.46)
V12 ln D
r

Equation (4.68) gives the line-to-line capacitance between the conductors. For the purpose
of transmission line modeling, we find it convenient to define a capacitance C between each
conductor and a neutral as illustrated in Figure 4.10.

Fig. 4.9: Illustration of capacitance to neutral.

Since the voltage to neutral is half of V12, the capacitance to neutral C = 2C12, or

2 0
C (F/m) (4.46)
D
ln
r

64
Capacitance of three-phase lines

Consider one meter length of a three-phase line with three long conductors, each with
radius r, with conductor spacing as shown in Figure 4.10.

Fig. 4.10: Three-phase transmission line.

Firstly, the potential difference between conductors i and j due to the presence of n charges
is

1 n Dkj
Vij 
2 0
q
k 1
k ln
Dki
(4.47)

Applying (4.47) to the three sections of the transposition, we have

1  D12 r D 
Vab (I)   qa ln  qb ln  qc ln 23 
2 0  r D12 D13 
1  D23 r D 
Vab (II)   qa ln  qb ln  qc ln 13  (4.48)
2 0  r D23 D12 
1  D13 r D 
Vab (III)   qa ln  qb ln  qc ln 12 
2 0  r D13 D23 

The average value of Vab is

1  D12 D23 D13 r3 D D D 

Vab   a
q ln  q ln  qc ln 12 23 13 
6 0  3 b
r D12 D23 D13 D12 D23D13 
 1
 (4.49)

1 
qa ln
 D12 D23 D13  3
 qb ln
r 

2 0
 D12 D23D13  3 
1
r


Similarly, we find the average voltage Vac as

 1

Vac 
1 
qa ln
 D12 D23 D13  3
 qc ln
r  (4.50)

2 0
 D12 D23D13  3 
1
r

65
Adding (4.49) and (4.50) and substituting for qb + qc = -qa, we have

 1

Vab  Vac 
1 
2qa ln
 D12 D23 D13  3
 qa ln
r 

2 0
 D12 D23D13  3 
1
r
 (4.51)
1


3qa D D D 
ln 12 23 13
3

2 0 r

For balanced three-phase voltages,

Vab  Van0o  Van  120o

(4.52)
Vac  Van0o  Van  240o

Therefore,

Vab  Vac  3Van (4.53)

Substituting in (4.51) the capacitance per phase to neutral is

qa 2 0
C  1
(F/m) (4.54)
Van
ln
 D12 D23D13  3

Capacitance of bundle three-phase lines

The procedure for finding the capacitance per phase for a three-phase transposed line with
bundle conductors follows the same steps as the procedure in the previous part. The capacitance
per phase is found to be

2 0
C 1
(F/m) (4.55)
D D D 
ln 12 23 13
3

rb

The effect of bundling is to introduce an equivalent radius rb. If d is the bundle spacing, we
obtain for the two-subconductor bundle

rb  r  d (4.56)

for the three-subconductor bundle

rb  3 r  d 2 (4.57)

for the four-subconductor bundle

66
 r b  1.09 4 r  d 3 (4.58)

4.3. ASSIGNMENT

67
 CHAPTER V
TRANSMISSION LINE MODEL AND PERFORMANCE

In Chapter 4 the per-phase parameters of transmission lines were obtained. This chapter
deals with the representation and performance of transmission lines under normal operating
conditions. Transmission lines are represented by an equivalent model with appropriate circuit
parameters on a “per-phase” basis. The model used to calculate voltages, currents, and power
flows depends on the length of the line. In this chapter the circuit parameters, voltage and current
relations are first developed for “short” and “medium” lines. Next, long line theory is presented
and expressions for voltage and current along the distributed line model are obtained.

5.1. SHORT LINE MODEL

Capacitance may often be ignored without much error if the lines are less than about 80 km
long, or if the voltage is not over 69 kV. The short line model is obtained by multiplying the
series impedance per unit length by the line length.

Z   r  j L  l
(5.1)
= R  jX

where r and L are the per-phase resistance and inductance per unit length, respectively, and l is
the line length. The short line model on a per-phase basis is shown in Figure 5.1. VS and IS are the
phase voltage and current at the sending end of the line, and VR and IR are the phase voltage and
current at the receiving end of the line.

Fig. 5.1: Short line model.

If a three-phase load with apparent power SR(3) is connected at the end of the transmission
line, the receiving end current is obtained by

68
 S R* (3 )
IR  (5.2)
3VR*

The phase voltage at the sending end is

VS  VR  ZI R (5.3)

and since the shunt capacitance is neglected, the sending end and the receiving end current are
equal, i.e.,

IS  IR (5.4)

The transmission line may be represented by a two-port networks as shown in Figure 5.2,
and the above equations can be written in terms of the generalized circuit constants commonly
known as the ABCD constants.

Fig. 5.2: Two-port representation of a transmission line.

VS  AVR  BI R
(5.5)
I S  CVR  DI R

or in matrix form

VS   A B  VR 
 I   C D   I  (5.6)
 S   R

According to (5.3) and (5.4), for short line model

A = 1; B = Z; C = 0; D=1 (5.7)

Voltage regulation is the change in voltage at the receiving end of the line when the load
varies from no-load to as specified full load at a specified power factor, while the sending-end
voltage is held constant. Expressed in percent of full-load voltage

VR (NL)  VR (FL)
%VR   100 (5.8)
VR (FL)

At no-load situation, IR = 0 and from (5.5)

69
 VS
VR (NL)  (5.9)
A

For short line, A = 1 so that VR(NL) = VS. Voltage regulation is a measure of line voltage drop and
depends on the load power factor. The effect of load power factor on voltage regulation is
illustrated by the phasor diagrams in Figure 5.3. The phasor diagrams are graphical
representation of (5.5) for lagging, unity and leading power factor loads.

Fig. 5.3: Phasor diagrams for a short transmission line.

