Conservative Power Theory

UNICAMP – UNESP Sorocaba
August 2012
Conservative Power Theory

A theoretical background to understand energy issues of
electrical networks under non-
non-sinusoidal conditions
and
to approach measurement, accountability and control
problems in smart grids
Paolo Tenti
Department of Information Engineering
University of Padova, Italy
1
Seminar Outline
1. Motivation of work
2. Mathematical and physical foundations of the theory
• Mathematical operators and their properties
• Instantaneous and average power & energy terms in poly-
phase networks
3. Definition of current and power terms in single-phase
networks under non-sinusoidal conditions
4. Extension to poly-phase domain: 3-wires / 4-wires
5. Sequence components under non-sinusoidal
conditions
6. Measurement & accountability issues
2
Why do we need to define power terms
Describe physical phenomena
energy transfer,
energy storage,
rate of utilization of power sources and distribution
infrastructure,
unwanted voltage and current terms, ….
Allow unambiguous measurement of quantities
load and source characterization,
revenue metering, …
Compensation
identify provisions which make the equipment or the
plant compliant with standards & regulations in terms
of symmetry, purity of waveforms, power factor …
3
Few basic questions
While the definition and meaning of instantaneous
power and its average value (active power) are
universally agreed, the situation is less clear with other
popular power terms
What is/means reactive power ?
What is/means distortion power ?
What is/means apparent power ?
These power terms are unambiguously defined when at
least the voltage supply is sinusoidal, but are matter of
controversial discussions (since nearly one century) in
case of distorted voltages and currents.
4
Milestones of power theory history
In the frequency domain
Budeanu (1927)
Sheperd & Zakikhani (1971)
Czarnecki (1984 …)
In the time domain
Fryze (1931)
Kusters & Moore (1975)
Depenbrock (1993)
Akagi & Nabae (1983)
No one of these theories was able to target all goals (characterization

of physical phenomena, load & line identification, compensation).
The time-
time-domain theory presented here tries to target all goals at the
same time.
It represents an outcome of a long
long--standing cooperation between
UNIPD, UNICAMP and UNESP. 5
5
Need for a revision of power terms
In modern scenarios (e.g., micro-grids) where:
the grid is weak,
frequency may change,
voltages may be asymmetrical,
distortion may affect voltages and currents,
are the usual definitions of reactive, unbalance and
distortion power still valid ?
Which is the physical meaning of such terms ?
Are they useful for compensation ?
To which extent are power measurements affected by
source non-ideality ?
It is possible to identify supply and load responsibility
on voltage distortion and asymmetry at a given network
port ? 6
Conservative
Power Theory
2. Mathematical and physical foundations
• Definition of mathematical operators and their
properties
• Definition of instantaneous power and energy terms
• Conservative quantities
• Selection of voltage reference
• Definition of average power terms and their physical
meaning in real networks
7
Mathematical operators for periodic
scalar quantities
Let T be the period of variables x and y, we define:
∫0 x(t ) dt
1 T
• Average value x= x =
T
( dx
• Time derivative x=
dt
x∫ = ∫ x (τ ) dτ
t
• Time integral 0
)
• Unbiased time integral x = x∫ − x∫
1 T
• Internal product x, y = ∫ x ⋅ y dt
T 0
• Norm (rms value) X = x = x, x
• Orthogonality x, y = 0
8
Mathematical operators for periodic
vector quantities
Let x and y be vector quantities of size N, we define:
N
• Scalar product xo y = ∑ xn y n
n =1
N
• Magnitude x = xo x = ∑ xn2
n =1
N
• Internal product x, y = x o y = ∑ x n , y n
n =1
N N
• Norm X= x = ∑ xn , xn = ∑ X n2
n =1 n =1
• Orthogonality x, y = 0
• The vector norm is also called collective rms value
9
Properties of mathematical operators
(valid for scalar and vector quantities)
The above operators have the following properties:

( (
x, x = 0 x, x = 0
• Orthogonality ) ⇒ )
x, x = 0 x, x = 0
• Equivalences
( ( ( (
x, y = − x , y x , y = − x ,y
) ) ) )
x, y = − x , y ⇒ x, y = − x , y
( ) ) ( ( ) ) (
x, y = − x , y = − x , y x, y = − x , y = − x , y
• For sinusoidal quantities
1 ( (2
) )2 x
X = x =ω x = x x + ω x = x 2 + 2 = 2X 2
2 2
ω ω
) 1
x, y = XY cosϕ x , y = XY senϕ
ω 10
Instantaneous power definitions
(for periodic variables)
Given the vectors of the N phase currents in and voltages
un measured at a generic network port we define:
N N
Instantaneous (active) power: p = u ⋅i = ∑ u n in = ∑ p n
n =1 n =1
N N
) )
∑nn ∑ n
Instantaneous reactive energy w = u ⋅ i = u i = w
(new definition): n =1 n =1
• Both quantities do not depend on the voltage reference

