Chapter 1.

Introduction to Electrical Energy System (Gii Thiu Chung V H Thng in Nng)

The electric power system is a network of interconnected components which generate electricity by converting different forms of energy, (potential energy, kinetic energy, or chemical energy are the most common forms of energy converted) to electrical energy; and transmit the electrical energy to load centers to be used by the consumer. The production and transmission of electricity is relatively efficient and inexpensive, although unlike other forms of energy, electricity is not easily stored and thus must generally be used as it is being produced. The electric power system consists of three main subsystems: the generation subsystem, the transmission subsystem, and the distribution subsystem. Electricity is generated at the generating station by converting a primary source of energy to electrical energy. The voltage output of the generators is then stepped-up to appropriate transmission levels using a step-up transformer. The transmission subsystem then transmits the power close to the load centers. The power is then stepped-down to appropriate levels. The distribution subsystem then transmits the power close to the customer where it is steppeddown to appropriate levels for use by a residential, industrial, or commercial customer. In this chapter, a brief description of the common methods of converting energy to electric power, and each power subsystem will be discussed.

Electrical energy is produced through an energy conversion process.

1.1

Sources of Energy

Electricity is produced by converting energy from one form to electricity.

The process used may be a direct conversion process, where the energy source is converted directly to electricity. An example of this is solar photovoltaic cells, which converts the energy found in solar radiation directly to electricity. An indirect conversion process consists of converting energy from one form, to an intermediate form, to electricity. Coal-fired generating plants are an example of this process, as the chemical energy released as heat by burning the coal is changed to rotating kinetic energy by the steam turbine, and then the rotating kinetic energy is converted to electricity. The majority of the electricity produced today is produced through an indirect energy conversion process. Major sources of energy for the production of electricity are fossil fuels, hydro energy, solar radiation, and nuclear energy. Fossil fuels are coal, petroleum, and natural gas. Fossil fuels are a finite, non-renewable resource, and are the primary source for the production of electricity. These fuels are burned to release their chemical energy, which produces heat to power steam turbines. The steam turbines power rotating electric generators, which turn kinetic energy into electricity. No energy conversion process converts all the energy present in one form

completely into the new form. Since the production of electricity from fossil fuels involves several energy conversion steps, the overall efficiency of a fossil fuel power plant is quite low, somewhere in the range of 40%. The environmental effects of electricity generation from fossil fuels are also a concern. Fly-ash is the physical matter left after coal is burned. Since fly-ash is harmful to humans, it must be disposed in an environmentally safe matter. Combustion of fossil fuels also produces carbon monoxide, carbon dioxide, sulfur dioxide, and nitrous oxides gases. These are the "greenhouse gases" that contribute to acid rain and global warming. Nuclear energy, just like fossil fuels, is a finite, non-renewable, energy source that uses an indirect conversion process to produce electricity. In the nuclear power plants, the fission reaction is used to produce electricity. The fission reaction involves the splitting of the nuclei of a heavy element. The heat output from this reaction is used to power a steam turbine, which is used to drive a rotating electric generator, just as in a fossil fuel power plant. The advantage of nuclear fission compared to fossil fuels is the energy content of fissionable materials. The energy content of uranium is approximately 1010 Btu/kg, which is about one million times the energy content of fossil fuels. The disadvantage of nuclear fission is the environmental cost. After fission, the nuclear fuel, the reactor vessel where fission occurs, and the steam pipes are highly radioactive. Also, plant failures can lead to the release of radioactive steam into the atmosphere. Solar radiation includes energy used directly as intercepted solar radiation, or indirectly as wind and hydropower. Solar radiation is a renewable energy source. The average incident power at the earth's surface is 182 W/m2, which corresponds to a daily average energy of 4.4 kWh/m2. Direct use of solar power includes active types involving photovoltaic cells, and passive types using radiation to heat solar collectors. Photovoltaic cells directly convert sunlight into electricity. Solar collectors are normally incorporated into a solar thermal system, converting sunlight into heat for various forms of use, including space heating, water heating, industrial process steam, and electricity production. At present, large-scale utilization of solar energy is limited by several factors, including the cost of solar cells and solar collector-heat exchanger systems, and by the requirement of an adequate energy storage system to smooth out the daily variation. Since solar cells generate direct current, inverting equipment is needed to obtain the desired alternating current for most large-scale operations. Since sunlight is available everywhere, use of solar radiation for energy production is not site specific. The best photovoltaic cells have efficiency in the 14% to 17% range. Photovoltaics are attractive from an environmental aspect, since there is no gases or wastes produced from the energy conversion process, and the fuel source does not have to be extracted from the ground. However, the most efficient solar cells use gallium arsenide, which is a toxic material. Solar cells are too new for an understanding of the disposal requirements and costs involved for worn out cells. Wind energy is a form of indirect use of solar radiation. Solar radiation produces wind by heating the air. During the day, the air over land is heated much faster than air over water bodies because the land absorbs much less sunlight, and the evaporation is less. The heated air over land expands, becomes lighter, and rises. The cooler, heavier air over

large water bodies moves in to replace the lighter and warmer air, creating a horizontal motion of air. During the night, since the land cools faster than water, the cool air moves seaward to replace the warm air that rises from the surface of the water. Wind energy to electricity is an indirect energy conversion process, because turbine type wind generators transform the kinetic energy of the wind into rotary-shaft motion and, in turn, to electrical energy. Since most wind turbine generators require a sustained wind speed of 20 km/hour, the location of the wind turbine generator is important, as a consistent adequate wind velocity must be present. This means that bulk wind generators are limited to specific sites, generally in coastal areas or mountain passes. Wind turbines have a maximum possible efficiency of 59.3%, with a more common efficiency of around 40%. The significant environmental problems associated with wind turbines are noise, aesthetics, and interaction with birds. Hydropower is also an indirect means of using solar power to produce electricity, since hydropower uses the stream-flow part of the hydrological cycle. In a hydropower plant, the potential energy of a mass of water in a reservoir a distance above the stream bed is converted to kinetic energy by flowing through a hydraulic turbine. The resulting kinetic energy of the turbine drives an electric generator. Hydropower is available where ever a suitable site exists having enough stream flow, potential drop, and area. Industrialized nations contain about 30% of all hydropower potential, and are responsible for about 80% of all electricity produced from hydro. Asia accounts for 30% of hydropower potential, and produces only 7% of electricity such produced. Africa accounts for 20% of hydropower potential, yet produces only 2% of electricity such produced. Hydropower is very attractive because it is a non-polluting renewable resource, but it can be very disruptive environmentally. The dam and reservoir effect the normal ecology of the stream and the surrounding habitat by altering water use, changing natural water flow cycles of the stream, and taking up land area for the reservoir. New dams normally requires the relocation of people and buildings. Tidal energy uses the tidal flow of oceans to run a hydropower plant to produce electricity. Basically, a dam encloses a tidal pool. The tidal pool fills during periods of high tide, then empties during periods of low tide. The water flow into and out of the pool drives a reversible hydraulic turbine. Since the turbine is reversible, the flow of water into and out of the bay may be used to produce electricity. Therefore, tidal power is available twice during each 12h 25min tidal period. The ideal sites for tidal energy have a large difference in tides.

1.2

Electromechanical Energy Conversion

The most common method for bulk power generation is by rotary generators located in
electric power stations. These generators are electromechanical energy converters, also known as electric machines. In practice, a mechanical prime mover coupled to the generator rotates the electrical conductors constituting the generator windings in a magnetic field, inducing a voltage in the generator windings. These windings supply electrical load on the generator. Conversely, if a current carrying conductor is placed in a magnetic field, the conductor experiences a force according to Ampere's law. In general,

electric machines are reversible, and capable of operating both as generators and motors. There are three major types of rotating electric machines: dc commutator, induction, and synchronous machines.
Energy
(fossil, hydro, wind, nuclear, etc.)

Prime Mover

Mechanical Energy

Electric Generator

Electric Energy

Figure 1-1 Prime Mover Driving an Electric Generator The source of mechanical energy for a rotating electric generator is known as the prime mover. The prime mover is directly coupled to the generator. Energy sources for prime movers are thermal, hydro, and wind. The prime movers normally are turbines, but some thermal units use internal-combustion engines. A turbine is mechanical device that is forced to rotate by the pressure of a gas (such as steam for thermal units or air for wind units) or fluid (such as water for hydro units). An electromechanical energy converter converts mechanical energy into electrical energy, and vice versa. A generator converts energy from mechanical to electrical form, and modulates in response to an electric signal. A motor converts energy from electrical to mechanical form, and modulates in response to an electrical signal derived from mechanical speed. Rotating machines, if lossless, operate on the principle of electromechanical power equivalence as given by Pmech=Mwm = vi=Pelectrical Where: M is mechanical torque (N-m), wm is mechanical angular velocity (rad/s), v is instantaneous electrical volts (volts), and i is instantaneous electrical current (amperes). Electric generators are governed by Faraday's law of electromagnetic induction: an electromotive force (emf) is induced in a conductor "cutting" magnetic lines of flux. Specifically if a conductor of length l (m) moves with a velocity u in a uniform magnetic field B (tesla), such that l, u, and B are mutually perpendicular, then the induced emf in the conductor is given by E = Blu. Example Problem 1.1. An ideal energy converter develops 500 N-m of torque while running at 3000 rpm. If the input voltage is 1000 volts, determine the input current for this generator. M m = vi

3000 x 2 500 M m 60 = = 157.08 amps i= v 1000 1.2. Calculate the power output of the ideal energy converter.

M m = Power 3000 x 2 Power = 500 = 157.08 kW 60

1.3

Power System Load

An electric load (or demand) is the power requirement of any device or equipment that
converts electric energy into light, heat, or mechanical energy. The power system load is the total of all such loads connected to the system. As such, the load is never constant, varying daily, weekly, monthly, and yearly as loads are added or subtracted from the power system. The minimum system load for a given period is called the base load. The maximum system load for a given period is known as the peak load or peak demand. The peak demand is usually quite short in duration. The operation of generation plants must be closely coordinated with the load demands to ensure that enough generation capacity is on line. On weekdays, the base load generally begins increasing at about 5:00 a.m., and hits peak load around 7:00 p.m. Maximum yearly peak loads generally occur during the summer. As more people and businesses connect to an electric system, the amount of load on the system will increase. Load forecasting is performed to ensure that power system generating capacity will be adequate to meet these future load demands. Power stations take years to build, so it is necessary to plan well in advance. One important part of load forecasting is the idea of the load growth rate. This is the estimated rate at which load on the power system will increase. It is generally based on historical data. The growth rate of the system load L is mathematically represented by dL = aL dt where a is the constant of proportionality, also known as the per-unit growth rate. The solution to this equation is written as L = L0 e at where L0 is the value of L at t = 0. At any two values of time, t1 and t2, the ratio of the corresponding L1 and L2 is L2 = e a (t2 t1 ) L1 This equation may be used to determine the time tk such that L2 = kL1 and t2-t1 = tk, given by ln k tk = a When talking about the growth rate of a quantity, the term "doubling time" is often used. The doubling time is the period necessary to double the initial value of load L, given a constant value of a. ln 2 0.693 = t2 = a a Doubling time is used to describe how long it will take, at a constant growth rate, to use twice what is currently used. For example, assuming a present peak energy demand of

10000 MW, and a 10 percent growth rate in the peak demand, the doubling time for energy demand is 6.93 years. This means that peak energy demand in about 7 years will equal 20 GW, or twice the current peak demand. Obviously, setting a steady growth rate for the use of any quantity is unrealistic, as the growth rate depends on many factors.

1.4

Environmental Impact of Electricity Generation and Transmission

All energy conversion methods used to produce electricity have some environmental
impact. The impact may have an active effect like the emission of airborne pollutants, or may have a passive effect like aesthetics or habitat modification. Even methods considered environmentally friendly, like wind, solar, and hydro, have some impact on the environment. Not only does the production of electricity have an environmental impact, but the transmission of electricity, with concerns over electromagnetic fields, aesthetics, and land use, has an impact as well. The whole cycle of electricity generation must be considered when looking at the environmental impact. This includes the production and transportation of fuel for the conversion process. This is especially true of fossil fuel and nuclear power plants, which use large quantities of fuel taken from the earth. Fossil fuel power plants generally have the most widespread effect on the environment, as the combustion process produces airborne pollutants that spread over a wide area. Nuclear power plants have the most potentially dangerous effect. An operating accident at a nuclear station could allow a large release of radioactive particles to occur. Solar, hydro, and wind power plants generally have a small effect on the environment. Fossil Fuel Power Plants Fossil fuel power plants produce environmental problems including land and water use, air emissions, thermal releases, climatic and visual impacts from cooling towers, solid waste disposal, ash disposal (for coal), and noise. Due to the need for large amounts of steam, plants can have a great effect on water use. For example, a typical 500 MW coal fired power plant uses 25 x 109 l/GW-year of water, which must be taken from a water source, and then cooled to return to the water source with as little environmental effect as possible. The biggest effect fossil fuel plants have overall is the emission of air pollutants, particularly SOX, NOX, CO, CO2, and hydrocarbons. CO, CO2 and hydrocarbons are the greenhouse gases, possibly responsible for global warming. SOX and NOX can produce acid when released into the atmosphere, leading to the production of acid rain. Table 1.1 list approximate amounts of airborne pollutants produced. Generally, air emissions are controlled by the use of scrubbers and precipitators located at the plant. Plant Type Coal CO 0.11 NOX 3.54 SO2 9.26 CO2 1090

Oil Gas

0.19 0.20

2.02 2.32

5.08 0.004

781 490

Table 1.1 Power Plant Emissions (g/kWh)

Hydro The use of hydropower to produce electricity can have both positive and negative effects on the environment. At some sites, a dam may help with flood control, flow regulation, or the reservoir may provide recreational opportunities. At other sites, the dam may have adverse effects on the hydrological cycle, water quality of the stream, stream ecology, fish migration, and cause the destruction of landscapes and ecosystems. Building new high-head dams requires the displacement and compensation of populations. Low-head dams generally have a benign effect on the environment. Dam failures can lead to catastrophic floods. Nuclear Power Plants Nuclear power plants have one environmental issue no other form of electrical power plant does. An accident at a nuclear power plant may release large amounts of radioactive particles, possibly resulting in a direct loss of life, and rendering a large land area immediately around the plant unlivable. The largest regular environmental impact is the disposal of the high level nuclear waste contained in spent fuel rods, as this waste must be stored safely for thousands of years. A long term issue is the decommissioning of nuclear power plants. Decommissioning is shutting down a nuclear plant after its operational life is over. At this point the entire reactor vessel becomes a high level radioactive waste that must be disposed. The current methods of decommissioning a plant are to completely remove and dispose of all radioactive components, to entomb the reactor in concrete, or simply to shut the plant down and restrict access until the radioactivity dies out. Transmission of bulk electricity from the generating station to the load uses wires suspended on large towers, known as transmission lines. Traditionally these lines have been viewed only as an aesthetic nuisance that could cause communications interference and be a hazard to low flying aircraft. Today, there are other issues considered about the effect of transmission lines on the environment. Greater concern is placed on the effect of the lines on the natural habitat. The major new issue is the effect of electromagnetic fields (EMFs) on human health.