As shown, the higher voltage regulation occurs for the lagging p.f. load, where VRNL
exceeds VRFL by the larger amount. A smaller or even negative voltage regulation occurs for the
leading p.f. load.

In practice, transmission-line voltages decrease when heavily loaded and increase when
lightly loaded. When voltages on EHV lines are maintained within ±5% of rated voltage,
corresponding the about 10% voltage regulation, unusual operating problems are not
encountered. 10% voltage regulation for lower voltage lines including transformer-voltage drops
is also considered good operating practice.

Once he sending-end power is obtained by the sending end voltage

SS (3 )  3VS I S* (5.10)

The total line loss is the given by

SL(3 )  SS (3 )  SR (3 ) (5.11)

and the transmission line efficiency is given by

PR (3 )
 (5.12)
PS (3 )

where PR(3) and PS(3) are the total real power at the receiving end and sending end of the line,
respectively.

Example 5.1:
70
 A 220-kV, three-phase transmission line is 40 km long. The resistance per phase is 5.15
Ω/km and the inductance per phase is 1.3263 mH/km. The shunt capacitance is negligible. Use
the short line model to find the voltage, power at the sending end, the voltage regulation and
efficiency when the line is supplying a three-phase load of

(a) 381 MVA at 0.8 p.f. lagging at 220 kV.

(b) 381 MVA at 0.8 p.f. leading at 220 kV.

Solution

(a) The series impedance per phase is

Z   r  jL  l  (0.15  j 2  60  1.3263  103 )  40  6  j 20 ()
The receiving end voltage per phase is
2200o
VR   1270o (kV)
3

The apparent power is

SR (3 )  381 cos1 0.8  38136.87o  304.8  j 228.6 (MVA)

The current per phase is given by

S *R ( 3 ) 381  36.87o  103

IR    1000 - 36.87o (A)
3VR* 3  1270 o

From (5.5) the sending end voltage is

VS  VR  ZI R  1270o  (6  j 20)  (1000 - 36.87o )  103

 144.334.93o (kV)

The sending end line-to-line voltage magnitude is

VS (L L)  3 VS  250 (kV)

The sending end power is

SS (3 )  3VS I S*  3  144.334.93o  100036.87o  103

 322.8 MW  j288.6 MVar
 43341.8o (MVA)

Voltage regulation is

250  220
%VR   100  13.6%
220

71
Transmission line efficiency is

PR (3 ) 304.8
   100  94.4%
PS(3 ) 322.8

(b) The current for 381 MVA with 0.8 leading p.f. is

S *R ( 3 ) 38136.87o  103
IR    100036.87o (A)
3VR* 3  1270o

The sending end voltage is

VS  VR  ZI R  1270o  (6  j 20)  (100036.87o )  103

 121.399.29o (kV)

The sending end line-to-line voltage magnitude is

VS (L L)  3 VS  210.26 (kV)

The sending end power is

SS (3 )  3VS I S*  3  121.399.29o  1000  36.87o  103

 322.8 MW  j168.6 MVar
 364.18  27.58o (MVA)

Voltage regulation is

210.26  220
%VR   100  4.43%
220

Transmission line efficiency is

PR (3 ) 304.8
   100  94.4%
PS(3 ) 322.8

5.2. MEDIUM LINE MODEL

As the length of line increases, the shunt capacitance must be considered. Line above 80
km and below 250 km in length are termed as medium length lines. For the medium lines, half of
the shunt capacitance may be considered to be lumped at each end of the line. This is referred to
as the nominal  model and is shown in Figure 5.4. Z is total series impedance of the line, and Y
is the total shunt admittance of the line given by

Y   g  jC  l (5.13)

72
 Fig. 5.4: Nominal  model for medium length line.

Under normal conditions, the shunt conductance per unit length, which represents the
leakage current over the insulators and due to corona, is negligible and g is assumed to be zero.
C is the line to neutral capacitance per km, and l is the line length. The sending end voltage and
current for the nominal  model are obtained as follows:

From KCL the current in the series impedance designated by IL is

Y
I L  I R  VR (5.14)
2

From KVL the sending end voltage is

VS  VR  ZI L (5.15)

Substituting for IL from (5.14), we obtain

 ZY 
VS   1  VR  ZI R (5.16)
 2 

The sending end current is

Y
I S  I L  VS (5.17)
2

Substituting for IL and VS

 ZY   ZY 
I S  Y 1  VR   1   IR (5.18)
 4   2 

Therefore, the ABCD constants for the nominal  model are given by

73
  ZY 
A  D  1  
 2 
BZ (5.19)
 ZY 
C  Y 1  
 4 

Note

- A medium-length line could also be approximated by the T circuit, lumping half of the
series impedance at each end of the line.
- Since we are dealing with a linear, passive, bilateral two-port network, the determinant
of the transmission matrix is unity, i.e.,
AD  BC  1 (5.20)

5.3. LONG LINE MODEL

For lines 250 km and longer and for more accurate solution, the exact effect of the
distributed parameters must be considered. In this section expressions for voltage and current at
any point on the line are derived. Then, based on these equations an equivalent  model is
obtained for long line. Figure 5.5 shows one phase of a distributed line of length l km.

Fig. 5.5: Long line with distributed parameters.