• Both quantities are conservative in every real network
11
Conservation of instantaneous
power and reactive energy
For every real network π, let u and i be the vectors of
the L branch voltages and currents, we claim that:
Branch voltages, their time derivative and unbiased
integral are consistent with network π, i.e. they
comply with KLV (Kirchhoff’s law for voltages)
Branch currents, their time derivative and unbiased
integral are consistent with network π, i.e. they
comply with KLC (Kirchhoff’s law for currents)
) ( ( )
Thus, according to u ⋅i = u ⋅i = u ⋅i = 0
Tellegen’s Theorem all ) )
quantities shown here are u ⋅i = u ⋅i = 0
( (
conservative u ⋅i = u ⋅i = 0
12
Average power definitions
(valid for periodic quantities)
Active power: P = p = u ,i
) )
Reactive energy: W = w = u ,i = − u ,i
Apparent power: A = u i = UI
P
Power factor: λ=
A
• All quantities are defined in the time domain.
• Reactive energy is a new definition, whose properties will be
analyzed in the following.
• Active power and reactive energy are conservative
quantities which do not depend on the voltage reference.
• Unlike P and W, apparent power A is non-conservative and
depends on the voltage reference. Skip voltage reference 13
Selection of voltage reference (1)
Cauchy-Schwartz P
u, i ≤ u i ⇒λ = ≤1
inequality: A
The equal sign is u ∝ i ⇒ u, i = u i ⇒ λ =1

possible if:
We select the voltage reference so as to ensure unity

power factor in case of symmetrical resistive load. This
gives a physical meaning to the apparent power, which
is the maximum active power that a supply line rated
for Vrms Volts and Irms Amps can deliver to a (purely
resistive and symmetrical) load.
14
N-phase systems without neutral wire
The proportionality condition u = Ri
between phase voltages and N N
currents for symmetrical resistive
load determines voltage reference n =1
∑
in = 0 ⇒ ∑ un = 0
n =1
Thus, the voltage reference must be selected to comply

with the zero-sum condition:
∑ (1un442−4u4
ref ) = 0
N N N
1
∑ un = 0 ⇒
3
mea sure
⇒ u ref =
N
∑ un measure
n =1 n =1 n =1
un
This choice minimizes the norm of the voltage vector
15
N-phase systems without neutral wire
Measurement of voltages and currents
SOURCE LINE LOAD
3
i3
u32 R3 L3
2
uSn u21 R2 L2
1
i1
R1 L1
n∈ {1,2,3} RLn LLn
Derivation of phase voltages

N
1
un =
N
∑ j ≠ n un j
j =1
1  N 2
N N
1
∑∑
2
u = = ∑ − u , n = 1÷ N
2
j ≠ n un j u n2 u 2
j≠n n j
2N n =1 j =1 N  j =1 

16
N-phase systems with neutral wire
In case of symmetrical resistive load the proportionality
condition between phase voltages and currents holds only
if the voltage reference is set to the neural wire.
u ref = u o = 0 ⇒ u n = R in , n = 0 ÷ N
Unity power factor may occur only if the neutral current is

disregarded for apparent power computation (only phase
currents are considered). Thus:
N  N  N  N 
A = P = U I, U = ∑ U n2  = ∑ n I=
U 2 
∑ n ≠
I 2 
∑ n I 2 

n =1  n =0  n =1  n =0 
17
N-phase systems with neutral wire
Measurement of voltages and currents
SOURCE LINE LOAD
3
i3
R3 L3
2
i2
R2 L2
1
i1
R1 L1
vSn u30 u20 u10 0
n∈ {0,1,2,3} RLn LLn
N N
Collective rms voltage
and current
U= ∑ U n2 I= ∑ I n2
n =1 n =1
Homopolar voltage 1 N
1 N
io
and current u =
z
N
∑ un i =
z
N
∑ in = − N
n=1 n =1
18
Power terms in passive networks
Resistor
u = Ri
i = Gu
2 2
PR = u , i = G u =R i
) )
WR = u , i = R i , i = 0
19
Inductor
di (
u=L = Li
dt
)
u
i=
L
)
u ) 2
PL = u , i = u , =0 WL = u , i = L i, i = L i
L
Inductor 1 2 1 2 WL
ε L = L i ⇒ ε L = ΕL = L i =
energy 2 2 2
20
Capacitor
du (
i=C = Cu
dt
)
i
u=
C
)
i ) 2
PC = u , i = ,i = 0 WC = − u , i = − u , C u = −C u
C
Capacitor 1 2 1 2 WC
energy
ε C = Cu ⇒ ε C = Ε C = C u =−
2 2 2
21
Active and reactive power absorption of a
linear passive network π
Remark: Whichever is
L=N+M+K π the origin of reactive
energy, including
N resistors
active and nonlinear
M inductors loads, it can be
compensated by
K capacitors
reactive elements
with proper energy
storage capability
Total active power and reactive energy
L N
P = ∑ ul , il =∑ PRn = PRto t
l =1 n =1
L M K M K
)
W =∑ ul , il = ∑ WLm + ∑ WCk = 2 ( ∑ Ε Lm − ∑ Ε Ck ) =2 (Ε Lto t − Ε Ctot )
l =1 m =1 k =1 m =1 k =1
22
Seminar Outline
phase networks
conditions
23
Conservative
Power Theory
3. Definition of current and power terms in
single-phase networks under non-
sinusoidal conditions
• Orthogonal current decomposition into active, reactive
and void terms
• Physical meaning of current terms
• Apparent power decomposition into active, reactive and
void terms
• Physical meaning of power terms
• Application examples 24
Orthogonal current decomposition in
single--phase networks
single
(voltage and current measured at a generic network port)
Current terms
i = ia + ir + iv = ia + ir + isa + isr + i g
14243
iv
• ia active current • isa scattered active current

