HARMONICS THEORY

Gary W. Chang Paulo F. Ribeiro

Siemens Power T&D BWX Technologies, Inc.
Brooklyn Park, MN, USA Lynchburg, VA 24505-0785

2.1 Introduction for all t. The smallest constant T that satisfies (2.1) is
called the period of the function. By iteration of (2.1), we
For most conventional analyses, the power system is have
essentially modeled as a linear system with passive
elements excited by constant- magnitude and constant- f ( t ) = f ( t + hT ), h = 0, ± 1, ± 2 , ... . (2.2)
frequency sinusoidal voltage sources. However, with the
widespread proliferation of power electronics loads Let a function f(t) be periodic with period T, then this
nowadays, significant amounts of harmonic currents are function can be represented by the trigonometric series
being injected into power systems. Harmonic currents not
only disturb loads that are sensitive to waveform ∞
∑
1
distortion, but also cause many undesirable effects on f (t ) = a0 + {ah cos( hω 0t ) + bh sin( hω 0t )} , (2.3)
power system elements. As a result, harmonic studies are 2 h =1
becoming a growing concern. where ω 0 = 2 π / T . A series such as (2.3) is called
Harmonics are usually defined as periodic steady state trigonometric Fourier series. It can be rewritten as
distortions of voltage and/or current waveforms in power ∞
systems. In the harmonic polluted environment, the theory
regarding harmonic quantities needs to be defined to
f ( t ) = c0 + ∑ ch sin( hω0t + φh ) , (2.4)
h =1
distinguish from those quantities defined for the
fundamental frequency. where c0 = a0 / 2 , ch = ah2 + bh2 , and φh = tan−1 ( ah / bh ) .
The purpose of this chapter is to present basic
harmonics theory. Initially, the Fourier series and analysis Observing (2.4), we see that the Fourier series expression
methods that can be used to interpret waveform of a periodic function represents a periodic function as a
phenomenon are reviewed. Some fundamentals of Fourier sum of sinusoidal components with different frequencies.
transforms used in today’s harmonics measurement The component of hω 0 is called the h-th harmonic of the
techniques are also introduced. The general harmonics periodic function. c0 is the magnitude of the dc component.
theory, the definitions of harmonic quantities, harmonic The component with h = 1 is called the fundamental
indices in common use, and power system response and component. ch and φh are known as the h-th order
solutions to harmonics are then described. harmonic magnitude and phase angle, respectively. The
magnitude and phase angle of each harmonic determine the
2.2 Fourier Series and Analysis resulting waveshape of f(t).
The theory of the Fourier series was first introduced by Equation (2.3) also can be represented by its complex
the French physicist and mathematician, Joseph Fourier, in form as
his article ‘Analytic Theory of Heat’ which was published
∞
in 1882. The theory involves expansions of arbitrary
functions in certain types of trigonometric series. It proves f (t ) = ∑ che jhω t ,0
(2.5)
that any periodic function in an interval of time could be h =−∞
represented by the sum of a fundamental and a series of where for h = 0, ± 1, ± 2 , ... ,
higher orders of harmonic components at frequencies
which are integral multiples of the fundamental 1 T /2
∫−T /2 f (t )e
− jhω 0 t
component. The series establishes a relationship between ch = dt . (2.6)
the function in time and frequency domains. Today, the T
theory has become the famous ‘Fourier series’ and it is one Orthogonal Functions
of the most important tools for engineers and scientists in
many applications. A set of functions {ϕh ( t )} is called orthogonal on an
interval α < t < β if all groups of any two functions ϕi ( t )
Fourier Series and ϕ j ( t ) in the set {ϕh ( t )} satisfy
A periodic function can be defined as any function for β  0, i ≠ j
which
f (t ) = f (t + T ) (2.1) ∫ ϕ (t )ϕ
α
i j ( t ) dt =
γ , i = j
, (2.7)
where γ is a nonzero value. It can be shown that synthesizes the frequency spectrum to regain the time
{1, cosω 0t , ..., coshω 0t , ...,sin ω 0t , ..., sinhω 0t , ...} is function.
an orthogonal set of sinusoidal functions on interval
− T / 2 < t < T / 2. Using the orthogonal relations, we can
show that Fourier coefficients a0 , ah , and bh of (2.3) are Discrete Fourier Transform
2 T /2 When the frequency domain spectrum and the time
a0 =
T −T /2 ∫f ( t ) dt , (2.8) domain function are both periodic sampled functions with N
samples per period, (2.14) and (2.15) can be represented by
2 T /2 the following so-called discrete Fourier transform (DFT)
ah = ∫
T −T /2
f (t ) cos( hω 0 t ) dt , and (2.9) pair:
N −1