The series impedance per unit length is shown by the lowercase letter z, and the shunt
admittance per phase is shown by the lowercase letter y, where z = r + jL and y = g + jC.
Consider a small segment of line x at a distance x from the receiving end of the line. The phasor
voltages and currents on both sides of this segment are shown as a function of distance. From
KVL we have

V (x   x)  V (x)  zxI (x) (5.21)

or
74
 V (x   x)  V (x)
 zI (x) (5.22)
x

Taking the limit as x0, we have

dV (x)
 zI (x) (5.23)
dx

Also, from KCL

I (x   x)  I (x)  yxV (x   x) (5.24)

or

I (x   x)  I (x)
 yV (x   x) (5.25)
x

Taking the limit as x0, we have

dI (x)
 yV (x) (5.26)
dx

Differentiating (5.23) and substituting from (5.26), we get

d 2V (x) dI (x)
2
z  zyV (x) (5.27)
dx dx

Let 2 = zy, the following second-order differential equation will result.

d 2V (x)
2
  2V (x)  0 (5.28)
dx

The solution of the above equation is

V (x)  A1e x  A1e x (5.29)

where  is known as the propagation constant, is a complex expression as

  zy   r  jL g  jC     j (5.30)

The real part  is known as the attenuation constant, and the imaginary component  is
known as the phase constant.  is measured in radian per unit length.

From (5.23), the current is

1 dV ( x ) 
I ( x)    A1e x  A2e x 
z dx z
(5.31)

y
z
 A1e x  A2e x 
75
or

I ( x) 
1
Zc
 A1e x  A2e x  (5.32)

where Zc is known as the characteristic impedance, given by

z
Zc  (5.33)
y

To find the constant A1 and A2 we note that when x = 0, then V(x) = VR and I(x) = IR. From
(5.29) and (5.32) these constants are found to be

VR  Z c I R
A1 
2
(5.34)
V  Zc I R
A2  R
2

Upon substitution in (5.29) and (5.36), the general expressions for voltage and current
along a long transmission line become

VR  Zc I R  x VR  Zc I R  x
V (x)  e  e (5.35)
2 2

VR VR
 IR  IR
Zc x Zc
I (x)  e  e  x (5.36)
2 2

The equation for voltage and currents can be rearranged as follows:

e x  e x e x  e x
V (x)  VR  Zc IR (5.37)
2 2

1 e x  e x e x  e x
I (x)  VR  IR (5.38)
Zc 2 2

Recognizing the hyperbolic functions sinh, and cosh, the above equations are written as
follows:

V (x)  cosh  xVR  Zc sinh  xI R (5.39)

1
I (x)  sinh  xVR  cosh  xI R (5.40)
Zc

We are particularly interested in the relation between the sending end and the receiving end
of the line. Setting x = l, V(l) = V(s) and I(l) = Is. the result is

76
 Vs  cosh  lVR  Zc sinh  lI R (5.41)

1
Is  sinh  lVR  cosh  lI R (5.42)
Zc

Therefore, the ABCD constants for long transmission line are

1
A  cosh  l C sinh  l
Zc (5.43)
B  sinh  l D  cosh  l

It is now possible to find an accurate equivalent  model, shown if Figure 5.6, to replace
the ABCD constants of the two-port network. Similar to the expressions (5.16) and (5.18)
obtained for the nominal , for the equivalent  model we have

 Z 'Y ' 
VS  1  VR  Z I R
'
(5.44)
 2 

 Z 'Y '   Z 'Y ' 

I S  Y 1 
'
VR  1   IR (5.45)
 4   2 

Comparing (5.44) and (5.45) with (5.41) and (5.42), respectively, and making use of the identity

 l cosh  l  1
tanh  (5.46)
2 sinh  l

The parameters of the equivalent  model are obtained.

sinh  l
Z '  Zc sinh  l  Z (5.47)
l

Y' 1  l Y tanh  l / 2
 tanh  (5.48)
2 Zc 2 2 l / 2

77
 Fig. 5.6: Equivalent  model for long length line.

5.4. ASSIGNMENT

78
 CHAPTER VI
POWER FLOW ANALYSIS

In the previous chapters, modeling of the major components of an electric power system
was discussed. This chapter deals with the steady-state analysis of an interconnected power
system during normal operation. Successful power system operation under normal balanced
three-phase steady-state conditions requires the following:

1. Generation supplies the demand (load) plus losses.

2. Bus voltage magnitudes remain close to rated values.
3. Generators operate within specified real and reactive power limits.
4. Transmission lines and transformers are not overloaded.

The power-flow computer program is the basic tool for investigating these requirements.
This program computes the voltage magnitude and angle at each bus in a power system under
balanced three-phase steady-state conditions. It also computes real and reactive power flows for
all equipment interconnecting the buses, as well as equipment losses.

Conventional nodal or loop analysis is not suitable for power-flow study because the input
data for loads are normally given in term of power, not impedance. Also, generators are
considered as power sources, not voltage or current sources. The power-flow problem is
therefore formulated as a set of nonlinear algebraic equations suitable for computer solution.

In this chapter, some basic methods including direct and iterative techniques for solving
algebraic equations. Then new solution methods for power-flow problem, Gauss-Seidel and
Newton-Raphson are presented. Finally, the DC power flow and the power-flow representation
of wind turbine generators are discussed.

6.1. BUS ADMITTANCE MATRIX

In order to obtain the node-voltage equations, consider the simple power system shown in
Figure 6.1 where impedances are expressed in per unit on a common MVA base and for
simplicity resistances are neglected.

79
 Fig. 6.1: The impedance diagram of a simple system

Since the nodal solution is based upon KCL, impedances are converted to admittance, i.e.,

1 1
yij   (6.1)
zij rij  jxij

The circuit has been redrawn in Figure 6.2 in term of admittances and transformation to
current sources. Node 0 (which is normally ground) is taken as reference.

Fig. 6.2: The admittance diagram for system of Figure 6.1.