• ir reactive current • isr scattered reactive current
• iv void current • ig generated current
Orthogonality
Orthogonality:: all terms in the above equations are orthogonal
2
i = ia + ir + iv = ia + ir + isa + isr + ig
2 2 2 2 2 2 2 2
25
single
Active current: the minimum current (i.e., with minimum

rms value) needed to convey the active power P flowing
through the port
u = port voltage
uu,, i P
ia = 2
u = 2 u = Geu U = rms value of port voltage
u U Ge = equivalent conductance
Pa = u, ia = Ge u, u = Ge U 2 = P Active current conveys full

) ) active power and zero
Wa = u , ia = Ge u , u = 0 reactive energy
26
single
Reactive current: the minimum current needed to

convey the reactive energy W flowing through the port
)
u, i ) W ) )
ir = ) 2 u = ) 2 u = Be u Be = equivalent reactivity
u U
)
Pr = u, ir = Be u, u = 0 Reactive current conveys full
) ) ) )2 reactive energy and no active
Wr = u , ir = Be u , u = BeU = W power
) Active and reactive current

ia , ir = Ge Be u, u = 0 are orthogonal
27
single
Void current: is the remaining current component

iv = i − ia − ir
Void current is not conveying active power or reactive energy
Pv = u, iv = u, i − u, ia − u, ir = P − Pa − Pr = 0
) ) ) )
Wv = u , iv = u , i − u , ia − u , ir = W − Wa − Wr = 0
Void current is orthogonal to active and reactive terms
iv , ia = Ge iv , u = Ge Pv = 0
)
iv , ir = Be iv , u = BeWv = 0
28
single
The void current reflects the presence of scattered active,

scattered reactive and load-generated harmonic terms
iv = isa + isr + ig
Scattered current terms: Load-generated current

Account for different values of harmonics:
equivalent admittance at Harmonic terms that exist in
different harmonics currents only, not in voltages
Scattered and load-generated

isa , isr = isa , i g = isr , i g = 0 harmonic currents are
orthogonal
Skip void current components 29
single
Scattered active current
For each co
co--existing harmonic components of voltage
and current we define:
define:
Harmonic active current terms
uk , ik Pk I k cos ϕ k
iak = 2
uk = 2 uk = uk = Gk u k
uk Uk Uk
Total harmonic active current
iha = ∑ iak Pha = ∑ Pk = Pa = P, Wha = 0
k∈K k∈K

isa = iha − ia = ∑ (Gk − Ge )uk Psa = Pha − Pa = 0, Wsa = 0
k∈K
30
single
Scattered reactive current
For each co
co--existing harmonic components of voltage
and current we define:
define:
Harmonic reactive current terms
)
u k , ik ) Wk ) ω k I k sin ϕ k ) )
irk = ) 2
uk = ) 2 uk = u k = Bk u k
uk Uk Uk
Total harmonic reactive current

ihr = ∑ irk Whr = ∑ Wk = Wr = W , Phr = 0
k∈K k∈K

isr = ihr − ir = ∑ (Bk − Be )u k
)
Wsr = Whr − W = 0, Psr = 0
k∈K
31
Apparent power decomposition in
single
A = u i = U I = P2 + Q2 +V 2
Active power: P = u ia = U I a
Reactive power: Q = u ir = U I r
Void power: V = u iv = U I v = S a2 + S r2 + V g2
Scattered active power: S a = u isa = U I sa
Scattered reactive power: S r = u isr = U I sr
Load-generated harmonic V = u i = U I
g g g
power:
32
Reactive Power
U and Û can be decomposed in U = U 2 + U 2 = U 1 + [THD(u )] 2
f h f
fundamental and harmonic
components ) )2 )2 )
U = U f + U h = U f 1 + [THD(u )]
) 2
(THD means total harmonic distortion)
)
Recalling that: Uf Uf =ω
U 1 + [THD (u )]2
We have: Q = U Ir = ) W = ω W
1 + [THD (u )]2
U )
Note that, unlike reactive energy W, REACTIVE POWER Q

IS NOT CONSERVATIVE. In fact, it depends on line
frequency and (local) voltage distortion.
Under sinusoidal conditions, the definition of Q
coincides with the conventional one 33
Void Power Terms
Void Power: V = U I v = S a2 + S r2 + Vg2
Scattered active power:

2
 Pk P  2
S a = U I sa = U 2
∑ −
U2 U2  k
U
k∈{K } k 
Scattered reactive power:

2
1 + THD u2 )2  Wk W  ) 2
S r = U I sr = ω U ∑  ) − )  Uk
 2 U2 
1 + THD u)2 k∈{K } U k 
Load
Load--generated harmonic power:
Vg = U I g
Skip examples 34
Application Examples
Example # 1
Voltage and Current : Resistive Load
1.5 1.5
1 1
u(t) u(t)
0.5
u P C C i P C C [p u ]
0.5
i(t) i(t)
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
0.1 0.11 0.12 0.13 0.14 0.15 0.16 0.1 0.11 0.12 0.13 0.14 0.15 0.16
T im e [s] T im e [s]
Sinusoidal voltage Non – sinusoidal voltage
Current = ipu(t)/2
35
Example # 1
Conservative Power Terms:
Terms: Resistive Load
p(t) = u(t)i(t) P = p = UIcosϕ p(t) = u(t)i(t) P = p
1 1
0.5 0.5
[p u ]
[p u ]
0 0
-0.5 -0.5
q(t) = ω û(t)i(t) Q = ω w(t) = UIsinϕ q(t) = ω û(t)i(t) Q
0.1 0.105 0.11 0.115 0.12 0.125 0.13 0.135 0.1 0.105 0.11 0.115 0.12 0.125 0.13 0.135
This example shows the correspondences between the