∑ f ( n∆T ) e
2 T /2
bh =
T ∫−T / 2
f ( t ) sin(hω 0 t )dt , (2.10) F ( k ∆Ω ) =
n= 0
− j 2 πkn / N
, (2.16)

where h = 1, 2, ... . and

N −1
For the set of complex valued functions {ϕh ( t )}, it can f ( n∆T ) = ∑ F ( k ∆Ω)e j 2 πkn / N
, (2.17)
be shown that (2.7) holds when ϕ j ( t ) is the complex k =0
conjugate of ϕi ( t ) . where k, n = 0, 1, ..., N-1, ∆Ω = 2π / ∆T and ∆T = T / N .
The DFT is often used in harmonic measurement because
Waveform Symmetry the measured data is always available in the form of a
A function f(t) is called an even function if it has the sampled time function. The sampled time function is
property represented by a time series of points of known magnitude
f ( −t ) = f (t ) , (2.11) separated by fixed time intervals of limited duration.
and it is called an odd function if Fourier analysis can be done by DFTs. The DFTs are
often calculated by the use of fast Fourier transform (FFT)
f ( − t ) = − f (t ) . (2.12) algorithm [1]. FFT techniques are very fast methods for
An even function is symmetrical to the vertical axis at the performing the DFT calculations (2.16) and (2.17) which
origin, and an odd function is anti-symmetrical to the allow the evaluation of a large number of functions. There
vertical axis at the origin. A function with a period of T is are a number of available FFT algorithms that can be easily
half-wave symmetry if it satisfies the condition used in harmonic analysis.
f ( t ) = − f ( t ± T / 2) . (2.13) 2.3 Basic Definitions of Harmonic Quantities
If f(t) has half-wave symmetry and is either an even or odd Conventionally, the definitions used to describe electric
function, then it has even or odd quarter-wave symmetry. quantities for power system study are for systems operating
The use of symmetry simplifies the calculation of Fourier in sinusoidal steady state. However, when harmonics are
coefficients in (2.8) - (2.10). present because of system nonlinearities, the definitions of
these electric quantities need to be modified from those
Fourier Transform appropriate for single-frequency systems.
The Fourier transform of a function f(t) is defined as This section gives the basic definition of power system
∞ harmonics and describes some useful definitions associated
F (ω ) = ∫ -∞
f ( t ) e - jωt dt , (2.14) with voltage, current, instantaneous power, average (active)
power, apparent power, reactive power, and power factor
and f(t) is called the inverse Fourier transform of F ( ω ) , computations under nonsinusoidal situations.
which is defined as Definition of Power System Harmonics
1 ∞
f (t ) =
2π −∞ ∫
F (ω )e jωt dω . (2.15) In the power system, the definition of a harmonic can be
stated as: A sinusoidal component of a periodic wave having
Equations (2.14) and (2.15) are often called the Fourier a frequency that is an integral multiple of the fundamental
transform pair, and they are used to map any function in frequency. Thus for a power system with f 0 fundamental
the interval of ( −∞ , ∞) in time or frequency domain into a frequency, the frequency of the h-th order of harmonic is
continuous function in the inverse domain. The key hf 0 . Harmonics are often used to define distorted sinewaves
property of the Fourier transform is its ability to examine a associated with currents and voltages of different amplitudes
function or waveform from the perspective of both the time and frequencies.
and frequency domains. A given function can have two One can compose a distorted periodic waveshape of any
equivalent modes of representations: one is in the time conceivable shape by using different harmonic frequencies
domain and is called f(t), and the other is in the frequency with different amplitudes. Conversely, one can also
domain and is called F ( ω ) . Equation (2.14) transforms decompose any distorted periodic waveshape into a
the time function into a frequency spectrum, and (2.15) fundamental wave and a set of harmonics. This
decomposition process is called Fourier analysis. With this
technique, we can systematically analyze the effects of ∞ ∞
nonlinear elements in power systems. P= ∑
h =1
Vh I h cos( θh − δh ) = ∑P
h =1
h . (2.22)
Most elements and loads in a power system respond the
same in both positive and negative half-cycles. The We see that each harmonic makes a contribution, either plus
produced voltages and currents have half-wave symmetry. or minus, to the average power. There are no contributions
Therefore, harmonics of even orders are not characteristic. to the average power from the voltage at one frequency and
Also, triplens (multiples of third harmonic) always can be the current at another. The average power generated by
blocked by using three-phase ungrounded-wye or delta harmonics is usually very small in comparison with the
transformer connections in a balanced system, because fundamental average power.
triplens are entirely zero sequence. For these reasons, By applying orthogonal relations, the rms values of
even-ordered and triplens are often ignored in harmonic (2.18) and (2.19) are proved to be
analysis. Generally, the frequencies of interests for
∞