Applying KCL to the independent nodes 1 through 4 results in

80
 I1  y10V1  y12 (V1  V2 )  y13 (V1  V3 )
I 2  y20V2  y12 (V2  V1 )  y23 (V2  V3 )
0  y23 (V3  V2 )  y13 (V3  V1 )  y34 (V3  V4 )
0  y34 (V4  V3 )

Rearranging these equations yields

I1  ( y10  y12  y13 )V1  y12V2  y13V3

I 2   y12V1  ( y20  y12  y23 )V2  y23V3
0   y13V1  y23V2  ( y13  y23  y34 )V3  y34V4
0   y34V3  y34V4

We introduce the following admittances

Y11  y10  y12  y13 Y12  Y21   y12

Y22  y20  y12  y23 Y13  Y31   y13
Y33  y13  y23  y34 Y23  Y32   y23
Y44  y34 Y34  Y43   y34

The node equation reduces to

I1  Y11V1  Y12V2  Y13V3  Y14V4

I 2  Y21V1  Y22V2  Y23V3  Y24V4
I 3  Y31V1  Y32V2  Y33V3  Y34V4
I 4  Y41V1  Y42V2  Y43V3  Y44V4

Extending the above relation to an n bus system, the node-voltage equation in matrix form is

 I1  Y11 Y12 Y1i Y1n  V1 

 I  Y Y22 Y2i Y2 n  V2 
 2   21  
    
    (6.2)
 Ii  Yi1 Yi 2 Yii Yin  Vi 
    
    
 I n  Yn1 Yn 2 Yni Ynn  Vn 

or

Ibus  YbusVbus (6.3)

where the Ibus is the vector of the injected bus currents (i.e., external current sources). The current
is positive when flowing towards the bus, and it is negative if flowing away from the bus. Vbus is
the vector of bus voltages measured from the reference node (i.e., node voltages). Ybus is known

81
as the bus admittance matrix. The diagonal element of each node is the sum of admittances
connected to it. It is known as the self-admittance or driving point admittance
n
Yii   yij j i (6.4)
j 0

The off-diagonal element is equal to the negative of the admittance between the nodes. It is
known as the mutual admittance or transfer admittance. It is to be noted here that if more than
one components are connected in parallel between the two nodes, the equivalent primitive
admittance of the components is first obtained before determining the entry in the Ybus.

Yij  Y ji  - yij (6.5)

When the bus current are known, (6.2) can be solved for the n bus voltages
1
Vbus  Ybus Ibus (6.6)

The inverse of the bus admittance matrix is known as the bus impedance matrix Zbus. The
admittance matrix obtained with one of the buses as reference is non-singular. Otherwise the
nodal matrix is singular.

Based on (6.3) and (6.4), the bus admittance matrix for the network in Figure 6.2 obtained
by inspection is

  j8.50 j 2.50 j5.00 0

 j 2.50  j8.75 j5.00 0
Ybus   
 j5.00 j5.00  j 22.50 j12.50 
 
 0 0 j12.50  j12.50 

6.2. SOLUTION OF NONLINEAR ALGEBRAIC EQUATIONS

The most common techniques used for the iterative solution of nonlinear algebraic
equations are Gauss-Seidel, Newton-Rapshon, and Quasi-Newton methods.

6.2.1. Gauss-Seidel method

General Gauss-Seidel method

The Gauss-Seidel method is also known as the method of successive displacements. To

illustrate the technique, consider the solution of the nonlinear equation given by

f ( x)  0 (6.7)

The above function is rearranged and written as

82
 x  g (x) (6.8)

If x(k) is an initial estimate of the variable x, the following iterative sequence is formed.

x( k 1)  g ( x( k ) ) (6.9)

A solution is obtained when the difference between the absolute value of the successive iteration
is less than a specified accuracy

x( k 1)  x( k )   (6.10)

Where  is the desired accuracy.

Example 6.2

Use the Gauss-Seidel method to find a root of the following equation

f ( x )  x3  6 x 2  9 x  4  0

Solution

Solving for x, the above expression is written as

1 6 4
x   x3  x 2   g ( x )
9 9 9

Applying the Gauss-Seidel algorithm, and use an initial estimate of

x(0)  2

From (6.9), the first iteration is

1 6 4
x(1)  g (2)   (2)3  (2)2   2.2222
9 9 9

The second iteration is

1 6 4
x(2)  g (2.2222)   (2.2222)3  (2.2222)2   2.5173
9 9 9

The subsequent iterations result in 2.8966, 3.3376, 3.95968, 3.9988 and 4.0000. The process is
repeated until the charge in variable is within the desired accuracy.

Note

- Gauss-Seidel method needs many iterations to achieve the desired accuracy.

- No guarantee for the convergence.
- If the initial estimate is chosen much different from the solution, the process would
diverge.
83
Gauss-Seidel method with an acceleration factor

In some cases, an acceleration factor can be used to improve the rate of convergence. If 
> 1 is the acceleration factor, the Gauss-Seidel algorithm becomes

x( k 1)  x(k)  [ g ( x( k ) )  x (k) ] (6.11)

Example 6.3

Use the Gauss-Seidel method with an acceleration factor of  = 1.25 to find a root of the
following equation

f ( x )  x3  6 x 2  9 x  4  0

Solution

Starting with an initial estimate of x(0) = 2, and using (6.11), the first iteration is

1 6 4
g (2)   (2)3  (2)2   2.2222
9 9 9

x(1)  x(0)  [ g ( x(0) )  x(0) ]  2  1.25[2.2222  2]  2.2778

The second iteration is

1 6 4
g (2.2778)   (2.2778)3  (2.2778)2   2.5902
9 9 9

x(2)  2.2778  1.25[2.5902  2.2778]  2.6683

The subsequent iterations result in 3.0801, 3.1831, 3.7238, 4.0084, 3.9978 and 4.0005. Care
must be taken not to use a very large acceleration factor since the larger step size may result in
an overshoot. This can cause an increase in the number of iterations or even result in divergence.