CPT theory and conventional theory
36
Example # 1 – Single-
Single-phase
Current Terms:
Terms: Resistive Load
1 1
0.5 0.5
i PCC [p u]
PCC
i PCC [p u]
0 0
-0.5 current -0.5
-1 -1
0.1 0.11 0.12 0.13 0.14 0.15 0.16 0.1 0.11 0.12 0.13 0.14 0.15 0.16
i PCC (t) = ia(t)
1 1
0.5 0.5
Active
i a [pu ]
i a [p u]
0 0
-0.5 current -0.5
-1 -1
0.1 0.11 0.12 0.13 0.14 0.15 0.16 0.1 0.11 0.12 0.13 0.14 0.15 0.16
0.5 0.5
0.25 Reactive 0.25
i r [pu ]
i r [pu ]
0 current 0
-0.25 -0.25
-0.5
i r (t)= 0 -0.5
0.1 0.11 0.12 0.13 0.14 0.15 0.16 0.1 0.11 0.12 0.13 0.14 0.15 0.16
0.5 0.5
0.25 Void 0.25
i v [pu ]
i v [p u]
0
current 0
-0.25 -0.25
-0.5
i v (t)= 0 -0.5
0.1 0.11 0.12 0.13 0.14 0.15 0.16 0.1 0.11 0.12 0.13 0.14 0.15 0.16

37
Example # 2
Voltage and Current : Ohmic-
Ohmic-inductive Load
1.5 1.5
1 1
u(t) i(t)
u(t) i(t) 0.5
0.5
u P C C iP C C [ p u ]
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.3 0.31 0.32 0.33 0.34 0.35 0.36
38
Example # 2
Conservative Power Terms
Terms:: Ohmic-
p(t) = u(t)i(t) P = p = UIcosϕ p(t) = u(t)i(t) P = p

1 1
0.5 0.5
[p u ]
[p u ]
0 0
q(t) = ω û(t)i(t) Q = ω w = UIsinϕ q(t) = ω û(t)i(t) Q

0.3 0.305 0.31 0.315 0.32 0.325 0.33 0.335 0.3 0.305 0.31 0.315 0.32 0.325 0.33 0.335
This example shows the correspondence between CPT

and conventional theory under sinusoidal conditions
39
Example # 2
Current Terms
Terms:: Ohmic-
1 1
0.5
i P C C [p u ]
0.5
PCC
i PC C [p u ]
0 0
-0.5 current -0.5
-1 -1
0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.3 0.31 0.32 0.33 0.34 0.35 0.36
1 1
0.5 0.5
Active
i a [p u ]
i a [p u ]
0 0
-0.5 current -0.5
-1 -1
0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.3 0.31 0.32 0.33 0.34 0.35 0.36
1 1
0.5
Reactive 0.5
i r [p u ]
i r [p u ]
0 0
-0.5 current -0.5
-1 -1
0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.5 0.5
0.25 i v (t)= 0 Void 0.25
i v [p u ]
i v [p u ]
0 0
-0.25 current -0.25
-0.5 -0.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.3 0.31 0.32 0.33 0.34 0.35 0.36

40
Physical meaning of void current
Example # 2
Void Current Terms:
Terms: Ohmic-
0.2
i v [pu] 0.1
0
-0.1
Void current
-0.2
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.2
0.1
i sa [pu]
0 Scattered active current

-0.1
-0.2
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.2
0.1
i sr [pu]
0
-0.1
-0.2
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.2
0.1
Load-generated
i g [pu]
0
-0.1
harmonic current
-0.2
0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s]
Non – sinusoidal voltage

41
Example # 3
Voltage and Current Conservative Power Terms
1.5
Distorting Load Distorting Load
Q=ωw P = p
1 1
u(t)
0.5
i(t)
[p u ]
0 0.5
-0.5
-1 0
q(t) = ω û(t)i(t) p(t) = u(t)i(t)
-1.5
0.3 0.305 0.31 0.315 0.32 0.325 0.33 0.335 0.3 0.305 0.31 0.315 0.32 0.325 0.33 0.335
T ime [s] T im e [s]
42
Void Current Terms:
Terms: Ohmic-
1
0.5
i PCC [pu ]
0 PCC current
-0.5
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
0.5
i a [pu]
0
Active current
-0.5
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
0.5
i r [pu ]
0
-0.5
Reactive current
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
0.5
Void current
i v [p u]
0
-0.5
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s]

43
Void Current Terms:
Terms: Ohmic-
0.8
0.4
i v [pu]
0
Void current
-0.4
-0.8
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.8
0.4
i sa [pu]
0
-0.4
-0.8
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.8
0.4
i sr [pu]
0
-0.4
-0.8
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.8
0.4
i g [pu]
0 Load-generated
-0.4
-0.8
harmonic current
0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s]