1.5

The Generation Subsystem

The electric power system consists of three main subsystems: the generation subsystem,
the transmission subsystem, and the distribution subsystem. Electricity is generated at the generating station by converting a primary source of energy to electrical energy.

Alternating current synchronous generators are the main source of electrical energy. These generators convert energy from a primary form to the electrical form. Energy is input to the generator in the form of mechanical torque from a prime mover or turbine. The turbine in turn is powered by a moving fluid either steam or water. The current sources of nearly all the electrical energy distributed come from: the conversion of chemical energy of fossil fuels, kinetic energy of water, and the nuclear fission energy. Some renewable sources of energy like wind, solar, and geothermal are also used. The synchronous generators typically range in size from 50 kW to over 1300 MW, and operate at voltage levels from 480 V to 25 kV. The output delivered by the generators is a balanced set of three phase ac voltages.
Extra-High-Voltage Transmission Substation 500 kV Turbine T Step-up Transformer

Generator 220 kV Transmission Lines Transmission Substation

110 kV

0.4 kV Distribution Transformer

Distribution Subsystem (35, 22,10, 6 kV)

Sub-Transmission Substation

Figure 1.2 Example of a Power System in Vietnam

In fossil fuel plants, the source of heat energy is due to the combustion of the fossil fuel. When coal is used, crushed coal is conveyed to a pulverizer where the fuel is ground to a consistency of face powder. The pulverized coal is then fed to burners where it undergoes combustion. The thermal energy produced is then used to heat water in a boiler to produce steam, and stored in the form of internal energy of the steam. The steam flows into the turbine where this internal energy is released as mechanical energy.

Alternatively, natural gas or petroleum can also be combusted to produce the source of heat energy. The turbine is a device used to convert the stored energy of high-pressure and hightemperature steam into rotational energy. The steam fed to the turbine is passed through a series of stages, each of which consist of stationary blades and moving blades attached to a rotor. The stationary blades accelerate the steam to high velocity, and this kinetic energy is converted into shaft rotation by the moving blades. The turbine then provides the rotational energy to the synchronous generator. Combustion turbines are an important part of the utility system generation mix. The combustion turbine has characteristics that make it attractive both for generation to meet peak loads and for base-load operation. Combined cycle plants using combustion turbines, heat recovery steam generators, and a steam turbine have the advantage of superior heat rates. In a simple cycle combustion turbine, air at atmospheric pressure and temperature is compressed to raise it temperature and pressure. The pressurized air is partly combusted with fuel and passed to the turbine. The fuel used could be gaseous, or liquid fuels, or combinations of different fuels. The gases exhausting from the combustion turbine are quite hot. In a combined cycle plant, these gases are used to generate steam for a steam turbine. This system incorporates some of the features of a conventional fossil-fired unit. With the advent of energy consciousness, and the drive to improve energy efficiency, cogeneration has become an important source of energy production. Electricity is produced sequentially with thermal energy. Two different schemes are normally used in co-generation. In the topping system electricity is first produced ("on top"). Waste heat is used in some industrial process or to provide heating. In the bottoming system, the heat energy is first used in the process. Waste heat from the process is used to produce electricity ("on bottom"). In a hydroelectric plant, the electrical energy is derived from falling water. Water is typically stored by a dam. It is then delivered using pen-stocks or pipes to the hydraulic turbine, which in turn supplies mechanical energy to the generator. Hydraulic turbines are of two basic types, impulse and reaction. The impulse turbine is also known as the Pelton wheel. It is used in high-head plants, where the head (fall of water) is 305 meters or more. The impulse turbine usually has a horizontal shaft. In the Pelton wheel, kinetic energy is derived by converting the fall of water into one or more high-velocity jets located around the periphery of the wheel and directed into spoon-shaped buckets. Reaction type turbines are of two general types, Francis and propeller. In both types the water passages are completely filled with water. The energy which drives the wheel is in both kinetic and "pressure head" form.

The output of the generators is fed to a unit transformer which steps-up the voltage to the appropriate transmission level.

1.6

The Transmission Subsystem

The electrical power produced at generating stations is transported or moved to areas

close to load locations by the transmission subsystem. Transmission lines are a major component of this subsystem. Transmission lines could be underground or overhead. The latter kind are more common. Transmission line voltages range from 110 kV to 765 kV. Overhead transmission lines span long distances and use bare conductors. As a result, they must be connected by insulated mechanical connections to towers or poles which support them. A transmission line consists of conductors, insulators, mechanical supports, and usually shield wires. Aluminum conductor, steel reinforced (ACSR), conductors are most commonly used for high-voltage transmission lines. Compared to other conductor material, they have low cost and high strength-to-weight ratio. Suspension type insulators are used. The material most commonly used for insulators is porcelain. New polymer materials are now being used for making insulators. Shield wires are used to protect the energized conductors from lightning. Underground transmission is mainly used in urban areas. Underground lines cost on an average, about 8-15 times more than overhead lines. Underground cables also have the disadvantage of poor accessibility. Transformers are an important part of the transmission subsystem. The transformer is a static device which transfers electrical energy from one circuit magnetically coupled with another. In doing so, they can transform voltage levels in the magnetically coupled circuits. Different types of transformers are used in the transmission subsystem. Power transformers are used to step-up, or step-down voltages. On load tap changing transformers are used to regulate voltages. Auto-transformers are used for most ties between moderate-voltage and high voltage transmission, high-voltage and extra-high voltage circuits. Phase shifting transformers are used to regulate the flow of real power. Switchgear constitutes an essential component of the transmission subsystem. Under normal operating conditions, it provides the means to perform routine switching operations, e.g., disconnecting and isolating various equipment for maintenance, inspection, or replacement, transferring load, isolating regulators, etc. Under abnormal conditions, switchgear provides the means for automatically isolating parts of the system in trouble to prevent damage and to localize the problem. The main components of the switchgear include, circuit breakers, disconnecting switches, fuse, instrument transformers, buses and connections, supporting insulators, protective and control relays, and control switches.

A substation houses all equipment involved in the switching or regulating of electricity. Substations can be large or small. Control can be automatic or manual.

1.7

The Distribution Subsystem

sub-transmission designates the circuits, which deliver energy from the transmission subsystem to the distribution subsystem. Usually the transmission substations supply the sub-transmission system, but it is still referred to as the subtransmission. Many sub-transmission systems were previously transmission lines. Load growth and demand for more power resulted in the transmission voltage being too low. As a result, in Vietnam, voltages from110 kV down to 35, 22, 10, 6 kV are found in subtransmission systems. The distribution subsystem includes the primary circuits and the distribution substations that supply them, the distribution transformers, the secondary circuits, including the components all the way up to the entrance of the customer's premises, and the protective and control devices. The primary circuit is three-phase and is operated in the 6-35 kV range. The secondary circuits serve most of the customers at levels of 220 volts singlephase, or 380 volts three-phase. The various other components used in the distribution subsystem include: distribution transformers, automatic circuit reclosers, cut-outs (a combination fuse and knife switch used on power poles), surge arresters, separable connectors, capacitors and voltage regulators, and metering.

The

1.8

Power Industry Structure

he traditional structure of the electric power system was a single company that owned and operated all of the components of the system. This included the generation, transmission and distribution subsystems along with all communications and control systems. This structure is changing dramatically since the deregulation of the industry was implemented in some industrialised countries in the early 1990s. Since this time, each company has been reorganized to fit the expected structure of the industry within the very near future. This structure assumes that the companies are segmented into generation (GENCO), transmission (TRANSCO), distribution (DISTCOS) and independent system operator (ISO) companies or independent contract administration (ICA) business units. The GENCOs are responsible for providing (selling) electric generation according to contracts along with any auxiliary services as required for the customer or the transmission grid. The TRANSCOs are responsible for the maintenance of the transmission equipment to enable the transmission grid to transport electricity according to schedules with auxiliary services as required. The DISTCOs are responsible for providing (purchasing) electricity along with any auxiliary services to the ultimate customer(s). The ICA is responsible for matching the schedules (transmission grid support, schedule control and accounting, control and communication facilities and personnel) such that the transmission grid can transport the electricity and such that

auxiliary services are available as required. The ISO is responsible for coordination of the maintenance schedules subject to the contracts and operating restrictions of the transmission system.

Chapter 2. Introduction to Polyphase Networks Analysis (Phn Tch Mch in Nhiu Pha)
Today, the largescale generation, transmission, and distribution of electric power is by means of the 3 phase ac system; that is, three individual single-phase voltages and currents having a 120 phase relationship to each other and intermingled on three wires (excluding a neutral). The three-phase system has been adopted because it provides for a constant rather than pulsating power flow to motors, and because it is an efficient system as far as the amount of copper required per kilowatt transmitted. Networks with different number of phases are used for some specialized applications. This chapter starts with an elementary discussion of alternating voltages and currents and with associated energy flows. It ends with an analysis with balanced three-phase circuits.

Systems with more than one phase are generally called polyphase.

2.1

Sinusoidal Steady State Analysis

At the nodes of a power system, the voltage waveform can be assumed to be purely
sinusoidal and of constant frequency. In these notes, we deal with the phasor representation of sinusoidal voltages and currents, and use the boldface quantity V and I to indicate these phasors. | V | and | I | designate the magnitudes of the phasors.

2.1.1

Phasor Representation

The phasor is a complex number that contains the amplitude and phase angle information of a sinusoidal function. Using Euler's identity, which relates the exponential function to the trigonometric function, the phasor concept can be developed.
e
j

= cos j sin

(2.1)

Equation (2.1) provides us an alternative way of expressing the cosine and sine function. The cosine function can be represented as the real part of the exponential function, and the sine function can be represented as the imaginary part of the exponential function as follows

cos = Re{e }

(2.2) (2.3)

and

sin = Im{ e }

Where Re represents the real part of and Im represents the imaginary part of. A sinusoidal voltage function (we have chosen to use the cosine function in analyzing the sinusoidal steady state), can be written in the form suggested by Eq. (2.2)

v = Vm cos( t + ) = Vm Re{e j (t + ) } = Vm Re{e jt e j } We can move the coefficient V m inside the argument, and also reverse the order of the two exponential functions inside the argument without altering the result. (2.4)

v = Re{V me j e jt }
j t

(2.5)

In Eq. (2.5) the coefficient of the exponential e is a complex number that carries the amplitude and phase angle of the given sinusoidal function. This complex number is by definition the phasor representation or phasor transform, of the given sinusoidal function. Thus j V = V me = P {V m cos(t + ) } (2.6) Where the notation P {V m cos(t + ) } depicts "the phasor transform of Vm cos( t + ) ." Hence, the phasor transform P transforms the sinusoidal function from the time domain to the complex-number domain. Equation (2.6) is the polar representation of a phasor, we can also obtain the rectangular representation of the phasor as
V = V m cos + jVm sin

(2.7)

2.1.2 Power Calculation in Single-Phase AC Circuits

Our aim is to determine the average power that is either delivered to or absorbed from a pair of terminals by a sinusoidal voltage and current. Figure 2.1 depicts the problem. Here, v and i are steady-state sinusoidal signals. With the use of passive sign convention; the power at any instant is
p = vi

(2.8)

Ieff + Veff _

Circuit

Figure 2.1 The Basic Calculation to Determine Average Power

The power is measured in watts when the voltage is in volts and the current is in amperes. First we write expressions for v and i v = V m cos( t + v ) and i = Im cos(t + i ) (2.9) (2.10)

Here, v is the voltage phase angle and i is the current phase angle. Using the instant when current is passing through a positive maximum as the reference for zero time, and expressing v and i with respect to this reference we have v = V m cos( t + v i ) = V m cos( t + ) i = I m cos t (2.11) (2.12)

Where = v i

The angle in these equations is positive for current lagging the voltage and negative for leading current. A positive value of p shows that energy is absorbed at the terminals. Substituting Eqs. (2.11) and (2.12) into Eq. (2.8), the instantaneous power is given by p = V m I m cos(t + ) cos(t ) (2.13)

The average power associated with sinusoidal signals is given by the average of the instantaneous power over one period

p avg

1 = T

t 0 +T

t0

pdt

(2.14)

where T is the period of the sinusoidal function. Substituting Eq. (2.13) into Eq. (2.14), the average power can be determined. More information, however can be obtained by expanding Eq. (2.13) using the trigonometric identity

cos( A) cos( B ) =

1 [cos( A B) + cos( A + B)] 2

Letting A = t + , and B = t , we obtain from Eq. (2.13)

p=

V m Im V I cos + m m cos( 2 t + ) 2 2

(2.15)

Expanding the second term on the right-hand side of Eq. (2.15) using the trigonometric identity cos( A + B) = cos Acos B sin A sin B , we get p= V m Im V I cos + m m cos( )cos 2t 2 2 V m Im sin sin 2 t 2

(2.16)

The average value of p is given by the first term on the right hand side of Eq. (2.16) because the integral of either cos 2 t or sin 2t over one time period is zero. Hence, the average power is V I (2.17) P = m m cos 2 P is also referred to as the real or active power. The third term on the right hand side of Eq. (2.16), the term containing sin , is alternatively positive and negative and has an average value of zero. This component of instantaneous power p is called the instantaneous reactive power. The maximum value of this pulsating power is designated as Q, and is called reactive power. Hence,
Q= Vm Im sin 2