∑V
harmonic analysis are limited to the 50th multiple. 1 T
Vrms = ∫ v 2 (t )dt = 2
h , (2.23)
One major source of harmonics in the power system is T 0
h =1
the static power converter. Under ideal operating and
conditions, the current harmonics generated by a p-pulse ∞

∑I
1 T
∫
line-commutated converter can be characterized by
Ih = I1 / h and h = pn ± 1 (characteristic harmonics) where I rms = i 2 ( t ) dt = 2
h , (2.24)
T 0
h =1
n = 1, 2, ... and p is an integral multiples of six. If 1) the respectively.
converter input voltages are unbalanced or 2) unequal
commutating reactances exist between phases or 3) The apparent power is given by
unequally spaced firing pulses are present in the converter
bridge, then the converter will produce non-characteristic S = Vrms Irms . (2.25)
harmonics in addition to the characteristic harmonics. A widely accepted definition of apparent power is
Non-characteristic harmonics are those that are not integer
multiples of the fundamental power frequency. S 2 = P2 + Q2 + D2 , (2.26)
The harmonic frequencies that are not integral multiples where Q is the reactive power defined as
of the fundamental power frequency are usually called
interharmonics. A major source of interharmonics is the ∞
cycloconverter [2]. One special subset of interharmonics
is called sub-harmonics. Sub-harmonics have frequency
Q= ∑V I
h =1
h h sin( θ h − δh ) , (2.27)
values that are less than that of the fundamental frequency.
Lighting flicker is one indication of the presence of sub- and D is defined as the distortion voltamperes which
harmonics. A well-known source of flicker is the arc correspond to the products of voltages and currents of
furnace [3]. different frequency components in (2.18) and (2.19).
Electric Quantities Under Nonsinusoidal Situation When harmonics are not present in (2.25), S is equal to
When steady-state harmonics are present, instantaneous V1 I 1 which is the conventionally defined apparent power at
voltage and current can be represented by Fourier series as fundamental frequency. Under the sinusoidal situation, the
follows: power equation relates mutually the average, reactive, and
∞ ∞ apparent power, and it is defined as
v(t ) = ∑ vh ( t ) = ∑ 2Vh sin( hω 0t + θh ) , (2.18)
(V1 I1 )2 = P12 + Q12 , (2.28)
h =1 h =1
∞ ∞ where Q1 = V1 I1 sin( θ1 − δ1 ) is the fundamental reactive
i(t ) = ∑ ih ( t ) = ∑ 2 Ih sin( hω 0t + δ h ) , (2.19) power defined in (2.27) for h = 1.
h =1 h =1
At present, there is still no consensus in the definitions
where the dc terms are usually ignored for simplicity, Vh and physical meanings regarding reactive power and
and Ih are rms values for h-th order of harmonic voltage distortion power among researchers and scientists [4-7]. In
and current, respectively. [8], some alternate definitions with interpretations on power
The instantaneous power is defined as definitions other than the above are described.