44
Seminar Outline
phase networks
conditions
45
Conservative
Power Theory
4. Extension to poly-phase domain:
3-wires / 4-wires
• Orthogonal current decomposition into active, reactive,

unbalance and void terms
• Physical meaning of current terms
• Active power decomposition into active, reactive,
unbalance and void terms
• Physical meaning of power terms
46
Orthogonal current decomposition
Extension to poly-
poly-phase: 3-
3-wires / 4
4--wires
In poly-phase systems, the current components (active,
reactive and void) can be defined for each phase:
Active current
u n , in Pn Pa = u , i a = P
ian = un = u n = Gn u n , n = 1 ÷ N
2
U n2 )
un Wa = u , i a = 0
Gn= equivalent phase conductance U I = u i ≠ P
a a
Reactive current
)
u n , in ) Wn ) ) Pr = u, i r = 0
irn = ) 2 u n = ) 2 u n = Bn u n , n = 1 ÷ N
)
un Un Wr = u , i r = W
) )
Bn = equivalent phase reactivity U Ir = u ir ≠ W
)
Void current Pv = u, i v = 0, Wr = u , i v = 0
ivn = in − ian − irn , n = 1 ÷ N U Iv > 0
47
Extension to poly-
poly-phase: 3-
3-wires / 4
4--wires
Active and reactive current terms can also be defined
collectively, i.e., by making reference to an equivalent
balanced load absorbing the same active power and
reactive energy of actual load:
Balanced Active currents: minimum collective
currents needed to convey active power P
u, i Gb= equivalent balanced conductance
P
i ba = u= u = Gb u
u
2
U 2 U I ab = u i ba = P, Qab = 0
Balanced Reactive currents: minimum collective

currents needed to convey reactive energy W
) Bb = equivalent balanced reactivity
u, i ) W ) b)
ir = ) 2 u = ) 2 u = B u
b ) b ) b
u U U I r = u i r = W , Prb = 0
48
Extension to poly-
poly-phase: 3-
3-wires / 4
4--wires
Unbalanced currents account for the asymmetrical
behavior of the various phases
Unbalanced Active currents

i ua = i a − i ba ⇒ ian
u
( )
= Gn − G b u n , n = 1 ÷ N Pau = Pa − Pab = 0
Wau = 0
Unbalanced Reactive currents
i ur = i r − i br ⇒ irn
u
( )
)
= Bn − B b u n , n = 1 ÷ N Pru = 0
Wru = Wr − Wrb = 0
49
Extension to poly-
poly-phase: 3-
3-wires / 4
4--wires
Void currents: as for single-phase systems, they
reflect the presence of scattered active, scattered
reactive and generated terms.
iv = i − ia − ir = i + i + i g
s s
a r
Scattered current terms: Load-generated harmonic

Account for different values of current:
equivalent admittance at Harmonic terms that exist in
different harmonics currents only, not in voltages
Pv = P − Pa − Pr = 0
Wv = W − Wa − Wr = 0
50
Extension to poly-
poly-phase: 3-
3-wires / 4
4--wires
Summary of current decomposition
i = ia + ir + iv = i + i + i + i + i + i + i g
b u b u s s
a a r r a r
i a active currents
• i ab balanced active currents
• i au unbalanced active currents
i r reactive currents
• i rb balanced reactive currents
• i ru unbalanced reactive currents
i v void currents
• i as scattered active currents
• i rs scattered reactive currents
• i g load-generated harmonic currents
51
Extension to poly-
poly-phase: 3-
3-wires / 4
4--wires
i = ia + ir + iv = i + i + i + i + i + i + i g
b u b u s s
a a r r a r
123 123 14243
ia ir iv
Each current component has a precise

PHISICAL MEANING and is computed in the time domain
Moreover, all current terms defined in the above equation

are ORTHOGONAL
ORTHOGONAL,, thus:
b 2 u 2 b 2 u 2 s 2 s 2 2
i = ia + ir + iv = i + i + i + i + i + i + ig
2 2 2 2
a a r r a r
52
Apparent power decomposition in
poly--phase: 3-
poly 3-wires / 4-
4-wires
A = U I = u i = P2 + Q2 + N 2 +V 2
Active power: P = U I ab = u i ba
Reactive power: Q = U I rb = u i br
power:: N = U I u = u i u = N a2 + N r2
Unbalance power
Void power: V = U I v = u i v = S a2 + S r2 + Vg2
53
Unbalance Power Terms
Unbalance power
power:: N= 2
Na + 2
Nr
Unbalance Active Power

N
Pn P2
N a = U I au = u i ua = U ∑U 2 − U2
n =1 n
Unbalance Reactive Power
1 + [THD (u )]2 N
Wn W 2
N r = U I ru = ω U ∑ U) 2 − U) 2
1 + [THD (u )]
) 2
n =1 n
Unbalance active and reactive power vanish

if the load is balanced
Skip examples 54
Example # 1 : 3
3--phase 3-
3-wire – Balanced load
(Resistive)

1.5
ua ub uc 1
p (t) & P ; w (t) & W [p u ]

1
ia ib
[p u ]
ia
0.5 0.5
PCC
0
& i
0
PCC
-0.5
u
-1
-0.5
-1.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.3 0.31 0.32 0.33 0.34 0.35 0.36
Sinusoidal voltage
Current = ipu(t)/2
55
Example # 1 : 3
3--phase 3-
3-wire – Balanced load
(Resistive)
1
[p u ]
0.5
0 Balanced active currents
an
-0.5
b
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
i r (t)= 0
[p u ]
0.5
0
Balanced reactive currents
r n
-0.5
b
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
0.5
i ua (t)= 0
[p u ]
0 Unbalanced active currents

an
-0.5
u
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
i ur (t)= 0
[p u ]
0.5
0 Unbalanced reactive currents
r n
-0.5
u
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
0.5 i v (t)= 0
[p u ]
0 Void currents
vn
-0.5
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s]
56
Example # 1 : 3
3--phase 3-
3-wire – Unbalanced load
(resistor connected between two phases)