(2.18)

P and Q carry the same dimension. However, in order to distinguish between real and reactive power, we use the term VArs (volt-ampere reactive) for reactive power. Since we have used current as the reference, Q is positive for inductors ( = 90) and negative for capacitors ( = -90). The angle is referred to as the power factor angle. The cosine of this angle is called the power factor. Lagging power factor implies that current lags voltage. Leading power factor implies that the current leads the voltage. The average power given by Eq. (2.17) and the reactive power given by Eq. (2.18) can be written in terms of effective or rms values
Vm I m cos 2 V I = m m cos 2 2 = Veff I eff cos P= and, by similar manipulation,

(2.19)

Q = Veff Ieff sin

(2.20)

Complex power is the complex sum of average real power and reactive power, or

S = P + jQ

(2.21)

Complex power has the same dimensions as real or reactive power. However, in order to distinguish complex power from real and reactive power, we use the term volt amps. Thus we use volt amps for complex power, watts for average real power, and vars for reactive power. We can think of P, Q, and |S | as the sides of a right-angled triangle as shown in Figure 2.2.

|S| Q

Figure 2.2 The Power Triangle

Combining Eqs. (2.17), (2.18), and (2.21) we get

S=

V m Im V I cos + j m m sin 2 2 V I = m m [ cos + j sin ] 2 V I 1 = m m e j = V m Im 2 2

(2.22)

Using effective values of the sinusoidal voltage and current, Eq. (2.22) becomes
S = Veff Ieff

(2.23)

Equations (2.22) and (2.23) show that if the phasor current and voltage are known at a pair of terminals, the complex power associated with that pair of terminals is either one half the product of the voltage and conjugate of the current or the product of the rms phasor voltage and the conjugate of the rms phasor current.
S= 1 VI* = P + jQ 2

(2.24)

or

S = Veff I*eff = P + jQ

(2.25)

Equations (2.24) and (2.25) have several useful variations. In order to demonstrate these variations, we first replace the circuit in the box in Figure 2.1 by equivalent impedance Z as shown in Figure 2.3.
Ieff + Veff _

Figure 2.3 General Circuit Replaced with Equivalent Impedance

Expressing the voltage as the product of the current times the impedance, we obtain
Veff = ZI eff

(2.26)

Substituting Eq. (2.26) into Eq. (2.25) yields S = ZI eff I eff

= I eff Z = I eff ( R + jX ) = I eff R + j I eff X = P + jQ
2 2 2 2

(2.27)

from which,
P = I eff R =
2

1 2 ImR 2 1 2 Im X 2

(2.28)

and
Q = I eff X =
2

(2.29)

In Eq. (2.29), X is the reactance of either the equivalent inductance or the equivalent capacitance of the circuit; it is positive for inductive circuits and negative for capacitive circuits. Another useful variation of Eq. (2.25) is obtained by replacing the current with the voltage divided by the impedance:

 Veff S = Veff Z If the Z is a pure resistance element,

Veff = Z

= P + jQ

(2.30)

P=
and if Z is a pure reactive element

Veff R

(2.31)

Q=

Veff X

(2.32)

X is positive for an inductor and negative for a capacitor in Eq. (2.32).

2.2

Balanced Three-Phase Circuits

Electric

power is supplied by three-phase generators. It is then transformed appropriately using transformers and transmitted in the form of three-phase power except at the lowest voltage levels of the distribution system where single phase power is used.

There are two main reasons for using three-phase power. First, the instantaneous power supplied to motors is constant torque and therefore motors run much smoother. Second, three phase power requires less conductor-cost than single-phase power for the same delivered power. Figure 2.5 represents a three-phase (3 ) circuit. The circuit is said to be a balanced 3 circuit if the impedances are equal and the three voltage source phasors are equal in magnitude and are out of phase with each other by exactly 120. In discussing threephase circuits, it is standard practice to refer to the three phases as a, b, and c. Furthermore, the a-phase is almost always used as the reference phase. The three voltages that comprise the three-phase set are referred to as the a-phase voltage, the bphase voltage, and the c-phase voltage.
a A

Van Vcn c Nn Vbn Bb C N B

Figure 2.4 Three-Phase Balanced Circuit

Since the phase voltages are out of phase by 120, two possible phase relationships can exist between the a-phase voltage and the b- and c- phase voltages. One possibility is that the b-phase voltage lags the a-phase voltage by 120, in which case the c-phase voltage must lead the a-phase voltage by 120. This phase relationship is known as the abc, or positive phase sequence. The other possibility is for the b-phase voltage to lead the a-phase voltage by 120, in which case the c-phase voltage must lag the a-phase voltage by 120. This phase sequence is known as the acb, or negative phase sequence. In phasor notation, the two possible sets of balanced phase voltages are
Va Va

Vb Vc Positive Sequence Vb Negative Sequence

Vc

Figure 2.5 Phasor Diagram: Three Phase Voltages

Va = Vm 0o Vb = Vm 120o Vc = Vm + 120o

(2.33)

and
Va = Vm 0o Vb = Vm + 120o Vc = Vm 120o

(2.34)

The phase sequence of the voltages given by Eq. (2.33) is the abc, or positive sequence. The phase sequence of the voltages given by Eq. (2.34) is the acb, or negative sequence. Another important characteristic of a set of balanced three-phase voltages is that the sum of the voltages is zero. Thus, from either Eq. (2.33) or Eq. (2.34)
Va + Vb + Vc = 0

(2.35)

This relationship holds for any set of balanced three-phase variables. Components of balanced three-phase circuits can be connected either in a Y-connection

or a - connection. The phase quantities and the line quantities for these connections are related as follows.

2.2.1

Delta Connection

Figure 2.6 illustrates a - connected balanced three-phase load. The relationships developed however, can be applied to any component, e.g., generator, transformer, etc.
_ a A

+
Vab _

IaA Z IAB _ V VCA _ AB + _ + VBC + Z IBC ICA Z

Vca

b + c

+ IbB B Vbc _ IaA

Figure 2.6 Delta Connection

In the - circuit, the line-to-line voltage Vab is equal to the phase voltage V AB = V . To demonstrate the relationship between the phase currents and line currents, we assume a positive phase sequence and let I represent the magnitude of the phase current. Then selecting IAB arbitrarily as a reference phasor we have

I AB = I 0o I BC = I 120
and
o

(2.36) (2.37) (2.38)

I CA = I + 120o

We can express the line currents in terms of the phase currents by direct application of Kirchoff's current law:
I aA = I AB I CA = I 0o I 120o

= 3I 30o

(2.39)

I bB = I BC I AB = I 120o I 0o

= 3I 150o
I cC = I CA I BC = I 120o I 120o

(2.40)

(2.41) = 3I 90o Comparing Eqs. (2.39)-(2.41) with Eqs. (2.36)-(2.38) we see that the magnitude of the line currents is 3 times the magnitude of the phase currents and that the set of line currents lags (leads) the set of phase currents by 30 for positive (negative) sequences.

2.2.2

Wye Connection

Figure 2.7 illustrates a Y- connected balanced three-phase load. The relationships developed however, can be applied to any component, e.g., generator, transformer, etc.
a + A _ IaA ZY IAN Vca Vab + IcC IbB VCN + C ZY ICN _ + VAN _ N _ VBN IBN ZY + B

c _ b

_ Vbc +

Figure 2.7 Wye Connection

In the Y-circuit, the line current IaA is equal to the phase current I AN = I . To demonstrate the relationship between the line-to-line voltages and the line-to-neutral voltages, we assume a positive, or abc, sequence and let V be the magnitude of the line to neutral or phase voltage. We arbitrarily choose the line-to-neutral voltage of a-phase as the reference. We then have VAN = V 0o (2.42)

VBN = V 120o VCN = V + 120o

(2.43) (2.44)

We can express the line-to-line voltages in terms of the line-to-neutral voltages by direct application of Kirchoff's voltage law:
VAB = VAN VBN = V V 120o = 3V 30o VBC = VBN VCN = V 120o V 120o = 3V 90o VCA = VCN VAN = V 120o V 0o = 3V 150o

(2.45) (2.46) (2.47)

Equations (2.45) - (2.47) reveal that the magnitude of the line-to-line voltage is 3 times the magnitude of the line-to-neutral or phase voltage, and the set of line-to-line voltages leads (lags) the set of line-to-neutral voltages by 30 for positive (negative) sequences.

2.2.3

Power Calculations in Balanced Three-Phase Circuits

The total power in a three-phase balanced circuit, i.e., power delivered by a three-phase generator or absorbed by three-phase load is determined by adding the power in each of the three phases. In a balanced circuit this is the same as multiplying the power in any one phase by 3 since the power is the same in all phases. If the magnitude V of the voltages to neutral for a Wye-connected circuit is

V = Van = Vbn = Vcn

and if the magnitude I of the phase current for a Wye-connected circuit is
I = I an = I bn = I cn

(2.48)

(2.49)

the total three-phase power is

P = 3V I cos

(2.50)

where is the phase angle difference between the phase current I , and the phase voltage V . If V L and IL are the magnitudes of line-to line voltage VL and line current IL , respectively, V V = L and I = I L (2.51) 3 Substituting Eq. (2.51) into Eq. (2.50) yields

P = 3V L I L cos
The total vars are

(2.52)

Q = 3 V I sin Q = 3 VL I L sin
and the voltamperes of the load are

(2.53) 2.54)

S = P + jQ S = P 2 + Q 2 = 3 VL I L
(2.55)

Equations (2.50), (2.54), and (2.55) are used for calculating P,Q, and S in balanced three-phase networks since the quantities usually known are line-to-line voltage, line current, and the power factor,

pf = cos
When we refer to a three-phase system, balanced conditions are assumed unless otherwise specified, and the terms voltage, current, and power, unless otherwise specified, are understood to mean line-to-line voltage, line current, and total three-phase power, respectively. If the circuit is -connected, the voltage across each phase is the line-to-line voltage, and the magnitude of the current through each phase is the magnitude of the line current divided by 3 (see Section 2.1.1), or

V = V L
The total three-phase power is

and

I =

IL 3

(2.56) (2.57)

P = 3V I cos
Substituting Eq. (2.56) in Eq. (2.57) we obtain

P = 3 V L I L cos

(2.58)

which is identical to Eq. (2.52). It follows that Eqs. (2.54) and (2.55) are also valid regardless of whether a particular circuit is connected or Y.

2.2.4

Per-Phase Analysis

From the analysis in Sections 2.2.1, and 2.2.2, we observe that in balanced three-phase circuits the currents and voltages in each phase are equal in magnitude and displaced from each other by 120. This characteristic results in a simplified procedure to analyze balanced three-phase circuits. In this procedure it is necessary only to compute results in

one phase and subsequently predict results in the other phases by using the relationship that exists among quantities in the other phases.

Chapter 3. Transformers and Per-Unit Systems (My Bin p v H n V Tng i)

A primary function of the transformer is to convert electrical energy at one voltage level
to voltages at another level. The transformation may be to increase voltages or to decrease voltages, depending on the application. Transformers are the essential parts of most power systems since they are untilised to interconnect different parts of transmission and distribution power grids that operate at different voltage levels. Transformers also are found in almost all power supplies for small power electronic equipment such as TV sets, laptops, chargers A key application of power transformers is to reduce the current before transmitting electrical energy over long distances through wires. Most wires have resistance and so dissipate electrical energy at a rate proportional to the square of the current through the wire. By transforming electrical power to a highvoltage, and therefore low-current form for transmission and back again afterwards, transformers enable the economic transmission of power over long distances. Consequently, transformers have shaped the electricity supply industry, permitting generation to be located remotely from points of demand. All but a fraction of the world's electrical power has passed through a series of transformers by the time it reaches the consumer. In this chapter, we will deal with some simplified models of transformers without discussing about electromagnetic elements in details.

3.1

The Ideal Transformer

For analysis of transformers, it is convenient to start by using the ideal transformer

relations. Suppose we consider two coupled coils on a steel core of high magnetic permeability, a simplified model for this transformer can be shown in Figure 3.1.
i1 N1 : N2 + v1 _ i2 + v2 _

Figure 3.1 Ideal Transformer

If the flux varies sinusoidally, there will be a sinusoidal voltage generated in each turn of each coil according to Faraday. This quantity is called volt-per-turn and occupies an important role in transformer design. If windings 1 and 2 have N1 and N2 turns, respectively, then:

v1 v = 2 N1 N 2

(3.1)

One of the ideal transformer relations is that there can be no energy absorbed, stored or lost in the device. Whatever complex power enters one winding must leave the other. Therefore, we have * v1i1* = v 2 i2 (3.2) and i1 N 2 = (3.3) i2 N 1 The transformer also tends to transform impedances. In Figure 3.3, some impedance is connected to one side of the ideal transformer. We can find an equivalent impedance Z ' viewed from the other side of the transformer.
I1 N1 : N2 + V1 _ Z I2

Figure 3.2 Impedance Coupling

Noting that

N1 I 1 and V2 = ZI 2 N2 The ratio between input voltage and current is: I2 =

N1 N V1 = Z I 1 = 1 V2 = ZV1 N N2 2 We may derive the expression of the equivalent impedance viewed from the primary side of the transformer:
' 2

N1 Z = N 2
'

(3.4)

From the ideal transformer relations we see that the voltamperes into one winding of a two-winding transformer must equal the voltamperes out of a second winding. The voltampere rating of a two winding transformer is then given as the voltampere rating of either winding since the two are equal. In large power transformers the nameplate gives a voltamperes (or kVA or MVA) rating for the device as well as the voltage ratings of the two windings. The current ratings then follow from these data since S = VI. Small transformers, for example, those used in electronic power supplies, are often rated by giving the voltage and current ratings of each winding, from which the voltampere rating would follow if desired. No mention was made of power in the above statements. In a practical transformer the relative phase angle of voltage and current has almost no effect on the voltage and current capabilities of the windings, and hence the magnitude of S is the important factor and how S is divided into P and Q is immaterial to the rating.