Gauss-Seidel method for n equations in n variables

f1 ( x1 , x 2 ,..., xn )  c1
f 2 ( x1 , x 2 ,..., xn )  c2
(6.12)
..............................
f n ( x1 , x 2 ,..., xn )  cn

Solving for one variable from each equation, the above functions are rearranged and written as

x1  c1  g1 ( x1 , x 2 ,..., xn )
x2  c2  g1 ( x1 , x 2 ,..., xn )
(6.13)
..............................
xn  cn  g n ( x1 , x 2 ,..., xn )
84
 The iteration procedure is initiated by assuming an approximate solution for each of the
independent variables  x1(0) , x2(0) ,..., xn(0)  . Equation (6.13) results in a new approximate solution

x(1)
1 , x2(1) ,..., xn(1)  . At the end of each iteration, the calculated values of all variables are tested

against the previous values. If all changes in the variables are within the specified accuracy, a
solution has converged, otherwise another iteration must be performed.

Note

The rate of convergence can often be increased by using a suitable acceleration factor ,
and the iterative sequence becomes

xi( k 1)  xi( k )   ( xi( kcal

1)
 xi( k ) ) (6.14)

6.2.2. Newton-Raphson method

Newton-Raphson method for one-dimensional equation

Consider the solution of the one-dimensional equation given by

f ( x)  c (6.15)

If x(0) is an initial estimate of the solution, and x(0) is a small deviation from the correct
solution, we must have

f ( x(0)  x(0) )  c (6.16)

Expanding the left-hand side of the above equation in Taylor’s series about x(0) yields
(0)
1  d2 f 
(0)
 df 
f ( x )    x(0)   2  ( x(0) )2  ...  c
(0)
(6.17)
 dx  2!  dx 

Assuming the error x(0) is very small, the higher-order terms can be neglected, which results in

(0)
 df 
f ( x )    x(0)  c
(0)
(6.18)
 dx 

Or
(0)
 df 
c (0)
 c  f ( x )    x(0)
(0)
(6.19)
 dx 

Adding x(0) to the initial estimate will result in the second approximation

85
 c (0)
x x (1) (0)
 (0) (6.20)
 df 
 
 dx 

Successive use of this procedure yields the Newton-Raphson algorithm

c ( k )  c  f ( x ( k ) )
c ( k )
x (k )
 (k ) (6.21)
 df 
 
 dx 
x ( k 1)  x ( k )  x ( k )

Example 6.4

Use the Newton-Raphson method to find a root of the following equation. Assume an initial
estimate of x(0) = 6.

f ( x )  x3  6 x 2  9 x  4  0

Solution

The analytical solution given by the Newton-Raphson algorithm is

df
 3x 2  12 x  9
dx

c(0)  c  f ( x(0) )  0  [(63 )  6  (62 )  9  (6)  4]  50

(0)
 df 
   3  (6)2  12  (6)  9  45
 dx 

c (0) 50
x (0)
 (0)
  1.1111
 df  45
 
 dx 

Therefore, the result at the end of the first iteration is

x(1)  x(0)  x(0)  6  1.1111  4.8889

The subsequent iterations result in

86
 13.4431
x (2)  x (1)  x (1)  4.8889   4.2789
22.037
2.9981
x (3)  x (2)  x (2)  4.2789   4.0405
12.5797
0.3748
x (4)  x (3)  x (3)  4.0405   4.0011
9.4914
0.0095
x (5)  x (4)  x (4)  4.0011   4.0000
9.0126

We see that Newton’s method converges considerably more rapidly than the Gauss-Seidel
method. The method may converge to a root different from the expected one or diverge if the
starting value is not close enough to the root.

Newton-Raphson method for n-dimensional equations

Consider the n-dimensional equations given by (6.22).

f1 ( x1 , x 2 ,..., xn )  c1
f 2 ( x1 , x 2 ,..., xn )  c2
(6.22)
..............................
f n ( x1 , x 2 ,..., xn )  cn

Expanding the left-hand side of the equation (6.22) in the Taylor’s series about the initial
estimates and neglecting all higher order terms, leads to the expression
(0) (0) (0)
 f   f   f 
f1 ( x )   1  x1(0)   1  x2(0)  ...   1  xn(0)  c1
(0)

 x1   x2   xn 

(0) (0) (0)
 f   f   f 
f 2 ( x )   2  x1(0)   2  x2(0)  ...   2  xn(0)  c2
(0)

 x1   x2   xn  (6.23)

(0) (0) (0)

 f   f   f 
f n ( x )   n  x1(0)   n  x2(0)  ...   n  xn(0)  cn
(0)

 x1   x2   xn 

or in matrix form

C (k)  J (k) X (k) (6.24)

where

 x1(k)   c1  f1(k) 
 (k)   
x c2  f 2(k) 
X (k)  2  and C (k)  
   
 (k)   (k) 
 xn  cn  f n 

87
  f (k)  f1 
(k)
 f1  
(k)

 1      
 x1   x2    xn  
 (k) (k) (k) 
  f 2   f 2   f 2  
  x1      
J (k)
  x2    xn   (6.25)
 
 (k) (k) (k) 
  f n   f n   f n  
  x      
 1   x2    xn  

And the Newton-Raphson algorithm for the n-dimensional case becomes

X (k )  X (k)  X (k) (6.26)

Note

- J(k) is called the Jacobian matrix. Elements of this matrix are the partial derivatives
evaluated at X(k). It is assumed that J(k) has an inverse during each iteration.
- This method, as applied to a set of nonlinear equations, reduces the problem to solving
a set of linear equations in order to determine the values that improve the accuracy of
the estimates.

6.3. POWER FLOW SOLUTION

Power flow studies, commonly known as load flow, form an important part of power
system analysis. They are necessary for planning, economic scheduling, and control of an
existing system as well as planning its future expansion. The problem consists of determining the
magnitudes and phase angle of voltages at each bus and active and reactive power flow in each
line.

In solving a power flow problem, the system is assumed to be operating under balanced
conditions and a single-phase model is used. Four quantities are associated with each bus,
including:

- Voltage magnitude V

- Phase angle 
- Active power P
- Reactive power Q

The system buses are generally classified into three types.