p ( t ) = v ( t )i ( t ) , (2.20) The concept of power factor originated from the need to

quantify how efficiently a load utilizes the current that it
and the average power over one period T of p( t ) is defined draws from the ac power system. Regardless of sinusoidal
as or nonsinusoidal situation, the total power factor is defined
1 T as
P=
T 0
p(t )dt . ∫ (2.21)
P
pf = , (2.29)
If we substitute (2.18) and (2.19) into (2.20) and make use S
of the orthogonal relations of (2.7), it can be shown that
where P is the average power contributed by the which is defined as the ratio of the rms value of the
fundamental frequency component and other harmonic harmonic components to the rms value of the fundamental
components, as shown in (2.22). In the next section, we component and usually expressed in percent. This index is
also will show the relationship between the power factor used to measure the deviation of a periodic waveform
and some harmonic distortion indices. containing harmonics from a perfect sinewave. For a perfect
sinewave at fundamental frequency, the THD is zero.
Phase Sequences of Harmonics Similarly, the measures of individual harmonic distortion for
For a three-phase balanced system under nonsinusoidal voltage and current at h-th order are defined as Vh V1 and
conditions, the h-th order of harmonic voltage of each Ih I1 , respectively.
phase can be expressed as Total Demand Distortion
vah ( t ) = 2Vh sin( hω 0t + θh ) , (2.30) The total demand distortion (TDD) is the total harmonic
current distortion defined as
vbh ( t ) = 2Vh sin( hω 0t − 2 hπ / 3 + θh ) , (2.31) ∞

vch ( t ) = 2Vh sin( hω 0t + 2 hπ / 3 + θh ) . (2.32) ∑ Ih2

h =2
Therefore, the harmonic phase sequence in a balanced TDD = , (2.34)
IL
three-phase system has the pattern shown in Table 1.1.
where I L is the maximum demand load current (15- or 30-
Table 1.1. Harmonic Phase Sequences in a Balanced minute demand) at fundamental frequency at the point of
Three-Phase Power System common coupling (PCC), calculated as the average current
of the maximum demands for the previous twelve months.
Harmonic Phase The concept of TDD is particularly relevant in the
Order Sequence application of IEEE Standard 519.
1 +
Telephone Influence Factor
2 -
3 0 Telephone influence factor (TIF) is a measure used to
4 + describe the telephone noise originating from harmonic
currents and voltages in power systems. TIF is adjusted
5 - based on the sensitivity of the telephone system and the
6 0 human ear to noises at various frequencies. It is defined as
. . ∞ ∞