3.2

Three-Phase Transformer

three-phase transformer is simply three single phase transformers. There are a number of ways of winding them, and a number of ways of interconnecting them. On either side of a transformer connection (i.e. the high voltage and low voltage sides), it is possible to connect transformers windings either line to neutral (wye), or line to line (delta). Thus we will have four connecting combinations: wye-wye, delta-delta, wyedelta, delta-wye. Ignoring all the imperfections, connection of transformers in either wye-wye or deltadelta is reasonably easy to understand. On the other hand, the interconnections of wyedelta or delta-wye transformer are a little more complex. Figure 3.3 shows a delta-wye connection, in what might be called wiring diagram form. A more schematic (and more common) form of the same picture is shown in Figure 3.4.

Yc

Yb

Ya

Figure 3.3 Delta-Wye Transformer Connection

Assume that N and NY are numbers of turns. If the three individual transformers are considered to be ideal, the following voltage and current constraints exist:

NY (va vb ) N N vbY = Y (vb vc ) N N (3.5) vcY = Y (vc va ) N N ia = Y (iaY icY ) N N ib = Y (ibY iaY ) N N ic = Y (icY ibY ) N where each of the voltages are line-neutral and the currents are in the lines at the transformer terminals. vaY =

Now, consider what happens if a -Y transformer is connected to a balanced three-phase voltage source, so that: va = Re V e jt

vb

2 j t 3 = Re V e

2 j t + 3 vc = Re V e Where: Re denotes the real part; V is the line-neutral (phase) voltage amplitude, an underline beneath the variable means it is a vector. Then the complex amplitudes on the wye side are:

j NY 6 V aY = 3 N Ve 2 2 j j NY j 3 NY 3 V bY = e e = 3 Ve 2 N N 2 5 j NY j 3 NY V cY = e 1 = 3 Ve 6 N N

N = Y N

2 j 1 e 3

Two observations should be made here: The ratio of voltages (that is, the ration of either line-line or line-neutral) is different from the turns ratio by a factor of 3 All wye side voltages are shifted in phase by 30o with respect to the delta side voltage.

It can be proved that impedances transform across transformers by the square of the voltage ratio, no matter what connection is used. As an example of some of the things said above, suppose that we read from the nameplate of a large transformer at a hydroelectric generating station the following rating data: 40 MVA, 115/24 kV. These data now tell us other things by using the ideal transformer relations, for example: N1 / N 2 = 115 / 24 = 4.79 since the voltage ratio and the turns ratio are the same under rating standards of large transformers. Also the rated current of the high voltage winding, which we call I1 is given by I 1 = 40 10 6 / 3 115 10 3

= 200.82 amperes rated current

and

I 2 = 40 10 6 / 3 24 10 3 = 962.25 amperes rated current for the low voltage winding It will be noted that this latter figure for I could also have been obtained by using the ideal transformer relation I1/I2 = N1/N2 if more convenient.

3.3

An Actual Transformer

The ideal transformer relations give very good answers to many transformer problems,
as in the examples preceding this section. For some problems, however, we must take account of the departures from perfection to get an adequate answer to a transformer problem. The first imperfection we will discuss is that of the core. The core is not infinitely permeable, it does require ampere turns to establish the flux, and in addition, there are internal energy losses in the core when the flux varies with time. Figure 3.4 shows the hysteresis properties of a transformer core. Each time the magnetic field is reversed, a small amount of energy is lost due to hysteresis within the core. For a given core material, the loss is proportional to the frequency, and is a function of the peak flux density (Bm) to which it is subjected.

Figure 3.4 (a) Hysteresis Loops of Steel; (b) The Normal Magnetization Curve

Transformer losses arising from the magnetic circuit, sometimes called iron loss. These losses are independent with load current, and may furthermore be expressed as "no-load" loss. Iron losses are caused mostly by hysteresis and eddy current effects in the core, and tend to be proportional to the square of the core flux for operation at a given frequency. The hysteresis loss occurs as an inherent property of the magnetic material. The internal structure of a ferromagnetic material is organized into domains and these domains are reoriented as the magnetic flux density - B vector goes through a cyclic change in magnitude or direction. An internal energy loss appearing is the result. The energy loss may be minimized by suitable alloying and heat treatment of the metal. The treatment processes may also affect the mechanical properties, however, so compromises must be made. If the frequency of the applied voltage were reduced but the range of B in the core maintained (by applying lower voltage), a similar loop would be observed but with smaller area than that originally observed. The reason for the larger area with higher frequency is the effect of eddy currents in the steel. The steel is a conductor and, as the flux in the steel varies, voltages are induced within the closed contours in the material. Currents flow as a result of the voltage and an I2R loss occurs known as eddy current loss. The loss is reduced by building the core from sheets of steel called laminations and by increasing the resistivity of the material by alloying. Since the core flux is proportional to the applied voltage, the iron loss can be represented by a resistance Rc (or a conductance Gc=1/Rc) in parallel with the ideal transformer. A core with finite permeability requires a magnetizing current IM to maintain the mutual flux in the core. The magnetizing current is in phase with the flux; saturation effects cause the relationship between the two to be non-linear, but for simplicity this effect tends to be ignored in most circuit equivalents. With a sinusoidal supply, the core flux lags the induced EMF by 90 and this effect can be modeled as a magnetising reactance

Xc ( or a susceptance Bc=1/Xc) in parallel with the core loss component. Rc and Xc are sometimes together termed the magnetising branch of the model. If the secondary winding is made open-circuit, the current I0 taken by the magnetising branch represents the transformer's no-load current. Figure 3.5 shows the use of a fictitious circuit added to the ideal transformer to account for exciting current.
B

Iex Gc Bc

Figure 3.5 Transformer Equivalent Circuit Accounted For Exciting Current

The equivalent circuit of Figure 3.5 is an approximation, but it is valid for most purposes. The advantages of using a linear model far outweigh the slight error introduced by a linear model. In any case the exciting current of a modern transformer of any size at all is only a very small percentage of the full load current, so if there is a slight error in the linear representation, it amounts to very little in terms of the total current passed by the transformer. The model to use for a certain study is of course a matter of experience and engineering judgment. For heavy load or short circuit studies the exciting current branch is normally omitted entirely. For light load or no load the exciting current branch may be included. For studies involving the wave form distortion of the exciting current the linear model is completely inadequate. Evaluation of Gc and Bc of Figure 3.5 involves an approximation to best model the actual transformer in some sense of a most useful model. The usual method of evaluating the parameters is to choose Gc and Bc so that the exciting current has the same rms value as the actual exciting current and the power loss in Gc is the same as the actual core loss. For example, suppose that a certain transformer is tested by applying rated voltage to a 10.5-kV winding with no load on the other winding, and it is observed that a current of 10 amperes flows and a power of 10000 watts is drawn. We solve for Gc and Bc as 2 Iex = EYc ; follows: P = Gc E 2 ; 10000 = Gc (10500) ; Gc = 90.7 10 6 siemen ;
B B B

10 = 10500Yc ;

Yc = 952.38 10 6 siemen ;

Yc2 = Gc2 + Bc2 ;

Bc = Yc2 Gc2 = 948.05 10 6 siemen .

In addition to the core loss and the requirement for exciting current, there are other ways in which an actual transformer differs from the ideal transformer model. For one thing, some of the applied voltage is absorbed in IR drop in the winding resistance. We can modify the ideal transformer model to account for winding resistance by adding two series resistors R1 and R2 in either side of the windings. Now, each winding has a

resistance which, while not zero, is kept low in order to minimize losses and increase efficiency. Flux leakage results in a fraction of the applied voltage dropped without contributing to the mutual coupling, and thus can be modeled as self-inductances Xl1 and Xl2 in series with the perfectly-coupled region. These series reactances (leakage reactance equivalent) play a significant factor in transformer performance.
R1 Xl1 Iex Gc Bc N1 N2 Xl2 R2

Figure 3.6 The Equivalent Circuit of An Actual Transformer

The secondary impedance R2 and Xl2 is frequently moved (or "referred") to the primary N side after multiplying the components by the impedance scaling factor ( 1 ) 2 N2
N1 N Xl2 2
2

R1

Xl1

N1 N 2

R2

N1

N2

Iex Gc Bc

Figure 3.7 Alternative Equivalent Circuit for the Transformer

The net series resistance and reactance are known simply as the impedance of the transformer. This is an item of data that is available from the manufacturer of the transformer. We designate the net resistance and reactance by the symbols Req and Xeq in the figure where: 2 Req = R1 + ( N1 / N2 ) R2
and

Xeq = Xl1 + ( N1 / N2 ) Xl 2
2

Open and short-circuit tests: The parameters of the equivalent circuits of Figure 3.8 may be determined by the designer from the physical dimensions and material properties of the transformer. On the other hand, an actual transformer may be tested electrically to determine these values. For the test, we apply rated voltage to the left side of the transformer of Figure 3.8(a) with the right side open-circuited. Since the output current is zero, the current through Req and Xeq is zero from properties of an ideal transformer. We thus "see" only the shunt branch and determine Gc and Bc from the instrument readings.
B

To determine the series impedances Req and Xeq, we short-circuit one side, and it is convenient to use the model of Figure 3.8(b) in this case. With a short circuit on N2, the voltage is zero on this winding and also is zero across N1 according to the properties of an ideal transformer. As a result we can ignore the shunt exciting current branch and we "see" only Req and Xeq.
Xeq Iex Gc Bc Req

(a)
Req Xeq Iex Gc Bc

(b) Figure 3.8 Simple Equivalent Circuit for the Transformer

3.4 Introduction to Per-Unit Systems

Per-unit systems are nothing more than normalizations of voltage, current, impedance
power, reactive power, and apparent power (volt-ampere). These normalizations of system parameters because they provide simplifications in many network calculations.

As we will discover, the transmission system and several portions of the distribution system are operated at voltages in the kV range. This results in large amounts of power being transmitted in the range of kilowatts to megawatts, and kilovoltamperes to megavoltamperes. As a result, in analysis, it is useful to scale, or normalize quantities with large physical values. This is commonly done in power system analysis and is referred to as the per-unit system. This helps in understanding how certain types of system behave. The numerical per-unit value of any quantity is its ratio to the chosen base quantity of the same dimensions. Thus a per-unit quantity is a normalized quantity with respect to a chosen base value.
Normalization of Voltage and Current:

The basis for the per-unit system of notation is the expression of voltage and current as fractions of base levels. Thus the first step in setting up a per-unit normalization is to pick up base voltage and current. Consider the simple situation in Figure 3.9. For this network the complex amplitudes of voltage and current are: V = IZ (an underline beneath the variable means it is a vector). We start by defining two base quantities, Vbase for voltage and Ibase for current. In many cases, these will be chosen to be nominal or rated values. For generating plants, for example, it is common to use the rated voltage and rated current of the generator as base qualities. In other situations, such as system stability studies, it is common to use a standard, system wide base system.
i1 + V _ Z + v1 _

Figure 3.9 Example

The per-unit voltage and current are then simply: V I v= , i= Vbase I base With V = IZ , we have

v=
Where the per-unit impedance is:

IZ I Z I base = = iz Vbase I base Vbase

I base Z = Vbase Z base This leads to a definition for a base impedance for the system: z=Z Vbase I base And there is also a base power, with for a single phase system is: Pbase = Vbase I base Z base =
As long as Vbase and Ibase are expressed in RMS (Root Mean Square). It is interesting to note that, as long as normalization is carried out in a consistent way, there is no ambiguity in per-unit notation. That is, peak quantities normalized to peak base will be the same, in per-unit, as RMS quantities normalized to RMS bases. This advantage is even more striking in polyphase systems.
Three Phase Systems:

In power system calculations the nominal voltage of lines and equipment is almost always known, so the voltage is a convenient base value to choose. The apparent power (volt-ampere - S) is usually chosen as a second base. In equipment this quantity is usually known and makes a convenient base. The choice of these two base quantities will automatically fix the base of current, impedance, and admittance. In a system study, the volt-ampere base can be selected to be any convenient value such as 100 MVA, 1000 MVA, etc. The same volt-ampere base is used in all parts of the system. One base voltage in a certain part of the system is selected arbitrarily. All other base voltages must be related to the arbitrarily selected one by the turns ratio of the connecting transformers. For single-phase systems or three-phase systems where the term current refers to line current, where the term voltage refers to line to neutral voltage, and where the term voltamperes refers to volt-amperes per phase, the following formulae relate the various quantities: S base ( ) I base = Vbase ( L N ) Z base = Vbase ( L N ) I base =

(V

base ( L N )

I baseVbase ( L N )

(V

base ( L N )

S base ( )

In performing per-phase analysis, the bases for the quantities in the circuit representation are volt-amperes per-phase or kilo-volt-amperes per phase, and volts or kilovolts from line to neutral. System specification is usually given in terms of total three-phase voltamperes or kilo-volt-amperes or mega-volt-amperes and line-to-line volts or kilovolts. This may result in some confusion regarding the relation between the per-unit value of line-to-line voltage and the per-unit value of phase voltage (line to neutral voltage). In a

per-phase circuit, the voltage required for the solution is the line to neutral voltage even though a line-to-line voltage may be specified as a base. The base value of the line to neutral voltage is the base value of the line-to line voltage divided by 3 . Since this is also the relation between line-to-line and line to neutral voltages of a balanced threephase system, the per-unit value of a line to neutral voltage on the line to neutral voltage base is equal to the per-unit value of the line-to-line voltage at the same point on the lineto -line voltage base if the system is balanced. Similarly, the three-phase volt-amperes is three times the volt-amperes per-phase, and the base value of the three-phase voltamperes is three times the base value of the per-phase volt-amperes. Therefore, the perunit value of the three-phase volt-amperes on the three-phase volt-ampere base is identical to the per-unit value of the volt-amperes per-phase on the volt-ampere perphase base. In a three-phase system, normally, a given value of base voltage is a line-to-line voltage, and a given value of base kilo-volt-amperes or base mega-volt-amperes is the total threephase base. The values of base impedance and base current can be computed from base values of voltage and volt-amperes as shown earlier in the section. If the base values of voltamperes and voltage are specified as the volt-amperes for the total three phases and voltage from line-to-line in a balanced three-phase system respectively, we have I base = Z base = Vbase ( L L ) 3I base = S base ( 3 ) 3Vbase ( L L )