88
 1. Slack bus (Swing bus): There is only one slack bus in a certain system, which for
convenience is numbered bus 1 in this text (bus 0 somewhere else). The magnitude and
phase angle of the voltage in slack bus are specified. The power flow program computes P
and Q.
2. Load bus (PQ bus): At these buses the active and reactive power are specified. The
magnitude and the phase angle of the bus voltages are unknown. Most buses in a typical
power-flow problem are load-bus type.
3. Voltage controlled bus (PV bus): At these buses, the active power and voltage magnitude
are specified. The phase angles of the voltages and the reactive power are to be
determined. Examples are buses to which generators, switched shunt capacitors, or static
var compensators are connected. Another example is a bus to which a tap-changing
transformer is connected, the power-flow program then computes the tap setting.

6.3.1. Power Flow Equation

Consider a typical bus of a power system network as shown in Figure 6.3. Transmission
lines are represented by their equivalent  models where impedances have been converted to per
unit admittances on a common MVA base.

Fig. 6.3: A typical bus of the power system.

Application of KCL to this bus results in

Ii  yi 0Vi  y i1 (Vi  V1 )  yi 2 (Vi  V2 )  ...  yin (Vi  Vn )

(6.27)
 ( yi 0  y i1  yi 2  ...  yin )Vi  yi1V1  yi 2V2  ...  yinVn

or
n n
Ii  Vi  yij   yijV j j i (6.28)
j 0 j 1

89
The active and reactive power at bus i is

Pi  jQi  Vi Ii* (6.29)

or

Pi  jQi
Ii  (6.30)
Vi *

Substituting for Ii in (6.28) yields

Pi  jQi n n
 Vi  ij y   yijV j j i (6.31)
Vi * j 0 j 1

From the above relation, the mathematical formulation of the power flow problem results
in a system of algebraic nonlinear equations which must be solved by iterative techniques.

6.3.2. Gauss-Seidel Power Flow Solution

In the power flow study, it is necessary to solve the set of nonlinear equations represented
by (6.31) for two unknown variables at each node. In the Gauss-Seidel method, (6.31) is solved
for Vi, and the iterative sequence becomes

Pi sch  jQisch
*(k)
  yijV j(k 1)
Vi
Vi (k 1)  j i (6.32)
 yij
where yij shown in lowercase letters is the actual admittance in per unit. Pi sch and Qisch are the
scheduled active and reactive powers expressed in per unit.

Note:

- For buses where active and reactive powers are injected into the bus, such as generator
buses, Pi sch and Qisch have positive values.

- For load buses where active and reactive powers are flowing away from the bus, Pi sch

and Qisch have negative values.

If (6.31) is solved for Pi and Qi, we have

 *( k )  ( k ) n n
 
Pi ( k 1)
  Vi Vi  yij   yijV j( k )   j i (6.33)
  j 0 j 1  

 ( k )  n n
 
Qi( k 1)   Vi * Vi ( k )  yij   yijV j( k )   j i (6.34)
  j 0 j 1  
90
 The power flow equation is usually expressed in terms of the elements of the bus
admittance matrix. Since the off-diagonal elements of the bus admittance matrix Ybus, shown by
uppercase letters, are Yij = -yij, and the diagonal elements are Yii = yij, (6.32) becomes

Pi sch  jQisch
*(k)
 YijV j(k 1)
Vi
Vi (k 1)  j i (6.35)
Yii

and

 ( k )  n
 
Pi ( k 1)   Vi * Vi ( k )Yii  YijV j( k )   j i (6.36)
  j 1  

 *( k )  ( k ) n
 
Q i
( k 1)
  Vi Vi Yii  YijV j( k )   j i (6.37)
  j 1  

Note:

- For P-Q buses, the active and reactive powers Pi sch and Qisch are known. Starting with
initial estimate, (6.35) is solved for the real and imaginary components of voltage.
- For voltage controlled buses where Pi sch and Vi are specified, first (6.37) is solved for

Qi( k 1) , and then is used in (6.35) to solve for Vi ( k 1) . However, since Vi is specified,

only the imaginary part of Vi ( k 1) is retained, and its real part is selected in order to
satisfy

e    f
(k 1) 2
i i 
(k 1) 2
V
2
(6.38)

Or

ei(k 1)  V   fi (k 1) 

2 2
(6.39)

- The rate of convergence is increased by applying an acceleration factor (range from 1.3
to 1.7) to the approximate solution obtained from each iteration.

Vi (k 1)  Vi (k)   (Vi (k)cal  Vi(k) ) (6.40)

- The process is continued until changes in the real and imaginary components of bus
voltages between successive iterations are within a specified accuracy.

ei( k 1)  ei( k )  

(6.41)
fi ( k 1)  fi ( k )  
91
6.3.3. Line Flows and Losses
After the iterative solution of bus voltages, the next step is the computation of line flows
and line losses. Consider the line connecting the two buses i and j in Figure 6.4.

Fig. 6.3: Transmission line model for calculating line flows.

The line current Iij, measured at bus i and defined positive in the direction i j is given by

Iij  Il  Ii 0  yij (Vi  Vj )  yi 0Vi (6.42)

Similarly, the line current Iji measured at bus j and defined positive in the direction ji is
given by

I ji   Il  Ii 0  yij (Vj  Vi )  yi 0V j (6.43)

The complex power Sij from bus i to j and Sji from bus j to i are

Sij  Vi Iij*
(6.44)
S ji  V j I *ji

The power loss in line i – j is the algebraic sum of the power flows determined as follows:

SL,ij  Sij  S ji (6.45)

Example 6.5:

Figure 6.4 shows the one-line diagram of a simple three-bus power system with generation
at bus 1. The magnitude of voltage at bus 1 is adjusted to 1.05 p.u. The scheduled loads at buses
2 and 3 are as marked on the diagram. Line impedances are marked in p.u on a 100-MVA base
and the line charging susceptances are neglected.