Observing Table 1.1, we find that the negative and zero ∑ ( whVh )2 ∑ ( wh Ih )2
h =1 h =1
sequences are also present in the system, and all triplens TIFV = or TIFI = , (2.35)
are entirely zero sequence. The above simple phase Vrms Irms
sequence pattern does not hold for the unbalanced system, where wh is a weighting accounting for audio and inductive
because harmonics of each order contain the three different coupling effects at the h-th harmonic frequency. Obviously,
sequences. It requires a more complicated analysis [9]. TIF is a variation of the previously defined THD where the
root of the sum of the squares is weighted using factors that
The definitions in (2.18) - (2.24) are also suitable for reflect the response in the voice band.
three-phase balanced system. However, for the
unbalanced system, the apparent power needs to be V·T and I·T Products
redefined and the consensus has yet to be reached.
Reference [10] provides some practical power definitions Another distortion index that gives a measure of
under unbalanced conditions. harmonic interference on audio circuits similar to TIF is the
V·T or I·T product, where V is rms voltage in volts, I is rms
2.4 Harmonic Indices current in amperes, and T is the TIF. In practice, telephone
interference is often expressed as V·T or I·T, which is
In harmonic analysis there are several important indices defined as
used to describe the effects of harmonics on power system
components and communication systems. This section ∞ ∞
describes the definitions of those harmonic indices in
common use [11-13].
V ⋅T = ∑ (w V
h =1
h h)
2
or I ⋅ T = ∑ ( wh Ih )2 , (2.36)
h =1
Total Harmonic Distortion (Distortion Factor) where wh is the same as previously described. If kV·T or
kI·T is used, then the index must be multiplied by a factor of
The most commonly used harmonic index is 1000. Equation (2.36) refers to the fact that the index is a
∞ ∞ product of harmonic voltage or harmonic current and the
∑ ∑ Ih2
corresponding telephone influence factor. Observing (2.35)
Vh2 and (2.36), we find that
h=2 h=2
THDV = or THDI = , (2.33) TIFV ⋅ Vrms = V ⋅ T and TIFI ⋅ I rms = I ⋅ T . (2.37)
V1 I1
C-Message Weighted Index Vrms = V1 1 + ( THDV / 100) 2 (2.41)
The C-message weighted index is similar to TIF, except and
that each weighting ch is used in place of wh . The
weighting is derived from listening tests to indicate the Irms = I1 1 + ( THDI / 100) 2 (2.42)
relative annoyance or speech impairment by an interfering by substituting (2.23) and (2.24) into (2.33). The total
signal of frequency f as heard through a “500-type” power factor in (2.29) becomes
telephone set. This index is defined as
P
∞ ∞ pf tot = . (2.43)
∑ (c V
h =1
h h)
2
∑ ( ch Ih ) 2 V1I1 1 + ( THDV / 100)2 1 + ( THDI / 100) 2
h =1
CV = or CI = . (2.38) In most cases, only very small portion of average power of
Vrms Irms P is contributed by harmonics and total harmonic voltage
distortion is less than 10%. Thus (2.43) can be expressed as
The relation between TIF weight and C-message weight is
P1 1
wh = 5ch f h , (2.39) pf tot ≈ ⋅
V1I1 1 + ( THD / 100) 2
I
where f h is the frequency of the h-th order harmonic. = cos( θ1 − δ1 ) ⋅ pf dist . (2.44)
Transformer K-Factor In (2.44), the first term, cos( θ1 − δ1 ) , is known as the
Transformer K-factor is an index used to calculate the displacement power factor, and the second term, pf dist , is
derating of standard transformers when harmonic currents defined as the distortion power factor. Because the
are present [14]. The K-factor is defined as displacement power factor is always not greater than one,
∞
we have
pf tot ≤ pfdist .
∑ h2 ( Ih / I1 )2 (2.45)
h =1 Obviously, for single-phase nonlinear loads with high
K= ∞
, (2.40)
current distortion, the total power factor is poor. It also
∑ ( Ih / I1 )2 should be noted that adding power factor correction
h =1 capacitors to such load is likely to cause resonance
where h is the harmonic order and Ih I1 is the conditions. An alternative to improve the distortion power
corresponding individual harmonic current distortion. factor is using passive or active filter to cancel harmonics
(2.40) is calculated based on the assumption that the produced by nonlinear loads.
transformer winding eddy current loss produced by each 2.5 Power System Response to Harmonics
harmonic current component is proportional to the square
of the harmonic order and the square of magnitude of the In comparison with the load, a power system is stiff
harmonic component. enough to withstand considerable amounts of harmonic
currents without causing problems. This means that the
The K-rated transformer is constructed to withstand system impedance is smaller compared to the load
more voltage distortion than standard transformers. The impedance. A power system itself is not a significant source
K-factor actually relates to the excessive heat that must be of harmonics. However, it becomes a contributor of
dissipated by the transformer. It is considered in the problems by way of resonance when severe distortion exists.
design and installation stage for nonlinear loads, and it is Assuming all nonlinear loads can be represented as
used as a specification for new or replacement power harmonic current injections, the harmonic voltage at each
source equipment. Table 2 shows typical commercially bus in a power system can be obtained by solving the
available K-rated transformers, where all regular following impedance matrix or nodal admittance equations
transformers fall into K-1 category. for all orders of harmonics under consideration:
Table 2: Commercially Available K-Rated Transformers Vh = Zh ⋅ Ih (2.46)
or
Category Ih = Yh ⋅ Vh , (2.47)
K-4
K-9 where Vh is the vector consisting of the h-th harmonic
voltage at each bus that is to be determined. Zh is the
K-13 system harmonic impedance matrix, Yh is the system
K-20 harmonic admittance matrix, and Ih is the vector of
K-30 measured or estimated harmonic currents representing the
K-40 harmonic-generating loads at connected busses.