(V

base ( L L )

3I baseVbase ( L N )

(V

base ( L L )

S base ( 3 )

Networks With Transformers:

One of the most important advantages of the use of per-unit systems arises in the analysis of networks with transformers. Properly applied, a per-unit normalization will cause nearly all ideal transformers to disappear from the per-unit network, thus greatly simplifying analysis.
I1 1 :N + V1 _ + V2 _ I2

Figure 3.10 An Ideal Transformer

To show how this comes about, consider the ideal transformer as shown in Figure 3.10. The ideal transformer imposes the constraints that: 1 V 2 = NV 1 , I 2 = I 1 N Note that an underline beneath the variable means it is a vector. Normalized to base quantities on the two sides of the transformer, the per-unit voltage and current are: V I v1 = 1 , i1 = 1 Vbase1 I base1
v2 = V2 I , i2 = 2 Vbase 2 I base 2

Note that if the base quantities are related to each other as if they had been processed by the transformer: 1 Vbase 2 = NVbase1 , I base 2 = I 1 N then v1 = v 2 and i 1 = i 2 , as if the ideal transformer were not there.
Transforming Form One Base To Another:

In most instances, the per-unit impedance of a component is specified on the rated component base which is different from the base selected for the part of the system in which the component is located. When performing calculations, all impedances in any one part of the system must be expressed on the same impedance base. As a result, it is necessary to have a means of converting per-unit impedances from one base to another. The process of changing this per-unit value of impedance to per-unit on a new base can be done as follows: Note that impedance (ordinary units) is given 2 (V ) = z new Z base _ new . Here, replace Z base = base , we can write: S base z old in Ohms by:

Z = z old Z base _ old

(V

base _ old

S base _ old

= z new

(V

base _ new

S base _ new

This yields a convenient rule for converting from the old base to the new one:
z new = S base _ new S base _ old

(V (V

base _ old

base _ new

) )

2 2

z old

In other word: Per Unit Z new = Per Unit Z old (

baseVold 2 baseVAnew ) ( ) baseV new baseVAold

Chapter 4. Electric Power Generation and Transmission Systems (H Thng Pht v Truyn Ti in)

The electrical power system is a network of interconnected components which generate

electricity by converting different forms of energy, (thermal, hydro, and nuclear are the most common forms of energy converted) to electrical energy; and transmit the electrical energy to load centers to be used by the consumer. Electricity is generated at the generating station by converting a primary source of energy to electrical energy. The output of the generators is then stepped-up to appropriate transmission levels using a step-up transformer. The transmission subsystem then transmits the power close to the load centers.

4.1

Electric Power Generation

Almost

all the energy conversion process take advantage of the synchronous AC generator coupled to a steam, gas, or hydro turbine such that the turbine converts steam, gas, or water flow into rotational energy, and the synchronous generator then converts the rotational energy of the turbine into electrical energy. It is the turbine-generator conversion process that is by far most economical and consequently most common in the industry today. In this section, we will briefly look at this conversion process with particular emphasis on the synchronous machine and the controls that are used to govern its behavior. A basic turbine-generator can be illustrated in Figure 4.1.
Steam Turbine Torque at speed Power at voltage Vt

Generator

+ ref

Governor

+ Vref

Excitation System

Vt

Figure 4.1 Block Diagram for Turbine-Generator System

The governor and excitation systems are known as feedback control systems because it is the feedback loops which provide for good control of certain parameters. The governor and excitation systems are typical feedback controllers in that the quantities to be controlled (speed and voltage, respectively) are also providing the feedback signal.

The generator is classified as a synchronous machine because it is only at synchronous speed that it can develop electromagnetic torque. If the nominal system frequency is f , synchronous speed is computed as 2 m = (e ) p where e = 2f is the frequency in rad/sec and p is the number of poles on the rotor of 120 the machine. The machine speed in RPM can be computed as N s = f. p The synchronous generator has two iron structures. The rotor is the revolving part of the machine, and is located inside the stator, which is the stationary part of the machine. Because hydro-turbines are relatively slow, the number of poles must be high in order to produce 50 Hz voltages (60 Hz in North America). Salient pole construction is simpler and more economical when a large number of poles are required. Steam plants, on the other hand, have very high speeds (1500 and 3000 RPM steam-turbine-generators are typical), and saliency would create significant mechanical stress at these speeds. Therefore, smooth or round rotor construction is employed for these generators. The two types of rotor construction are illustrated in Figure 4.2.

(a)

(b)

Figure 4.2 Salient Pole (a) and Round (b) Rotor Construction

A magnetic field is provided by the DC-current carrying field winding, which induces the desired AC voltage in the armature winding. The field winding is always located on the rotor where it is connected to an external DC source via slip rings and brushes or to a revolving DC source via a special brushless configuration. The armature consists of three windings, all of which are wound on the stator, physically displaced from each other by 120 degrees. It is through these windings that the electrical energy is produced and distributed. A typical layout for a 2 pole, round rotor machine would appear as in Figure 4.3.

A C B

B A

Figure 4.3 Winding Layout for Two-Pole Round Rotor Synchronous Machine

The DC current in the revolving field windings on the rotor produces a revolving magnetic field. We denote the flux associated with this field that links the armature windings as f (the subscript f indicates field windings). By Faradays Law of Induction, this rotating magnetic field will induce voltages in the three armature windings. Because these three windings are physically displaced by 120 degrees (for a two-pole machine), the induced voltages will be phase displaced in time by 120 degrees. If each of the three armature windings is connected across equal impedances, balanced three phase currents will flow in them. These currents will in turn produce their own magnetic fields. We denote the flux associated with each field as a , b , and c . The resultant field with associated flux obtained as the sum of the three component fluxes a , b , and c is the field of armature reaction. We designate the associated flux as ar . Using electromagnetic field theory and a trigonometric identity, one can show that ar revolves at the same velocity as the rotor. Therefore the two fields represented by f and ar are stationary with respect to each other. The armature field is effectively locked in with the rotor field and the two fields are said to be rotating in synchronism. The total resultant field is the sum of the field from the rotor windings and that associated with armature reaction: r = ar + f .

d r dt where is the number of winding turns. Because r is a sinusoidal function of time, the negative sign captures the fact that the induced voltage will lag the flux by 90 degrees. Letting E r , E ar , and E f be the voltages induced in winding a by the fluxes r ,
A voltage is induced in each of the three armature windings according to v = N

ar , and f , respectively, we can represent the relationships in time between the various
quantities using the phasor diagram, illustrated in Figure 4.4.

ar

Ef

Ear Ia Er

Figure 4.4 Phasor Diagram for Synchronous Generator

Regarding Figure 4.4, take note that All voltages lag their corresponding fluxes by 90 degrees. The current in winding a, denoted by I a , is in phase with the flux it produces ar If I a = 0 (no load conditions), then ar = 0 , and in this case, r = F , and Er = E F . All resistances have been neglected.

A Generator Equivalent Circuit can be illustrated in Figure 4.5.

jX Ia jXa Ef + Er _ jXl + Vt _ Load

Figure 4.5 Equivalent Circuit Model of Synchronous Generator

Defining X s = X l + X ar as the synchronous reactance, we have that Vt = E f jX s I a The phasor diagram corresponding to the equivalent circuit, when the load is inductive, is shown in Figure 4.6.

Ef -jXSIa Vt

Ia

Figure 4.6 Phasor Diagram for Equivalent Circuit Inductive Load

The phasor diagram corresponding to the equivalent circuit, when the load is capacitive, is shown in Figure 4.7.
Ia Ef

Vt
-jXSIa

Figure 4.7 Phasor Diagram for Equivalent Circuit Capacitive Load

When the load is inductive, the current I a lags the voltage Vt ; the generator is said to be operating lagging. When the load is capacitive, the current I a leads the voltage Vt ; the generator is said to be operating leading. The angle between I a and Vt is i, i.e., I a =| I a | I if Vt = | Vt | 0. This implies that i < 0 lagging, i > 0 leading The excitation voltage magnitude | E f | is higher in the lagging case. We sometimes refer to the lagging case as overexcited operation; here we have that | E f | cos >| Vt | , where is the angle between E f and Vt . The leading case results in under-excited operation; in this case we have E f cos < Vt .
Power Relationships: From our equivalent circuit in Figure 4.5, we write that Vt = E f jX s I a . Solving for I a yields

Ia =

E f Vt jX s

If is the angle at which the excitation voltage leads the terminal voltage. Therefore,

Ia =

| E f | Vt 0 jX s

| E f | cos + j | E f | sin Vt jX s

| E f | cos Vt jX s

j | E f | sin jX s

Ia =

| E f | sin
Xs

| E f | cos Vt j Xs
E f sin Xs

Equating real and imaginary parts of the above equation, we have I a cos = and I a sin = yields Pout = 3Vt I a cos = E f cos Vt Xs

. Multiplying both sides of the previous equations by 3Vt 3Vt E f sin Xs

3Vt Xs
2

Qout = 3Vt I a sin =

3Vt E f cos Xs

We can note that, reactive power out of the machine is positive when the machine is operated overexcited, i.e., when it is lagging implying < 0 . The above power relationships are based on the assumption that stator winding resistance is zero. The electrical power output Pout can be plotted against the power angle , resulting in sinusoidal variation as shown in Figure 4.8.

Pout Pmax

0 90
o

180o

Figure 4.8 Power Angle Curve

For simplicity, and without loss of generality, we neglect all real power losses associated with windage and heat loss in the turbine and friction in turbine and generator bearings. Continuing with the assumption that stator winding resistances are zero, in steady-state operation, the mechanical power input to the machine is equal to the electrical power:

Pmech = Pout . (In reality, Ploss > 0 in steady-state operation so that Pmech = Pout + Ploss ). Consider what happens to this lossless machine operating at Pout = Pmax ( = 90 ) when the steam valve opening is increased so that Pmech becomes slightly larger. In this case, the power angle increases beyond 90 , and the electrical power begins to decrease. However, the mechanical power is only dependent on the steam valve opening, i.e., it is unaffected by the decrease in Pout . This can only mean that Pmech > Pout . The difference Pmech Pout causes the machine to accelerate beyond its synchronous speed. When this happens, we say that the machine has pulled out, gone out of step, or lost synchronism. The generation level at which this happens is called the pull out power. It is given by 3Vt E f Pmax = Xs This limit is lower when the generator is under-excited (leading current) because E f is lower. Note that the excitation control system is used on synchronous machines to regulate terminal voltage, and the turbine-governor system is used to regulate the speed of the machine.

4.2

Electric Power Transmission

Electric power transmission, a process in the delivery of electricity to consumers, is the

bulk transfer of electrical power. Typically, power transmission is between the power plant and a substation near a populated area. Electricity distribution is the delivery from the substation to the consumers. Electric power transmission allows distant energy sources (such as hydroelectric power plants) to be connected to consumers in population centers. We may think of the transmission system as providing the medium of transportation for electric energy, but one must realize that this transportation system is unlike most in that the transportation takes place almost instantaneously. In addition, the transmission system is a highly integrated system; that is, a change in the status of any one component can significantly affect the operation of the entire system. Due to the large amount of power involved, and in order to enable long distance power transfer at lower current levels and therefore minimize | I 2 R | losses, transmission normally takes place at high voltage (110 kV or above). Electricity is usually transmitted over long distance through overhead power transmission lines. Underground cables are used predominantly in densely populated areas due to its high cost of installation and maintenance, and they are also often used for transmission under rivers, lakes, and bays. A power transmission system is sometimes referred to colloquially as a "grid"; however, for reasons of economy, the network is not a mathematical grid. Redundant paths and lines are provided so that power can be routed from any power plant to any load center, through a variety of routes, based on the economics of the transmission path and the cost of power. In this section, we will focus on transmission lines; substation equipment such as transformers, relays, and circuit breakers, will not be discussed.

4.2.1 Transmission Line Components

The basic components of a transmission line are the supports (towers), insulators, and the conductors. Operation of a transmission line is also dependent on fault detection equipment, voltage control equipment, and the bus arrangement at the terminals. In this chapter, we will focus on conductor characteristics. A single transmission circuit is comprised of three phases. Each phase may consist of a single conductor, or each phase may be bundled in that it consists of two or more conductors suspended from the same insulator string. The bundled design is usually used at the high voltage levels, because it minimizes power loss due to corona, which occurs when the surface voltage gradient of a conductor gets so high, that it causes partial dielectric breakdown in the surrounding air, and ionization occurs. Conductors are always bare in order to allow maximum heat dissipation. In addition to the phase conductors, one or two grounded shield wires are also strung along the top of the tower in order to protect the phase conductors from lightning strokes. Shield wires on transmission lines may include optical fibers (OPGW Optical Ground Wire), used for communication and control of the power system. Today, almost all conductors utilize aluminum in their construction because aluminum is relatively inexpensive, and lightweight. The most common types of conductors are commonly referred to by the following acronyms: AAC (all-aluminum conductor), AAAC (all-aluminum-alloy conductor), and ACSR (aluminum conductor, steel-reinforced). There are three main parameters of a conductor that are of concern to us. These are the resistance, inductance, and the shunt capacitance. We will discuss some of the influencing effects regarding these characteristics. Because we consider only balanced three-phase operation, discussion is limited to positive sequence quantities only.
Conductor Resistance: Conductors of any material have resistance. Although conductor resistance is small enough so that it does not appreciably contribute to voltage drop, it is of considerable interest in systems analysis because it causes | I | 2 R losses. The conductor resistance to direct current is given by R = l / A , where is the resistivity, | l | is the length, and A is the cross-sectional area of the conductor. However, there are two other important considerations regarding the DC resistance: (1) Values of resistivity are given for a specified temperature (e.g. 20 degrees C), and resistivity increases approximately linearly with temperature; (2) Because conductors are actually made with strands of material that are spiraled around a central core, the length used to compute resistance should be greater than the length of the conductor itself.

Due to the skin effect, resistance to AC is usually higher than the resistance to DC because AC causes current distribution in the conductor to be non-uniform; typically, more current tends to flow at the surface of the conductor than in the interior.