92
 Fig. 6.4: One-line diagram of Example 6.5 (admittances in p.u on 100-MVA base).

(a) Using the Gauss-Seidel method, determine the phasor values of the voltage at the load
buses 2 and 3 accurate to four decimal places.
(b) Find the slack bus active and reactive power.
(c) Determine the line flows and line losses. Construct a power flow diagram showing the
direction of line flow.

Example 6.6:

Figure 6.5 shows the one-line diagram of a simple three-bus power system with generator
at bus 1 and 3. The magnitude of voltage at bus 1 is adjusted to 1.05 p.u. Voltage magnitude at
bus 3 is fixed at 1.04 p.u with an active power generation of 200 MW. A load consisting of 400
MW and 250 MVar is taken from bus 2. Line impedances are marked in p.u on a 100-MVA
base, and the line charging susceptances are neglected. Obtain the power flow solution by the
Gauss-Seidel method including line flows and line losses.

Fig. 6.4: One-line diagram of Example 6.6 (admittances in p.u on 100-MVA base).

6.3.4. Newton-Raphson Power Flow Solution

Newton-Raphson method has some advantages over the Gauss-Seidel method as follows:

- It is less prone to divergence with ill-conditioned problems.

- The number of iterations required to obtain a solution is independent of the system
size.
93
 - For large power system, the Newton-Raphson method is found to be more efficient and
practical.

For the typical bus of the power system shown in Figure 6.5, the current entering bus i is
given by (6.28). This equation can be rewritten in terms of the bus admittance matrix as
n
Ii  YijV j (6.46)
j 1

In the above equation, j includes bus i. Expressing this equation in polar form, we have
n
Ii   Yij V j ij   j (6.47)
j 1

The complex power at bus i is

Pi  jQi  Vi * Ii (6.48)

Substituting from (6.47) for Ii in (6.50)

n
Pi  jQi  Vi   i   Yij V j ij   j (6.49)
j 1

Separating the real and imaginary parts,

n
Pi   Yij V j Vi cos(ij   i   j ) (6.50)
j 1

n
Qi   Yij V j Vi sin(ij   i   j ) (6.51)
j 1

Equations (6.50) and (6.51) constitute a set of nonlinear algebraic equations in terms of the
independent variables, voltage magnitude in p.u and phase angle in radians. We have two
equations for each load bus, given by (6.50) and (6.51), and one equation for each voltage-
controlled bus, given by (6.51).

The scalar Taylor’s series indicates that

f ( x )
f ( x  x )  f ( x )  x  h.o.t (6.52)
x

where h.o.t stands for “higher-order terms”. In the case of n scalar equations, we have

94
 f1 ( x ) f ( x )
f1 ( x  x )  f1 ( x )  x1  ...  1  h.o.t
x1 xn
f 2 ( x ) f ( x )
f 2 ( x  x )  f 2 ( x )  x1  ...  2  h.o.t
x1 xn (6.53)

f n ( x ) f ( x )
f n ( x  x )  f n ( x )  x1  ...  n  h.o.t
x1 xn

Expanding (6.50) and (6.51) in Taylor’s series about the initial estimate and neglecting all
higher order terms results in the following set of linear equations.

 P   J1 J 2    
 Q    J J 4    V 
(6.54)
   3

Where

 P2(k)   Q2(k)    2(k)    V2(k) 

       
P   ; Q   ;    ; V   
 Pn 
(k)
 Qn 
(k)
  n 
(k)  (k) 
  Vn 

 P2(k) P2(k)   P2(k) P2(k) 

  V  Vn 
 n   2 
 2 
J1   ; J2   ;
 (k)   (k) 
 Pn Pn(k)   Pn Pn(k) 
  2  n    V2  Vn 


 Q2(k) Q2(k)   Q2(k) Q2(k) 

  V  Vn 
 n   2 
 2 
J3   ; J4   
 (k) (k) 
 (k) 
 Qn Qn   Qn Qn(k) 
  2  n    V2  Vn 


In these above equation, bus 1 is assumed to be the slack bus. J 1 to J4 is the four parts of
the Jacobian matrix in which the elements are the partial derivatives of (6.50) and (6.51),
evaluated at  i(k) and  Vi (k) .

Note

- If m buses of the system are voltage-controlled, m equations involving Q and V and

the corresponding columns of the Jacobian matrix are eliminated. Accordingly, there

95
 are (n-1) active power constraints and (n-1-m) reactive power constraints, and the
Jacobian matrix is of order (2n  2  m)  (2n  2  m) .
- The order of sub-matrix of Jacobian matrix are

J1: (n  1)  (n  1)

J2: (n  1)  (n  1  m)

J3: (n  1  m)  (n  1)

J4: (n  1  m)  (n  1  m)

The diagonal and the off-diagonal elements of J1 are

Pi
  Yij V j Vi sin(ij   i   j ) (6.55)
 i j i

Pi
  Yij Vi V j sin(ij   i   j ) j i (6.56)
 j

The diagonal and the off-diagonal elements of J2 are

Pi
 2 Vi Yii cosii   V j Yii cos(ij   i   j ) (6.57)
 Vi j i

Pi
 Yij Vi cos(ij   i   j ) j i (6.58)
 Vj

The diagonal and the off-diagonal elements of J3 are

Qi
  Yij V j Vi cos(ij   i   j ) (6.59)
 i j i

Qi
  Yij Vi V j cos(ij   i   j ) j i (6.60)
 j

The diagonal and the off-diagonal elements of J4 are

Qi
 2 Vi Yii sin ii   V j Yii sin(ij   i   j ) (6.61)
 Vi j i

Qi
  Yij Vi sin(ij   i   j ) j i (6.62)
 Vj

The term Pi (k) and Qi(k) are the difference between the scheduled and calculated values,
known as the power residuals, given by
96
 Pi (k)  Pi sch  Pi (k) (6.63)

Qi(k)  Qisch  Qi(k) (6.64)

The new estimates for bus voltage are

 i(k 1)   i( k )   i(k) (6.65)

Vi (k 1)  Vi ( k )   Vi (k) (6.66)

The procedure for power flow solution by the Newton-Raphson method is as follows:

1. For load buses, where Pi sch and Qisch are specified, voltage magnitudes and phase angles

are set equal to the slack bus values, i.e., Vi (0)  1.0 and  i(0)  0.0 . For voltage-regulated

buses, where Vi (0) and Pi sch are specified, phase angles are set equal to the slack bus

angle, i.e.,  i(0)  0.0 .