Distortion Power Factor In (2.46), Zh can be obtained by using a Z-bus building

algorithm for each harmonic of interest or from the inverse
When voltage and current contain harmonics, it can be of Yh in (2.47), but the harmonic effects on different power
shown [15] that system components and loads need to be properly modeled
[16]. Approaches for harmonic analysis based on (2.46) or
(2.47) are commonly called current injection methods. Figure 2.2 shows the circuit topology in which parallel
These approaches are usually used in conjunction with resonance is likely to occur. Parallel resonance occurs when
fundamental frequency load flow computations. Through the parallel inductive reactance and the parallel capacitive
providing the network harmonic impedance or admittance reactance of the system are equal at certain frequency, and
and harmonic currents injected by nonlinear loads for all the parallel combination appears to be a very large
harmonics under consideration, the individual and total impedance to the harmonic source. The frequency where the
harmonic voltage distortions at each bus can be large impedance occurs is the resonant frequency. When
determined. Reference [16] also describes some other parallel resonance exists on the power system, significant
harmonic analysis methods. voltage distortion and current amplification may occur. The
highly distorted bus voltage may cause distorted currents
Observing (2.46), we see that system harmonic flowing in adjacent circuits. The amplified current may
impedance plays an important role in the system response result in equipment failure.
to harmonics, especially when resonance occurs in the
system. Resonance is defined as an amplification of power XL
system response to a periodic excitation when the
Power
Ih System
excitation frequency is equal to a natural frequency of the
system. For a simple LC circuit excited by a harmonic XC
current, the inductive and capacitive reactance seen from
the harmonic current source are equal at the resonant
Figure 2.2. Parallel Resonance
frequency f r = 1 / ( 2 π LC ) .
In a power system, most significant resonance problems When parallel resonance occurs in the circuit of Figure
are caused by a large capacitor installed for displacement 2.2, the resonant frequency can be determined by
power factor correction or voltage regulation purposes.
XC MVA SC
The resonant frequency of the system inductive reactance hr = = , (2.49)
and the capacitor reactance often occurs near fifth or XL MVAR CAP
seventh harmonic. However, resonant problems occurring
at eleventh or thirteenth harmonic are not unusual. There where MVA SC is the short-circuit MVA at the harmonic-
are two types of resonances likely to occur in the system: generating load connection point to the system and
series and parallel resonance. Series resonance is a low MVAR CAP is MVAR rating of the capacitor. It should be
impedance to the flow of harmonic current, and parallel understood that this approximation is only accurate for
resonance is a high impedance to the flow of harmonic systems with high X/R ratios.
current.
Another resonant scheme is shown in the distribution
Series Resonance network of Figure 2.3. If some of the feeder inductance
As shown in Figure 2.1, if the capacitor bank is in appears between groups of smaller capacitor banks, the
series with the system reactance and creates a low system may present a combination of many series and
impedance path to the harmonic current, a series resonance parallel resonant circuits, although the resonant effects are
condition may result. Series resonance may cause high somewhat less than that caused by one large resonant
voltage distortion levels between the inductance and the element. For this type of resonance problem, more
capacitor in the circuit due to the harmonic current sophisticated harmonic analysis programs must be employed
concentrated in the low impedance path it sees. Series to predict the harmonic characteristics of the system.
resonance often causes capacitor or fuse failures because Substation XL
of overload. The series resonant condition is given by
XC XC Ih
hr = , (2.48)
XL
where hr is the harmonic order of resonant frequency. Figure 2.3. Distributed Resonance