Conductor Inductance: Current flowing in a single conductor generates a magnetic field surrounding the conductor. We may denote the alternating current as i (t ) because it varies with time;

consequently, so will the magnetic field. We characterize this magnetic field with magnetic flux (t ) in Webers. The amount of flux which links this conductor is the flux linkage (t ) = N (t ) in Weber-turns, where N is the number of turns linked. In the case of a single conductor, N = 1 . We assume here that (t ) is given on a per unit length basis. According to Faradays Law, the magnitude of the induced voltage will be d (t ) vL (t ) = dt in units of induced volts per unit length of conductor. If the magnetic field is set up in a medium of constant permeability, then (t ) = Li (t ) where L is a constant. Substitute (t ) into equation of the induced voltage, we have di (t ) v L (t ) = L dt The constant L is defined as the inductance of the conductor, and it relates the voltage induced by the changing magnetic field to the rate of change of current: v (t ) L= L di (t ) / dt Here, L is given in units of henries per unit length. If we consider sinusoidal steadystate quantities, we have VL = jLI where the quantity X L = L is defined as the inductive reactance per unit length of the conductor, and V L represents the voltage drop per unit length across the conductor carrying current I. For a power transmission line, the inductive reactance is higher than the resistance by a factor of about 2 to 3 for lower voltage transmission lines, increasing to a factor of about 20 to 30 for the highest voltage transmission line. Consequently, inductive reactance is the dominant factor in computing voltage drop and power flow across a line.
Conductor Shunt Capacitance: Let us consider again a single conductor such that the return path is located far away from this conductor, that there is a charge on the peripheral of the conductor that is uniformly distributed throughout its length, and that an equal and opposite charge is distributed along the earth below the conductor. This charge generates an electric field emanating from the conductor directed towards the ground. If the conductor voltage is alternating, then we may denote it as v (t ) because it varies with time; consequently, so will the electric field. We characterize the electric field with the charge q (t ) in coulombs per unit length of conductor. The flow of charge per unit length caused by the changing voltage is current, given by

dq (t ) dt If the electric field is set up in a medium of constant permittivity, then q (t ) = Cv (t ) , where C is a constant. Substitution of q (t ) into iC(t) yields i C (t ) = dv (t ) dt The constant C is defined as the capacitance per unit length of the conductor, and it relates the current resulting from the changing electric field to the rate of change of voltage: iC (t ) = C

C=

iC (t ) dv(t ) / dt

If we consider sinusoidal steady-state quantities, then we can write I C = jCV , where the quantity BC = C is the capacitive susceptance per unit length of the conductor to ground. For three phase transmission lines, the capacitance between the phases is normally much larger than the capacitance to ground. Note the j in the above equation causes I C to lead V by 90 degrees. Therefore line capacitance, also called line charging, contributes leading current. In a power system, when the loads are heavy, currents of high magnitude in the lines result in heavy reactive losses ( | I | 2 X L ). The current in this case is lagging (since I = (V S V R ) / jX L ), and the effect of the leading current from the line capacitance is to bring the total current angle closer to zero degrees. This effect is desirable since the same real power flow can be obtained for a smaller current magnitude: P = 3 Re(VI * ) = 3V | I | cos , where is the angle between the current and voltage phasors; if decreases, then | I | can decrease for constant P , assuming that V remains approximately constant. Reducing current magnitude will decrease both real losses( | I | 2 R ) and reactive losses ( | I | 2 X L ). Another equally valid way to think about this is that the line inductance absorbs reactive power (MVAr) whereas the line capacitance produces reactive power (+MVAr). There is also a real part, the conductance, caused mainly by insulator leakage. However, this component is usually very small and negligible. It is rarely modeled in power system studies.
Lumped-Circuit Equivalent: The pi-equivalent lumped parameter model is used for most power flow analysis applications. This model represents the distributed effects of the series resistance and inductance and the shunt capacitance with composite or lumped values. Figure 4.9 illustrates the model.

p Y/2

jXL

Y/2

Figure 4.9 Pi-Equivalent Model of a Transmission Line

For transmission lines less than about 200 km in length, the lumped parameters are computed as simply the product of the per-unit length parameter and the line length. For lines that exceed 200 km, the model is still used but one needs to more rigorously derive appropriate expressions based on equivalent terminal characteristics.

4.2.2 Complex Power Transmission

In this section, we will develop equations for computing real and reactive power flow in a transmission line. We consider a transmission line that interconnects two buses p and q in the fig. 4.9. Therefore we may denote the series impedance as Z pq = R + jX L and the shunt admittance at each end of the line as Y pp = Yqq = Y / 2 where Y is the total line charging given by Y = jBC . In addition, we may represent the series impedance as R jX L 1 = 2 ). We will R + jX L R + X L 2 investigate the power flow into the transmission line from bus p in terms of two components: the flow into the p-side shunt capacitance, and the flow into the p to q series impedance. The total flow into the line from the p bus will then be the sum of these two components. Similar analysis will apply for the flow from bus q into the transmission line (which is just the negative of the flow from the transmission line into the q bus). admittance; thus we have Y pq = G jB ( Y pq = G jB = Using phasor notation, we denote the per unit voltages at the p and q buses as V p p and Vq q , respectively.
Flow Into Charging Capacitance: Let the current flowing into the p-side charging capacitance be | I Sp | . This current is given by (4.1) I Sp = V p p (Y pp ) = V p p ( jBC / 2)

The per phase complex power flowing into the p-side charging capacitance is then given by 2 S Sp = V p p ( I Sp ) * = V p p [V p p ( jBC / 2)]* = jV p ( BC / 2) (4.2)

where the asterisk indicates complex conjugation. Therefore the power flow into the charging capacitance is purely reactive, and the negative sign indicates that vars are being supplied to the network, not absorbed from it.
Flow Into Series Impedance: The per phase complex power flowing into the series impedance from the p bus is given by

S pq = V p p ( I pq ) *
where | I pq | is the current flowing into the series impedance, computed as
I pq = [V p p V q q ]Y pq = [V p p V q q ](G jB )

(4.3)

(4.4)

However, from eqn. 4.3, we see that | I pq | must be conjugated. Recall that if x = ab where a and b are two complex numbers, then x * = a * b * . Applying this relation to equation 4.4, we have that I pq = [V p p V q q ]* (G jB) * = [V p p V q q ](G + jB)
*

Substitution into eqn. 4.3 yields S pq = V p p [V p p V q q ](G + jB) = [V p V pV q ( p q )](G + jB)

2

Recalling that V p V q ( p q ) = V p V q cos( p q ) + jV p V q sin( p q ) , we may rewrite the last expression as S pq = [V p V pV q cos( p q ) jV pV q sin( p q )](G + jB)
2

Denoting p q = p q , carrying out the indicated multiplication, and then collecting real and imaginary parts, we have that S pq = V p G V p V q G cos p q + V pV q B sin p q +
2

j[V p B V pV q B cos p q V p V q G sin p q ]

2

(4.5)

The real and reactive power flow from bus p to bus q, measured at bus p, are then given by 2 (4.6) Ppq = V p G V pV q G cos p q + V pV q B sin p q Q pq = V p B V pV q B cos p q V pV q G sin p q
2

(4.7)

These equations are appropriate for computing real and reactive power flow across a transmission line. If voltages are given as phase to neutral, in volts, and impedances as per phase, in ohms, then the computed power quantities are per phase, in watts and vars. These equations can also be used if voltages and impedance values are given in per unit; in this case, the computed power quantities are also per unit, and multiplication by the system 3-phase base gives the three phase power flowing across the transmission line. Normally, R << X L so that we can assume R=0, and this implies G=0; then eqns. 4.6 and 4.7 become Ppq =
2

V pVq XL

sin p q

(4.8) (4.9)

Q pq =

V p V pVq cos p q XL

V p (V p Vq cos p q ) XL

These equations are very frequently used to analyse how certain flows are affected by design or operational actions that might be under consideration. In order to avoid system stability problems, the angular difference p q = p q across a transmission line is rarely allowed to exceed about 40 degrees; typically, angle differences are less than 20 to 30 degrees. This means that the arguments of the trigonometric functions in eqns. 4.8 and 4.9 are small angles. For small angles (given in radians) these trigonometric functions may be approximated with sin p q p q and
cos p q 1.0 , causing eqns. 4.8 and 4.9 to simplify to

Ppq =
Q pq =

V pV q p q XL
XL

V p (V P V q )

From these two equations, we see that real and reactive flows are heavily influenced by the the series reactance. Real power flow is closely related to differences between bus angles; this is especially apparent if we realize that normally, bus voltage magnitudes do not deviate substantially from 1.0 per unit. Reactive power flow is closely related to differences in bus voltage magnitudes. These concepts are fundamental to understanding control of real and reactive power flow in a transmission system.

Chapter 5. Introduction to Load Flow (Tnh Ton Ch Xc Lp HT)

In power engineering, the power flow study (also known as load-flow study) is an
important tool involving numerical analysis applied to a power system. Unlike traditional circuit analysis, a power flow study usually uses simplified notation such as a one-line diagram and per-unit system, and focuses on various forms of AC power (i.e: reactive, real, and apparent) rather than voltage and current. It analyses the power systems in normal steady-state operation. There exist a number of software implementations of power flow studies. In addition to a power flow study itself, sometimes called the base case, many software implementations perform other types of analysis, such as fault analysis and economic analysis. In particular, some programs use linear programming to find the optimal power flow, the conditions which give the lowest cost per kW generated. The great importance of power flow or load-flow studies is in the planning the future expansion of power systems as well as in determining the best operation of existing systems. The principal information obtained from the power flow study is the magnitude and phase angle of the voltage at each bus and the real and reactive power flowing in each line.

5.1

Power Flow

However, under balanced conditions (the currents in all three phases are equal in magnitude and phase separated by 120), we may analyze the three phase system using a per-phase equivalent circuit consisting of the a-phase and the neutral conductor. Per-unitization of a per-phase equivalent of a three phase, balanced system results in the per-unit circuit. It is the per-unitized, per-phase equivalent circuit of the power system that we use to formulate and solve the power flow problem. It is convenient to represent power system networks using the one-line diagram, which can be thought of as the circuit diagram of the per-phase equivalent. To show the power flow studies, consider the system with oneline diagram shown in Figure 5.1.

AC power transmission systems are comprised of three phase circuits.

Bus 1 G1 Z13 Z34 Bus 2 G2 Z24

Bus 3

Bus 5 Z35

P3+jQ3

P4+jQ4 Bus 4

Figure 5.1 System One-line Diagram

In this example system, the buses can be categorized as follows: Buses 1 and 2 are categorized as PV buses since their power (P) and voltage (V) magnitudes can be specified. Most generator buses fall into this category, independent of whether it also has load; exceptions are buses that have reactive power injection at either the generators upper limit (Qmax) or its lower limit (Qmin), and the system swing bus (we describe the swing bus below). There are also special cases where a non-generator bus (i.e., either a bus with load or a bus with neither generation or load) may be classified as type PV, and some examples of these special cases are buses having switched shunt capacitors or static var systems (SVCs). The real power injections of the type PV buses are chosen according to the system dispatch corresponding to the modeled loading conditions. The voltage magnitudes of the type PV buses are chosen according to the expected terminal voltage settings, sometimes called the generator set points, of the units. Buses 3 and 4 are all type PQ buses. For these buses we know P and Q but not |V | or angle . All load buses fall into this category, including buses that have not either load or generation. The real power injections of the type PQ buses are chosen according to the loading conditions being modeled. The reactive power injections of the type PQ buses are chosen according to the expected power factor of the load. Bus 5 can be referred to as the swing bus. Two other common terms for the swing bus type are slack bus and reference bus. In general, there is only one swing bus, and it can be designated by the engineer to be any generator bus in the system. For the swing bus, we know |V| and . The fact that we know is the reason why it is sometimes called the reference bus. Physically, there is nothing special about the swing bus; in fact, it is a mathematical artifact of the solution procedure. At this point in our treatment of the power flow problem, it is most appropriate to understand this last statement in the following way. The generation must supply both the load and the losses on the circuits. Before solving the power flow problem, we will know all injections at PQ buses, but we

will not know what the losses will be as losses are a function of the flows on the circuits which are yet to be computed. So we may set the real power injections for, at most, all but one of the generators. The one generator for which we do not set the real power injection is the one modeled at the swing bus. Thus, this generator swings to compensate for the network losses, or, one may say that it takes up the slack. Therefore, rather than call this generator a |V| bus (as the above naming convention would have it), we choose the terminology swing or slack as it helps us to better remember its function. The voltage magnitude of the swing bus is chosen to correspond to the typical voltage setting of this generator. The voltage angle may be designated to be any angle, but normally it is designated as 0o. Because the real power injection of the swing bus is not set by the engineer but rather is an output of the power flow solution, it can take on mathematically tractable but physically impossible values. Therefore, the engineer must always check the swing bus generation level following a solution to ensure that it is within the physical limitations of the generator.

5.2

Bus Admittance Matrix

The

bus admittance matrix Ybus is an important network description of the interconnected power system. In the bus admittance matrix representation, current injections at a bus are analogous to power injections. Current injections may be either positive (into the bus) or negative (out of the bus). Unlike current flowing through a branch (and thus is a branch quantity), a current injection is a nodal quantity. The admittance matrix, a fundamental network analysis tool that we shall use heavily, relates current injections at a bus to the bus voltages. Thus, the admittance matrix relates nodal quantities. It is easier to show the ideas of developing and using a Ybus matrix by introducing a simple example.