2. For load buses, Pi (k) and Qi(k) are calculated from (6.50) and (6.51) and Pi (k) and Qi(k)
are calculated from (6.63) and (6.64).
3. For voltage-controlled buses, Pi (k) and Pi (k) are calculated from (6.50) and (6.64),
respectively.
4. The elements of the Jacobian matrix (J1, J2, J3 and J4) are calculated from (6.55) to (6.62).
5. The linear simultaneous equation (6.54) is solved directly by optimally ordered triangular
factorization and Gaussian elimination.
6. The new voltage magnitudes and phase angles are computed from (6.65) and (6.66).
7. The process is continued until the residuals Pi (k) and Qi(k) are less than the specified
accuracy
Pi ( k )  
(6.67)
Qi( k )  

Example 6.7

Figure 6.5 shows the one-line diagram of a simple three-bus power system with generation
at bus 1. The magnitude of voltage at bus 1 is adjusted to 1.05 p.u. The scheduled loads at buses
2 and 3 are as marked on the diagram. Line impedances are marked in p.u on a 100-MVA base
and the line charging susceptances are neglected.

97
 Fig. 6.5: One-line diagram of Example 6.5 (admittances in p.u on 100-MVA base).

(a) Using the Newton-Raphson method, determine the phasor values of the voltage at the
load buses 2 and 3 accurate to four decimal places.
(b) Find the slack bus active and reactive power.
(c) Determine the line flows and line losses. Construct a power flow diagram showing the
direction of line flow.

6.3.5. Fast Decoupled Power Flow Solution

For power system transmission lines having a very high X/R ratio, the following
assumptions are used:

- The active power changes P are less sensitive to changes in the voltage magnitude
and are most sensitive to changes in phase angle .
- The reactive power is less sensitive to changes in angle and are mainly dependent on
changes voltage magnitude.

Thus, it is reasonable to set elements J2 and J3 of the Jacobian matrix to zero. The Equation
(6.54) becomes

 P   J1 0    
 Q    0 J    V  (6.68)
   4 

Or

P
P  J1     

(6.69)
Q
Q  J 4   V   V
V

Note

- This method has advantage of requiring considerably less time to solve compared to
the time required for the solution of (6.54)
98
 - Considerable simplification can be made to eliminate the need for recomputing J1 and
J4 during each iteration.

The diagonal elements of J1 described by (6.55) may be written as

Pi n
  Yij V j Vi sin(ij   i   j )  Vi Yii sin ii
2

 i j 1

Replacing the first term of the above equation with -Qi, as given by (6.53), results in

Pi
 Qi  Vi Yij sin ii  Qi  Vi Bii
2 2

 i

where Bii  Yii sin ii is the imaginary part of the diagonal elements of the bus admittance matrix.

In typical power system, the self-susceptance Bii >> Qi, and we may neglect Qi. Further

simplification is obtained by assuming Vi  Vi , which yields

2

Pi
  Vi Bii (6.70)
 i

Under normal operating conditions,  j   i is quite small. Thus, assuming that

ii   j   i  ii , the off-diagonal elements of J1 becomes

Pi
  Vi V j Bij (6.71)
 j

Further simplification is obtained by assuming V j  1

Pi
  Vi Bij (6.72)
 j

Similarly, the diagonal elements of J4 described by (6.61) may be written as

Qi n
  Vi Yii sin ii   Yii Vi V j sin(ij   i   j ) (6.73)
 Vi j 1

Replacing the second term of the above equation with –Qi, results in

Qi
  Vi Yii sin ii  Qi (6.74)
 Vi

Again, since Bii >> Qi, Qi may be neglected and we have

99
 Qi
  Vi Bii (6.75)
 Vi

Likewise in (6.62), assuming ii   j   i  ii , yields

Qi
  Vi Bij (6.76)
 Vj

With these assumptions, equation (6.69) take the following form

P   B '  
(6.77)
Q   B ''   V

where B’ and B’’ are the imaginary part of the bus admittance matrix Ybus. Since the elements of
this matrix are constant, they need to be triangularized and inverted only once at the beginning of
the iteration. The successive voltage magnitude and phase angle changes are

1 P
    B ' 
V
(6.78)
Q
'' 1
 V    B 
V

Note

- B’ is of order of (n-1), B’’ is of order of (n-1-m), where m is the number of voltage-

regulated buses.
- This method requires more iterations than the Newton-Raphson method, but requires
considerable less time per iteration, and a power flow solution is obtained very rapidly.
- This technique is very useful in contingency analysis where numerous outages are to be
simulated or a power flow solution is required for on-line control.

Example 6.8

Figure 6.6 shows the one-line diagram of a simple three-bus power system with generation
at bus 1. The magnitude of voltage at bus 1 is adjusted to 1.05 p.u. The scheduled loads at buses
2 and 3 are as marked on the diagram. Line impedances are marked in p.u on a 100-MVA base
and the line charging susceptances are neglected.

100
 Fig. 6.6: One-line diagram of Example 6.5 (admittances in p.u on 100-MVA base).

(a) Using the fast decoupled method, determine the phasor values of the voltage at the load
buses 2 and 3 accurate to four decimal places.
(b) Find the slack bus active and reactive power.
(c) Determine the line flows and line losses. Construct a power flow diagram showing the
direction of line flow.

6.3.6. Assignment

101