1.5
1
ua ub uc 1
p (t) & P ; w (t) & W [p u ]

[p u ]
ia ia
0.5
0.5
PCC
ib
0
& i
-0.5 0
u PCC
-1
-0.5
-1.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s]
T im e [s]
Sinusoidal voltage
Current = ipu(t)/2
57
Example # 1 : 3
3--phase 3-
3-wire – Unbalanced load
(resistor connected between two phases)
1
0.5
[p u ]

an
-0.5
b
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
[p u ]
0.5 i r (t)= 0
0
r n
-0.5
b
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
0.5
[p u ]

an
-0.5
u
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
[p u ]
0.5
r n
-0.5
u
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
1
0.5
i v (t)= 0
[p u ]
0 Void currents
vn
-0.5
i
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s]
58
Example # 3 : 3
3--phase 3-
3-wire
Three--phase RL + Single-
Three Single-phase R load
1.5
1
ua ua uc 1
p ( t) & P ; w (t ) & W [p u ]
ia
[p u ]
0.5
ia
ib
PCC
0.5
0
& i
-0.5
PCC
0
u
-1
-1.5 -0.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s] T ime [s]
Symmetrical non-
non-sinusoidal voltage
59
Example # 3 : 3
3--phase 3-
3-wire
1
Three--phase RL + Single-
Three Single-phase R load
0.5
[p u ]

an
-0.5
bi
-1
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.5
[p u ]
0.25
0
r n
-0.25
b
i
-0.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.5
0.25
[p u ]

an
-0.25
u
i
-0.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.5
[p u ]
0.25
r n
-0.25
u
i
-0.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36
0.5
0.25
[p u ]
0 Void currents
vn
-0.25
i
-0.5
0.3 0.31 0.32 0.33 0.34 0.35 0.36
T im e [s]
Symmetrical non-
non-sinusoidal voltage
60
Sharing of compensation duties
Orthogonal current terms
terms::
i = ia + ir + iv = ia + ia
b u
+ i r + i r + i sa
b u
+ i sr + i g
123 123 14243
ia ir iv
Each current component has

a precise PHYSICAL MEANING
Balanced Active currents convey active power P

Balanced Reactive currents convey reactive power Q
Unbalanced Active and Reactive currents account for
asymmetrical behavior of the various phases
Void currents reflect the presence of different behavior at
different frequencies and/or generated current harmonics 61
Sharing of compensation duties
i = ia + ir + iv = i + i + i + i + iv
b
a
b
r
u
a
u
r
Unbalance
Reactive compensation Harmonic
compensation requires controllable compensation
reactances (extended requires high-
Steinmetz approach) frequency response
Stationary Compensators
(reactive impedances) Quasi-
& Stactionary Passive filters &
Quasi-Stationary Compensators Switching Power
Compensators (SVC, (SVC) Compensators
Static VAR Compensators) (SPC=APF+SPI) 62
Effect of compensation on
power terms
Active power (balanced):
(balanced): P = U I b
a
Compensation
) b
(balanced):: Q = U I r
Reactive power (balanced) SVC
→ 0
u2 u2 SVC
→ 0
Unbalance power:
power: N = UI = U
u
Ia + Ir
Void power:
power: V = U Iv SPC
→ 0
APPARENT POWER
A = U I = P2 + Q
X2 + N
X2 +V
X2 SVC
 →
,SPC
A = U I ba = P
63
Seminar Outline
phase networks
conditions
64
Conservative
Power Theory
5. Sequence components under non-
sinusoidal conditions
1. Problem statement
2. Goal of decomposition
3. Derivation of generalized symmetrical components in
the time domain (extension of Fortescue’s approach)
4. Analysis of generalized symmetrical components in
the frequency domain
5. Orthogonality of sequence components
6. Application examples
Skip 65
1. Problem statement (1)
Symmetrical components are very useful to

simplify the analysis of three-phase networks
under sinusoidal conditions
It is important to extend the definition and

application of symmetrical components to
non-sinusoidal periodic operation
66
1. Problem statement (2)
Given periodic three-phase variables fa(t), fb(t), fc(t) of
period T we define the following symmetry
properties:
Homopolarity (zero symmetry)
f a (t ) = f b (t ) = f c (t )
Positive (direct) symmetry

 T  2T 
f a (t ) = f b  t +  = f c  t + 
 3  3 
Negative (inverse) symmetry
 T  2T 
f a (t ) = f b  t −  = f c  t − 
 3  3 
67
2. Goal of decomposition
Given a set of generic three-phase variables:
f a (t )
f = f b (t )
f c (t )
we decompose them in the orthogonal form:
f = fz+fh= fz+f p
+fn+ fr
where:
f z are zero-sequence (homopolar) components
f h are non-zero sequence (heteropolar) components
f p are positive-sequence components
f n are negative-sequence components
f r are residual components
68
3. Derivation of generalized
symmetrical components
Zero sequence (homopolar) component
1
f a (t ) + f b (t ) + f c (t )
f = f
z z
(t ) 1 f (t ) =
z
3
1
Heteropolar components
f ah (t ) f a (t ) − f z (t )
fh = f − f z = f bh (t ) = f b (t ) − f z (t )
f ch (t ) f c (t ) − f z (t )
69
Positive sequence component
1 h  T  2T 
f (t ) =  f a (t ) + f bh  t +  + f ch  t +
p