Figure 5.2 shows a network represented in a hybrid fashion using one-line diagram representation for the nodes (buses 1-4) and circuit representation for the branches connecting the nodes and the branches to ground. The branches connecting the nodes represent lines. The branches to ground represent any shunt elements at the buses, including the charging capacitance at either end of the line. All branches are denoted with their admittance values yij for a branch connecting bus i to bus j and yi for a shunt element at bus i. The current injections at each bus i are denoted by Ii.
1 y12 2 y14 y23 I2 y1 y2 y3 3 y34 4 I4

I1

I3 y4

Figure 5.2 Example Network

Kirchoffs Current Law (KCL) requires that each of the current injections be equal to the sum of the currents flowing out of the bus and into the lines connecting the bus to other buses, or to the ground. Therefore, recalling Ohms Law, I=V/z=Vy, the current injected into bus 1 may be written as: I1=(V1-V2)y12 + (V1-V4)y14 + V1y1 (5.1)

To be complete, we may also consider that bus 1 is connected to bus 3 through an infinite impedance, which implies that the corresponding admittance y13 is zero. The advantage to doing this is that it allows us to consider that bus 1 could be connected to any bus in the network. Then, we have: I1=(V1-V2)y12 + (V1-V3)y13 + (V1-V4)y14 + V1y1 (5.2)

Note that the current contribution of the term containing y13 is zero since y13 is zero. Rearranging eq. 5.2, we have: I1= V1( y1 + y12 + y13 + y14) + V2(-y12) + V3(-y13) + V4(-y14) Similarly, we may develop the current injections at buses 2, 3, and 4 as: I2 = V1(-y21) + V2( y2 + y21 + y23 + y24) + V3(-y23) + V4(-y24) I3 = V1(-y31)+ V2(-y32) + V3( y3 + y31 + y32 + y34) + V4(-y34) I4 = V1(-y41)+ V2(-y42) + V3(-y34)+ V4( y4 + y41 + y42 + y43) (5.4) (5.3)

where we recognize that the admittance of the circuit from bus k to bus i is the same as the admittance from bus i to bus k, i.e., yki=yik From eqs. (5.3) and (5.4), we see that the current injections are linear functions of the nodal voltages. Therefore, we may write these equations in a more compact form using matrices according to:
I 1 y1 + y12 + y13 + y14 I y 21 2 = I 3 y 31 y 41 I 4 y12 y 2 + y 21 + y 23 + y 24 y 32 y 42 y13 y 23 y 3 + y 31 + y 32 + y 34 y 43 V1 V 2 V3 y 34 y 4 + y 41 + y 42 + y 43 V4 y14 y 24

(5.5)

The matrix containing the network admittances in eq. (5.5) is the admittance matrix, also known as the Y-bus, and denoted as:
y1 + y12 + y13 + y14 y 21 Y = y 31 y 41 y12 y 2 + y 21 + y 23 + y 24 y 32 y 42 y13 y 23 y 3 + y 31 + y 32 + y 34 y 43 (5.6) y 34 y 4 + y 41 + y 42 + y 43 y14 y 24

Denoting the element in row i, column j, as Yij, we rewrite eq. (5.6) as:

Y11 Y Y = 21 Y31 Y41

Y12 Y22 Y32 Y42

Y13 Y14 Y23 Y24 Y33 Y34 Y43 Y44

(5.7)

where the terms Yij are not admittances but rather elements of the admittance matrix. Therefore, eq. (5.5) becomes:

I 1 Y11 Y12 I Y 2 = 21 Y22 I 3 Y31 Y32 I 4 Y41 Y42

Y13 Y23 Y33 Y43

Y14 V1 Y24 V2 Y34 V3 Y44 V4

(5.8)

By using eq. (5.7) and (5.8), and defining the vectors V and I, we may write eq. (5.8) in compact form according to:
V1 V V = 2 , V3 V4 I1 I I = 2 I3 I4
I = YV

(5.9)

We make several observations about the admittance matrix given in eqs. (5.6) and (5.7). These observations hold true for any linear network of any size: Y matrix is a symmetric, i.e., Yij=Yji. and square matrix that completely describes the configuration of power transmission lines. In realistic systems which are quite large containing thousands of buses, the Y matrix is quite sparse. Each bus in a real power system is usually connected to only a few other buses through the transmission lines. The admittances Y11, Y22,.. Ynn are called the self-admittances at the nodes, and each equals the sum of all the admittances terminating on the node identified by the repeated subscripts ( Yii = yi +
k =1, k i

ik

). The other admittances are the mutual admittances of the

nodes, and each equals the negative of the sum of all admittances connected directly between the nodes identified by the double subscripts (Yij=-yji). For small transmission systems of about less than 10 nodes or busses, Y matrix can be calculated manually. But for a realistic system with relatively large number of nodes or busses, say 1000 nodes, a computer program for computing Ybus is more practical to use.
Y Matrix Example: Consider the network given in Figure 5.1, where the numbers indicate admittances.

1 I1 1-j4 2

2-j3 3-j5 I2 3 2-j3

I4

I3 j0.2

j0.2

j0.3

j0.4

Figure 5.2 Example Circuit

The admittance matrix is given by inspection as:

Y11 Y Y = 21 Y31 Y41

Y12

Y13

Y22 Y23 Y32 Y33 Y42 Y43

Y14 3 j 6.8 1 + j 4 0 2 + j3 Y24 1 + j 4 4 j8.7 3 + j5 0 = Y34 0 3 + j 5 5 j 7.6 2 + j3 Y44 2 + j 3 0 2 + j3 4 j 5.8

5.3

Power Flow Equations

The net complex power injection into a general bus i can be expressed as Si=SGi-SDi,
where SGi is the generator power, SDi is the local load power. In this section, we desire to derive an expression for this quantity in terms of network voltages and admittances. All quantities are assumed to be in per unit, so we may utilize single-phase power relations. Drawing on the familiar relation for complex power, we may express Sk as:

Sk=VkIk*
From eq. (5.8), we see that the current injection into any bus k may be expressed as

(5.10)

I k = YkjV j
j =1

(5.11)

note that the Ykj terms are admittance matrix elements and not admittances. Substitution of eq. (5.11) into eq. (5.10) yields:
N N * * S k = Vk Y V = V k Ykj V j kj j j =1 j =1 *

(5.12)

Recall that Vk is a phasor, having magnitude and angle, so that Vk=|Vk|k. Also, Ykj, being a function of admittances, is therefore generally complex, and we define Gkj and Bkj as the real and imaginary parts of the admittance matrix element Ykj, respectively, so that Ykj=Gkj+jBkj. Then we may rewrite eq. (5.12) as
B

S k = Vk Ykj V j = Vk k (G kj + jBkj ) * V j j
* * j =1 j =1 N

= Vk k (Gkj jBkj ) V j j
j =1 N

)
(5.13)

= Vk k V j j (G kj jBkj ) = Vk V j ( k j ) (Gkj jBkj )

j =1 j =1

A phasor may be expressed as complex function V=|V|=|V|{cos+jsin}, we may rewrite eq. (5.13) as
S k = Vk V j ( k j ) (G kj jB kj )
j =1 N N

of

sinusoids,

i.e.,

= Vk V j (cos( k j ) + j sin( k j ) )(G kj jB kj )

j =1

(5.14)

= Pk + jQ k If we now perform the algebraic multiplication of the two terms inside the parentheses of eq. (5.14), and then collect real and imaginary parts, we can express eq. (5.14) as two equations, one for the real part, Pk, and one for the imaginary part, Qk, according to:
Pk = V k V j (G kj cos( k j ) + B kj sin( k j ) )
N j =1 N

Q k = V k V j (G kj sin( k j ) B kj cos( k j ) )
j =1

(5.15)

5.4

The Power Flow Problem

It is possible to specify six quantities at each bus: voltage magnitude (V) and angle (), current magnitude and angle, real and reactive power. These are, of course, inter-related so that any two of these are specified by the other four, and the network itself provides two more constraints. Typical combinations are: Generator bus: Active power PGi and terminal voltage |Vi| are specified (by varying turbine power and generator field current). Load bus: Real and reactive power are specified. Swing bus: This is a voltage source, of constant magnitude and phase angle.
Consider a power system network having N buses, NG of which are voltage-regulating generators. One of these must be the swing bus. Thus there are NG-1 type PV buses, and N-NG type PQ buses. We assume that the swing bus is numbered bus 1, the type PV

The typical load flow problem involves buses with different constraints.

buses are numbered 2,, NG, and the type PQ buses are numbered NG+1,, N. It is typical that we know, in advance, the following information about the network: The admittances of all series and shunt elements (the Y-bus can be obtained), The voltage magnitudes Vk, k=1,, NG, at all NG generator buses, The real power injection of all buses except the swing bus, Pk, k=2,, N The reactive power injection of all type PQ buses, Qk, k=NG+1, , N The last two statements indicate power flow equations for which we know the injections, i.e., the values of the left-hand side of eqs. (5.15). The total number of the power flow equations is (N-1)+(N-NG)=2N-1-NG. We repeat the power flow equations here, but this time, we denote the appropriate number to the right.
Pk = Vk V j (G kj cos( k j ) + Bkj sin( k j ) ) ,
N j =1 N

k = 2,..., N

Qk = Vk V j (G kj sin( k j ) Bkj cos( k j ) ) ,

j =1

(5.16)
k = N G + 1,..., N

We are trying to find the following information about the network: The angles for the voltage phasors at all buses except the swing bus (it is 0 at the swing bus), i.e., k, k=2,,N The magnitudes for the voltage phasors at all type PQ buses, i.e., |Vk|, k=NG+1, , N The above statements imply that we have N-1 angle unknowns and N-NG voltage magnitude unknowns, for a total number of unknowns of (N-1)+(N-NG)=2N-1-NG. Referring to the power flow equations, eq. (5.15), we see that there are no other unknowns on the right-hand side besides voltage magnitudes and angles (the real and imaginary parts of the admittance values, Gkj and Bkj, are known). Thus we see that the number of equations having known left-hand side (injections) is the same as the number of unknown voltage magnitudes and angles. Therefore it is possible to solve the system of 2N-NG-1 equations for the 2N-NG-1 unknowns. However, we note from eq. (5.16) that these equations are not linear, i.e., they are nonlinear equations. This nonlinearity comes from the fact that we have terms containing products of some of the unknowns and also terms containing trigonometric functions of some of the unknowns. Because of these nonlinearities, we are not able to put them directly into the familiar matrix form of Ax=b (where A is a matrix, x is the vector of unknowns, and b is a vector of constants) to obtain their solution. We must therefore resort to some other methods that are applicable for solving nonlinear equations. We can define the vector of unknown angles and voltage magnitudes |V| as follows: |V N G +1| 2 |V | = 3 , |V| = NG + 2 (5.17) M M |V N| N

Then we can define the vector x as the composite vector of unknown angles and voltage magnitudes.
2 x1 x 2 3 M M N x N 1 = x= = |V| |V N G +1| x N |V N + 2| x N +1 G M M |V | x 2 N 1 N N G

(5.18)

With this notation, we see that the right-hand sides of eqs. (5.16) depend on the elements of the unknown vector x. Expressing this dependence more explicitly, we rewrite eqs. (5.16) as
Pk = Pk ( x ) , Q k = Qk ( x ) , k = 2 ,...,N k = N G + 1,...,N

(5.19)

In eqs. (5.19), Pk and Qk are the specified injections (known constants) while the righthand sides are functions of the elements in the unknown vector x. Bringing the left-hand side over to the right-hand side, we have that
Pk ( x ) Pk = 0 , Qk ( x ) Qk = 0 , k = 2 ,...,N k = N G + 1,...,N

(5.20)

We now define a vector-valued function f(x) as:

f1 ( x) P2 ( x) P2 P2 0 M M M M f N 1 ( x) PN ( x) PN PN 0 =0 = = f ( x) = = f N ( x) Q N G +1 ( x) Q N G +1 Q N G +1 0 M M M M f x ( ) Q x Q Q ( ) 0 N N N 2 N 1 N G

(5.21) Equation (5.21) is in the form of f(x)=0, where f(x) is a vector-valued function and 0 is a vector of zeros; both f(x) and 0 are of dimension (2N-1-NG)1, which is also the dimension of the vector of unknowns, x. The vector of Pks and Qks is called the mismatch vector.

5.5

Newton-Raphson Iteration

There are two basic methods for solving the power flow problem: Gauss-Siedel (GS)
and Newton-Raphson (NR). Both of these methods are iterative root finding schemes. The GS and NR methods are often classified as root finding schemes because they are geared towards solving equations like f(x)=0 (or f(x)=0). The solution to such an equation, call it x* (or x*), is clearly a root of the function f(x) (or f(x)). The idea of the method is as follows: one starts with an initial guess which is reasonably close to the true root, then the function is approximated by its tangent line (which can be computed using the tools of calculus), and one computes the x-intercept of this tangent line (which is easily done with elementary algebra). This x-intercept will typically be a better approximation to the function's root than the original guess, and the method can be iterated. Figure 5.3 shows one iteration of Newton's method, we see that x(n + 1) is a better approximation than x(n) for the root x of the function f(x).
f(x)

x x(n+1) x(n)

Figure 5.3 Illustration of One Iteration of Newton's Method

Suppose we have some current approximation x(n). Then we can derive the formula for a better approximation, x(n + 1) by referring to the diagram below. We know from the definition of the derivative at a given point that it is the slope of a tangent at that point. That is
f (x
' ( n)

f ( x ( n) ) 0 ) = (n) x x ( n +1)

( n +1)

=x

(n)

f ( x ( n) ) ' (n) f (x )

We start the process off with some arbitrary initial value x0. The closer to the zero, the better. But, in the absence of any intuition about where the zero might lie, a "guess and check" method might narrow the possibilities to a reasonably small interval by appealing

to the intermediate value theorem. The method will usually converge provided this initial guess is close enough to the unknown zero. Furthermore, for a zero of multiplicity 1, the convergence is at least quadratic in a neighborhood of the zero, which intuitively means that the number of correct digits roughly at least doubles in every step.
NR Iteration Example:

Consider the scalar function f(x)=x2+x-6. This function may be easily factored to find the roots as x*=2 and -3. We first need the derivative: f(x)=2x+1. Let try an initial guess of x(0)=4. following provides the first two iterations: f(x(0))=f(4)=42+4-6=14 f(x(0))=f(4)=2(4)+1=9 x(0)= -f(x(0))/f(x(0))= -14/9=-1.55556 x(1)=x(0)+x(0)=4+(-1.55556)=2.444444

The

f(x(1))=f(2.444444)=2.41975 f(x(1))=f(2.44444)=5.88889 x(1)= -f(x(1))/f(x(1))= -2.41975/5.88889=-0.410901 x(2)=x(1)+x(1)=2.444444+(-0.410901)=2.033543 f(x(2))=f(2.033543)=0.16884 f(x(2))=f(2.03354)=5.067086 x(2)= -f(x(2))/f(x(2))= -0.16884/5.067086=-0.033321 x(3)=x(2)+x(2)=2.033543+(-0.033321)=2.00022