XL 2.6 Solutions to Harmonics

XC
Power Passive harmonic filters are an effective mitigation
Ih System method for harmonic problems. The passive filter is
generally designed to provide a path to divert the
troublesome harmonic currents in the power system.
Two common types of filters are the series and the shunt
Figure 2.1. Series Resonance filters. The series filter is characterized as a parallel
resonant and blocking type which has a high impedance at
Parallel Resonance its tuned frequency. The smoothing reactor used in power
electronics device is an example. The shunt filter is
characterized as a series resonant and trap type which has a
low impedance at its tuned frequency. The single tuned 2. R. F. Chu and J. J. Burns, “Impact of Cycloconverter
LC filter is the most common design in power systems. Harmonics,” IEEE Trans. on Industry Applications,
More detailed information on harmonic filter design and Vol. 25, No. 3, May/June 1989, pp. 427-435.
applications can be found in [12,17]. 3. R. C. Dugan, “Simulation of Arc Furnace Power
Systems,” IEEE Trans. on Industry Applications, IA-
Harmonic currents in a power system can also be 16(6), Nov/Dec 1980, pp.813-818.
reduced by providing a phase shift between nonlinear loads 4. A. E. Emanuel, “Powers in Nonsinusoidal Situations -
on different branches. One popular method called phase A Review of Definitions and Physical Meaning,” IEEE
multiplication is to operate separate six-pulse static Trans. on Power Delivery, Vol. 5, No. 3, July 1990,
converters (12-pulse and higher) in series on the dc side pp.1377-1389.
and in parallel on the ac side through the phase-shifting (∆- 5. A. E. Emanuel, “On the Definition of Power Factor and
∆ and ∆-Υ) transformers [18] so that there is self- Apparent Power in Unbalanced Polyphase Circuits,”
cancellation of some harmonics. Sometimes, a specially IEEE Trans. on Power Delivery, Vol. 8, No. 3, July
designed transformer (zigzag) is used to trap triplen 1993, pp.841-852.
harmonic currents and to prevent the currents flowing back 6. L. S. Czarnecki, “Misinterpretations of Some Power
to the source from the nonlinear load. This zigzag Properties of Electric Circuits,” IEEE Trans. on Power
transformer is usually designed to provide a low harmonic Delivery, Vol. 9, No. 4, October 1994, pp.1760-1769.
impedance between its windings compared to the source 7. P. S. Filipski, Y. Baghzouz, and M. D. Cox,
harmonic impedance. Thus there are circulating harmonic "Discussion of Power Definitions Contained in the IEEE
currents between the nonlinear load and the transformer. Dictionary,” IEEE Trans. on Power Delivery, Vol. 9,
Active filtering techniques [19] have drawn great No. 3, July 1994, pp.1237-1244.
attention in recent years. By sensing the nonlinear load 8. “Nonsinusoidal Situations: Effects on the Performance
harmonic voltages and/or currents, active filters use either of Meters and Definitions of Power,” IEEE Tutorial
1) injected harmonics at 180 degrees out of phase with the Course 90 EH0327-7-PWR, IEEE, New York, 1990.
load harmonics or 2) injected/absorbed current bursts to 9. K. Srinivasan, “Harmonics and Symmetrical
hold the voltage waveform within an acceptable tolerance. Components,” Power Quality Assurance, Jan/Feb 1997.
These approaches provide effective filtering of harmonics 10. IEEE Working Group on Nonsinusoidal Situations,