3  3  3 
p T p 2T 
f ap (t ) = f p
(t ), f bp (t ) = f  t − , fc p
(t ) = f t − 
 3  3 
Negative sequence component
1 h  T  2T 
f (t ) =  f a (t ) + f bh  t −  + f ch  t −
n

3  3  3 
n T n 2T 
f an (t ) = f n
(t ), f bn (t ) = f  t + , f cn (t ) = f  t + 
 3  3 
70
Residual components
 T  2T 
f ah (t ) + f ah  t −  + f ah  t − 
 3  3 
3
f ar (t )  T  2T 
f bh (t ) + f bh  t −  + f bh  t − 
f r = f br (t ) =  3  3 
3
f cr (t )  T  2T 
f ch (t ) + f ch  t −  + f ch  t − 
 3   3 
3
Note: these components are computed independently

for each phase and vanish in sinusoidal operation
71
Resulting decomposition
f z
(t ) + f p (t ) + f n (t ) + f ar (t )
f a (t )
f = f b (t ) = f z
(t ) + f p  t − T  + f n  t + T  + f br (t )
 3  3
f c (t )
 2T  n 2T 
f z (t ) + f p  t −  + f t +  + f c (t )
r
 3   3 
72
4. Analysis in the frequency domain
Expressing variables fa(t), fb(t), fc(t) in Fourier series:
∞ ∞
f a (t ) = ∑ f ak (t ) = ∑ 2 Fak sin (kω t + α ak )
k =1 k =1
∞ ∞
f b (t ) = ∑ f bk (t ) = ∑ 2 Fbk sin (kω t + α bk )
k =1 k =1
∞ ∞
f c (t ) = ∑ f ck (t ) = ∑ 2 Fck sin (kω t + α ck )
k =1 k =1
we can determine, for each harmonic, the zero, positive,

and negative components f zk(t), f pk(t), f nk(t). Instead,
residual harmonic components are zero because
harmonic quantities are sinusoidal.
73
4. Analysis in the frequency domain
Contribution of harmonic sequence components to
generalized sequence components
Harmonic order : k = 3m + 1 ∀m ∈ [0, ∞ ]

f kp ⇒ f p , f kn ⇒ f n , f kz ⇒ f z
Harmonic order : k = 3m + 2 ∀m ∈ [0, ∞ ]

f kp ⇒ f n , f kn ⇒ f p , f kz ⇒ f z
Harmonic order : k = 3m ∀m ∈ [0, ∞]

f kp ⇒ f r , f kn ⇒ f r , f kz ⇒ f z
74
5. Orthogonality of components
Given two sets of three-phase quantities f and g,
their sequence components obey the following
general rules:
Scalar product
f z o gh = f z o g p = f z o gn = f z o gr = 0
Internal product
f p, gn = f p, gr = f n, gr = 0
f , g = f z, g z + f h, gh = f z, gz + f p, g p + f n, gn + f r, gr
Norm
2 2 2 2 2 2 2
f = f z
+ f h
= f z
+ f p
+ f n
+ f r
75
6. Application Example
1 1 1 1
va ( t ) = sin( ωt ) + sin( 3ωt ) + sin( 5ωt ) + sin( ωt ) + ;
3 5 10 10
2π 1 2π 1 2π 1 1
vb ( t ) = sin( ωt − ) + sin( 3ωt + ) + sin( 5ωt − ) + sin( ωt ) + ;
3 3 3 5 3 10 10
2π 1 2π 1 2π 1 1
vc ( t ) = sin( ωt + ) + sin( 3ωt − ) + sin( 5ωt + ) + sin( ωt ) + ;
3 3 3 5 3 10 10
Line to Neutral Voltages
1.5
0.5
0
V
-0.5
-1
-1.5
0 2 4 6 8 10 12 14 16 18 20 76
ms
1 1 1 1
3 5 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
Generalized Zero Sequence Voltages
1.5
0.5
0
V
-0.5
-1
-1.5
0 2 4 6 8 10 12 14 16 18 20 77
ms
1 1 1 1
3 5 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
Generalized Direct Sequence Voltages
1.5
0.5
0
V
-0.5
-1
-1.5
0 2 4 6 8 10 12 14 16 18 20
78
ms
1 1 1 1
3 5 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
Generalized Inverse Sequence Voltages
1.5
0.5
0
V
-0.5
-1
-1.5
0 2 4 6 8 10 12 14 16 18 20 79
ms
1 1 1 1
3 5 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
2π 1 2π 1 2π 1 1
3 3 3 5 3 10 10
Residual Voltages
1.5
0.5
0
V
-0.5
-1
-1.5
0 2 4 6 8 10 12 14 16 18 20
80
ms
9. Summary - 1
An extension of the sequence components in case of
non-sinusoidal periodic operation has been proposed.
It has been shown that three-phase currents (or
voltages) cannot always be derived from generalized
positive sequence, generalized negative sequence and
generalized zero-sequence components. A residual
component may be required.
To compute the generalized positive and negative
sequence components, the zero-sequence components
should first be subtracted from the phase quantities, in
contrast with the sinusoidal case where this is not
necessary.
In the sinusoidal case the residual component is absent
and the other components reduce to the classical
symmetrical components.
81
9. Summary - 2
The generalized positive sequence, negative sequence,
and zero-sequence components have complete phase
symmetry. This implies that the three-phase analysis can
be reduced to a single-phase analysis.
The residual components do not have the same
symmetry, and the corresponding three-phase analysis
cannot be reduced to single-phase analysis. It
corresponds to a periodic time function in each of the
three phases with a period which is 1/3 of the line period;
this simplifies the analysis because only 1/3 of the period
must be studied.
82
1.5
1
Voltage
0.5
phase a
0
V
-0.5
-1
-1.5
0 2 4 6 8 10 12 14 16 18 20
ms
1.5
Voltage
Derivation of 1
phase b
homopolar 0.5
component 0
Homopolar
V
-0.5
component
-1
-1.5
0 2 4 6 8 10 12 14 16 18 20
ms
1.5
1
Voltage
phase c
0.5
0
V
-0.5
-1
-1.5
0 2 4 6 8 10
ms
12 14 16 18 20 83
1.5 1.5
Voltage Opposite of the