Note that by the third iteration, as it is getting very close to the correct solution, the algorithm has almost obtained quadratic convergence.
Newton Raphson for the Multidimensional Case:

Assume we have n nonlinear algebraic equations and n unknowns characterized by f(x)=0, and that we have guessed a solution x(0). Then f(x(0))0 because x(0) is just a guess. But there must be some x(0) which will make f(x(0) + x(0))=0. We can expand the function f(x) in a Taylor series, as follows:
f1 ( x f 2 (x
(0)

+ x ) = f1 ( x ) + f1 ' ( x ) x
( 0) ( 0) ( 0)

( 0)

( 0)

+ x ) = f 2 ( x ) + f 2 ' ( x ) x
( 0) (0) (0)

(0)

1 ( 0) ( 0) f1 ' ' ( x )( x ) 2 + ... = 0 2 1 ( 0) ( 0) + f 2 ' ' ( x )( x ) 2 + ... = 0 2 + 1 ( 0) (0) f n ' ' ( x )( x ) 2 + ... = 0 2

(5.22)

M
f n (x
(0)

+ x ) = f n ( x ) + f n ' ( x ) x
(0) (0) ( 0)

(0)

Equations (5.22) may be written more compactly as

f (x
(0)

+ x ) = f ( x ) + f ' ( x ) x
( 0) (0) (0)

( 0)

1 (0) ( 0) f ' ' ( x )( x ) 2 + ... = 0 2

(5.23)

Assuming the guess is a good one such that x(0) is small, then the higher order terms are also small and we can write
f (x
(0)

+ x ) = f ( x ) + f ' ( x ) x
( 0) ( 0) ( 0)

( 0)

=0

(5.24)

Since we have n functions and n variables, we could compute a derivative for each individual function with respect to each individual unknown, like fk(x)/xj, which gives the derivative of the kth function with respect to the jth unknown. Thus, there will be a number of such derivatives equal to the product of the number of functions by the number of unknowns, in this case, nn. Thus, it is convenient to store all of these derivatives in a matrix. This matrix has become quite well-known as the Jacobian matrix, and it is often denoted using the letter J. But how should the nn derivatives be stored in this matrix J? The rows of J should be ordered in the same order as the functions, that is, the kth row should contain the derivatives of the kth functions. In eq. (5.24), since the product f(x(0)) x(0) must provide a correction to the function f(x(0)+x(0)), i.e., since f(x(0)) = f(x(0)) x(0), it must be the case that elements of any row of the matrix J must be ordered so that the term in the jth column contains a derivative with respect to the jth unknown of the vector x. The reasoning in the last paragraph suggests that we write the Jacobian matrix as:
f1 ( x ( 0) ) x1 f 2 ( x ( 0) ) J = x 1 M ( 0) f n ( x ) x 1
(0) (0) f1 ( x ) f1 ( x ) L x 2 x n (0) (0) f 2 ( x ) f 2 ( x ) L x2 x n M M M (0) (0) f n ( x ) f n ( x ) L x2 x n

(5.25)

In eq. (5.24), taking f(x(0)) to the right hand side, we have

f ' ( x ) x
( 0) ( 0)

= f (x )
( 0)

(5.26)

or, in terms of the Jacobian matrix J, we have:

Jx

(0)

= f (x )
( 0)

(5.27)
1

(0)

= f '(x )
(0)

f (x ) = J
(0)

f (x )

(0)

(5.28)

This equation provides the basis for the update formula to be used in the first iteration of the multi-dimensional case. This update formula is:
x
(1)

=x

( 0)

+ x

( 0)

=x

( 0)

f (x )

( 0)

(5.29)

and from eq. (5.29), we may infer the update formula for any particular iteration as:
x
( i +1)

=x

(i )

+ x

(i )

=x

(i )

f (x )

(i )

(5.30)

For problems of relatively small dimension, where the inverse of the Jacobian is easily obtainable, eq. (5.30) is an appropriate update formula. In general, however, it is a good rule, in programming, to always avoid matrix inversion if at all possible, because for high-dimension problems, as is usually the case for large scale power networks, matrix inversion is very time consuming.
NR Iteration Example for the Multidimensional Case:

Solve the following two equations algebraically and using NR: 1.05x12-(x1+3)x2+3.2025x1+0.1575=0, x12 +x2-1.9525=0 The steps for the algebraic solution are to first solve both equations for x2, resulting in Equating these two x2=(1.05x12+3.2025x1+0.1575)/(x1+3) and x2=1.9525-x12. expressions for x2, and manipulating, results in a cubic x13+4.05x12+1.25x1-5.7=0. This expression may be factored as: (x1-0.95)(x1+2)(x1+3)=0, and we see that the solutions to the cubic in x1 are 0.95, -2 and 3. Plugging these values for x1 back into either expression for x2 yields, respectively, 1.05, -2.0475, and -7.0475, and therefore there are three solutions to the original problem; they are: (x1, x2)=(0.95,1.05), (-2,-2.0474), (-3,7.0475). Now lets solve this same problem using NR. Define the functions f1(x1,x2) = 1.05x12-(x1+3)x2+3.2025x1+0.1575 and f2(x1,x2) = x12 +x2-1.9525. Then the Jacobian matrix is: f 1 ( x1 , x 2 ) f1 ( x1 , x 2 ) 2.1x1 x 2 + 3.2025 x1 3 x1 x 2 = J = 2 x1 1 f 2 ( x1 , x 2 ) f 2 ( x1 , x 2 ) x x 1 2

Lets act like we do not know the solution and guess at (x1(0), x2(0))=(1,1). Then the Jacobian J, evaluated at this guessed solution, is 2.1x1 x 2 + 3.2025 x1 3 4.3025 4 ( 0) (0) = J ( x1 , x 2 ) = 2 x1 1 x(0) 2 1 Inverting the Jacobian results in: 1 4 0.0812843 0.32513717 4.3025 4 1 1 1 = = J = 1 12.3025 2 4.3025 0.162568584 0.349725665 2 We also need to evaluate:
f ( x ( 0) ) 1.05 x12 ( x1 + 3) x 2 + 3.2025 x1 + 0.1575 0.41 (0) f ( x ) = 1 (0) = = 2 x1 + x 2 1.9525 x ( 0 ) = (1.,1) 0.0475 f 2 ( x )

We can now update the solution using eq. (5.29), as (1) (0) (0) (0) (0) 1 x = x + x = x J f (x )
x (1) 1 0.081284 0.3251372 0.41 1 0.0487706 0.951229 = 1 (1) = = = x 2 1 0.1625686 0.3497257 0.0475 1 0.050041 1.050041

We see that the first update results in a solution that is very close to the actual solution of (0.95,1.05). This good performance is due to the fact that we made a good initial guess. If the second iteration will be done, we can get an almost exact solution. Similarly, we can repeat the above procedure, but starting from other points, e.g., (-1.9,-1.9) and (2.5,6) to get the other two solutions. Writing a simple program in will greatly reduce the effort. In general, we usually need to iterate several times in order to obtain a satisfactory solution. There are two ways to do determine when the solution is satisfactory. The first method is to test the maximum change in the solution elements from one iteration to the next, and if this maximum change is smaller than a certain predefined tolerance, then stop. This means to compare the maximum absolute value of elements in x against a small predefined number. The second method is to test the maximum absolute value in the function elements of the most current iteration f(x), and if this maximum value of elements in f(x) is smaller than a certain predefined tolerance, then stop. This means to compare the maximum absolute value of elements in f(x) against a small predefined number. This is the most common stopping criterion for power flow solutions, and the value of each element in the function is referred to as the power mismatch for the bus corresponding to the function. For type PQ buses, we test both real and reactive power mismatches. For type PV buses, we test only real power mismatches.

5.6

Application to Power Flow Equations

Recall

Lets consider the use of (5.30) to solve the power flow equations (5.21).
equations (5.16), (5.18), (5.21), (5.30) here for convenience:

Pk = Vk V j (G kj cos( k j ) + Bkj sin( k j ) ) ,

N j =1 N

k = 2,..., N

Qk = Vk V j (G kj sin( k j ) Bkj cos( k j ) ) ,

j =1

(5.16)
k = N G + 1,..., N

2 x1 x 2 3 M M N x = N 1 x= = |V| |V NG +1| x N |V N + 2| x N +1 G M M |V | x2 N 2 N G N Substitution of the above expressions into equation (5.21), we have:
f1 ( x) P2 ( x) P2 P2 M M M f N 1 ( x) PN ( x) PN PN = = f ( x) = = f N ( x) Q N G +1 ( x) Q N G +1 Q N G +1 M M M Q N Q N ( x) Q N f 2 N 1 N G ( x) N V2 V j (G 2 j cos( 2 j ) + B2 j sin ( 2 j )) P2 j =1 0 M N M ( ( ) ( ) ) + cos sin V V G B P N j Nj N j Nj N j N 0 j =1 = = = 0 N 0 V PN G +1 N G +1 V j G N G +1, j sin N G +1 j + B N G +1, j sin N G +1 j j =1 M M 0 N V N V j (G N , j sin ( N j ) + B N , j sin ( N j )) PN j =1

))

(5.21)
x
( i +1)

=x

(i )

+ x

(i )

=x

(i )

f (x )

(i )

An essential step in applying NR to the power flow problem is to enable calculation of the Jacobian elements, given for the general case by eq. (5.25) as

 f1 ( x ( 0) ) x1( 0) f 2 ( x ) J = x 1 M ( 0) f n ( x ) x 1

(0) (0) f1 ( x ) f1 ( x ) L x2 x n (0) (0) f 2 ( x ) f 2 ( x ) L x2 x n M M M (0) (0) f n ( x ) f n ( x ) L x2 x n

(5.25)

Note that there are only two kinds of equations (real power equations and reactive power equations), and that there are only two kinds of unknowns (voltage angle unknowns and voltage magnitude unknowns). Therefore, there are only four basic types of derivatives in the Jacobian. There are four types of derivatives as JP, JQ, JPV, JQV, where the first superscript indicates the type of equation we differentiate, and the second superscript indicates the unknown with respect to which we differentiate. We may then identify an individual element of each sub-matrix as:
J jk
P

Vk Vk k k Taking the partial derivatives of eq. (5.16), yields expressions as follows:

Pj

J jk

Q j

J jk

PV

Pj

J jk

QV

Q j

(5.31)

J jk
J jj

=
=

Pj ( x) k Pj ( x)
j

= V j Vk (G jk sin( j k ) B jk cos( j k ) )
= Q j ( x) B jj V j
2

J jk J jj

= =
=

Q j ( x) k Q j ( x) k Pj ( x)
Vk

= V j Vk (G jk cos( j k ) + B jk sin( j k ) ) = Pj ( x) G jj V j
2

J jk

PV

= V j (G jk cos( j k ) + B jk sin( j k ) )

J jj
J jk

PV

=
=

Pj ( x) Vj
Q j ( x) Vk

Pj ( x) Vj

+ G jj V j

QV

= V j (G jk sin( j k ) B jk cos( j k ) )

J jj

QV

Q j ( x) Vj

Q j ( x) Vj

B jj V j

We are now in a position to provide the algorithm for using NR to solve the power flow problem. The NR algorithm, for application to the power flow problem, is: 1. Specify:

 All admittance data Pd and Qd for all buses Pg and |V| for all PV buses |V| for swing bus, with =0 2. Set the iteration counter j=1. Use one of the following to guess the initial solution. Flat Start: Vk=1.0 0 for all buses. Hot Start: Use the solution to a previously solved case for this network. 3. Compute the mismatch vector for x(j), denoted as f(x) in eq. (5.21). In what follows, we denote elements of the mismatch vector as Pk and Qk corresponding to the real and reactive power mismatch, respectively, for the kth bus. This computation will also result in all necessary calculated real and reactive power injections. Perform the following stopping criterion tests: If |Pk|< P for all type PQ and PV buses and If |Qk|< Q for all type PQ buses, Then go to step 5 Otherwise, go to step 4. 4. Find an improved solution as follows: Evaluate the Jacobian J at x(j). Denote this Jacobian as J(j) Solve for x(j) from:
P = Q

( j)

( j)

or

( j)

=J

[ ]

(j) -1

P Q

where we must use factorization with the left equation if the system is large, but if the system is not large, we may use the right hand equation. Compute the updated solution vector as x(j+1)= x(j)+ x(j). Return to step 3 with j=j+1. 5. Stop. The above algorithm is applicable as long as all PV buses remain within their reactive limits. To account for generator reactive limits, we must modify the algorithm so that, at each iteration, we check to ensure PV bus reactive generation is within its limits. In this case, steps 1-4 remain exactly as given above, but we need a new step 5 and 6, as follows: 5. Check reactive limits for all generator buses as follows: a. For all type PV buses, perform the following test: If Qgk>Qgk,max, then Qgk=Qgk,max and CHANGE bus k to a type PQ bus (see step 6a) If Qgk< Qgk,min, then Qgk=Qgk,min and CHANGE bus k to a type PQ bus (see step 6b) b. For all type PQ generator buses, perform the following test: If Qgk=Qgk,max and |Vk|>|Vk,set| or if Qgk=Qgk,min and |Vk|<|Vk,set|, then

 CHANGE this bus back to a type PV bus (see step 6b) 6. If there were no CHANGES in Step 5, then stop. If there were one or more CHANGES in step 5, then modify the solution vector and the mismatch vector as follows: a. For each CHANGE made in step 5-a (changing a PV bus to a PQ bus): NG=NG-1 Include the variable Vk to the vector x and the variable Vk to the vector x. Include the reactive equation corresponding to bus k to the vector f(x). Modify the Jacobian by including a column to JPV and including a row to JQ and JQV. b. For each CHANGE made in Step 5-b (changing a PQ generator bus back to a PV bus): NG=NG+1 Remove the variable Vk to the vector x and the variable Vk from the vector x. Remove the reactive equation corresponding to bus k from the vector f(x). Modify the Jacobian by removing a column to JPV and removing a row from JQ and JQV. After modifications have been made for all CHANGES, go back to Step 4.

When the algorithm stops, then all line flows may be computed using * * S jk = V j I jk = V j [V j Vk ]Y jk