and eliminate some adverse effects of passive filters such “Practical Definitions for Powers in Systems with
as component aging and resonance problems. Nonsinusoidal Waveforms and Unbalanced Loads: A
Harmonic standards provide useful preventive solutions Discussion,” IEEE Trans. on Power Delivery, Vol. 11,
to harmonics. Recent standards such as IEEE 519-1992 No. 1, January 1996, pp. 79-101.
[11] and IEC 1000-3-2 [20] emphasize placement of limits 11. “Recommended Practices and Requirements for
on harmonic currents produced by nonlinear loads for Harmonic Control in Electric Power Systems," IEEE
customers and network bus harmonic voltage distortion for Standard 519-1992, IEEE, New York, 1993.
electric utilities. 12. J. Arrillaga, D. A. Bradley, and P. S. Bodger, Power
2.7 Summary System Harmonics, John Wiley & Sons, New York,
1985.
For harmonic studies, Fourier series and Fourier 13. G. T. Heydt, Electric Power Quality, Stars in a Circle
analysis are fundamental concepts. Many FFT algorithms Publications, West LaFayette, IN, 1991.
have been implemented for DFT computations on
measuring harmonics. 14. “IEEE Recommended Practice for Establishing
Transformer Capability When Supplying
In nonsinusoidal situations, the conventional electric Nonsinusoidal Load Currents,” ANSI/IEEE Standard
quantities used in sinusoidal environment need to be C57.110-1986, IEEE, New York, 1986.
redefined. However, power definitions as well as harmonic 15. W. M. Grady and R. J. Gilleskie, “Harmonics and How
phase sequences under unbalanced three-phase systems are They Relate to Power Factor,” Proceedings of PQA93,
still under investigation. Several harmonic indices have San Diego, CA, 1993.
been defined for the evaluation of harmonic effects on
power system components and communication systems. 16. Task Force on Harmonics Modeling and Simulation,
"Modeling and Simulation of the Propagation of
To predict precisely the power system response to Harmonics in Electric Power Networks Part I :
harmonics requires accurate models for power system Concepts, Models and Simulation Techniques," IEEE
elements and harmonic-generating loads. A simple Trans. on Power Delivery, Vol.11, No.1, January
technique for harmonic analysis is the current injection 1996, pp. 452-465.
method, which is performed in the frequency domain. 17. E. W. Kimbark, Direct Current Transmission, Vol. 1,
Other analysis methods include time domain and John Wiley & Sons, New York, 1971.
frequency/time domain techniques. Solutions to harmonics
can be classified as remedial and preventive. Passive and 18. N. Mohan, T. M. Undeland, and W. P. Robbins, Power
active filters are widely-used remedial solutions, and Electronics - Converters, Applications, and Design,
harmonic standards provide the best solution before actual John Wiley & Sons, New York, 1995.
harmonic problems occur. 19. W. M. Grady, M. J. Samotyj, and A. H. Noyola,
“Survey of Active Power Line Conditioning
References Methodologies,” IEEE Trans. on Power Delivery, Vol.
1. A. V. Oppenheim and R. W. Schafer, Discrete-Time 5, No. 3, July 1990, pp. 1536-1542.
Signal Processing, Prentice-Hall, Inc., Englewood
Cliffs, NJ, 1989.
20. “Limits for Harmonic Current Emmisions,”
International Electrotechnical Commission Standard
IEC 1000-3-2, March 1995.