1 1
phase a homopolar
0.5 0.5
component
0 0
V
-0.5 -0.5
-1 -1
-1.5 -1.5
1.5
0 2 4 6 8
ms
10 12 14 16 18 20
1.5
0 2 4 6 8 ms
10 12 14 16 18 20
1
Opposite of the
1
Derivation of homopolar
heteropolar 0.5 0.5
component
component 0 0
V
V
-0.5 -0.5
-1
Voltage -1
-1.5
phase b -1.5
0 2 4 6 8 ms
10 12 14 16 18 20
0 2 4 6 8
ms
10 12 14 16 18 20
1.5 1.5
Opposite of the
1 1
homopolar
0.5 0.5
component
0 0
V
V
-0.5 -0.5
-1
Voltage -1
.
phase c
-1.5 -1.5
0 2 4 6 8
ms
10 12 14 16 18 20
84
0 2 4 6 8
ms
10 12 14 16 18 20
1.5
1.5
1
1
0.5
0.5
0
0
V
-0.5
-0.5
Voltage -1
-1
phase a -1.5
-1.5 0 2 4 6 8 10 12 14 16 18 20
0 10 20 30
ms
40 50 60
ms
1.5 1.5
Derivation of positive 1 1
sequence component 0.5 0.5
(phase a) V 0 0
V
-0.5 -0.5
Voltage -1 -1 Positive
phase b -1.5 -1.5
sequence component
0 10 20 30 40 50 60 0 2 4 6 8 10 12 14 16 18 20
1.5
ms 1.5 ms
1 1
0.5 0.5
0 0
V
V
-0.5 -0.5
Voltage -1 -1
phase c -1.5
0 10 20 30
ms
40 50 60
-1.5
0 2 4 6 8 10
ms
12 14 16 18 20 85
1.5
1.5
1
1
0.5
0.5
0
0
V
V
-0.5
-0.5
Voltage -1
-1
phase a -1.5
-1.5 0 2 4 6 8 10 12 14 16 18 20
0 10 20 30 40 50 60
ms ms
1.5 1.5
Derivation of negative 1 1
sequence component 0.5 0.5
(phase a) V 0 0
V
-0.5 -0.5
Voltage -1 -1 Negative
phase b sequence component
-1.5 -1.5
0 10 20 30 40 50 60 0 2 4 6 8 10 12 14 16 18 20
1.5
ms 1.5
ms
1
1
0.5
0.5
0
0
V
V
-0.5
-0.5
Voltage -1
-1
phase c -1.5
-1.5
0 10 20 30 40 50 60
0 2 4 6 8 10
ms
12 14 16 18 20
86
ms
1.5 1.5
Opposite of the
1 1
homopolar comp.
0.5 0.5
0 0
V
-0.5 -0.5
-1 -1
-1.5 -1.5
0 2 4 6 8 10 12 14
14 16 18
18 20
20 0 2 4 6 8 10 12 14 16 18 20
ms 1.5
ms
1.5
1
Opposite of the positive
1
Derivation of residual sequence comp.

0.5 0.5
component (phase a)
V 0 0
V
-0.5 -0.5
-1 -1
-1.5 -1.5
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
ms ms
1.5
1
Opposite of the negative
sequence comp.
0.5
0
V
V
-0.5
Residual component
-1
ms
-1.5
0 2 4 6 8 10 12 14 16 18 20
87
ms
Seminar Outline
3. Instantaneous and average power & energy terms in
poly-phase networks
conditions
7. Measurement & accountability issues (basic
approach)
88
Measurement & accountability issues
Active and reactive current (and power) terms are
affected by the presence of negative-sequence, zero-
sequence and harmonic voltages
A proper accountability approach must be adopted to
depurate the power and current terms from the effects
of voltage non-idealities, which are not under load
responsibility
If we assume that the supply voltages are sinusoidal

and symmetrical with positive sequence, we have:
U n = U fp , n = 1 ÷ 3 ⇒ U = 3 U fp
) U fp ) 3 U fp
Un = , n =1÷ 3 ⇒ U =
ω ω
89
Phase current and power terms
Assuming that the equivalent phase resistance
remains the same irrespective of supply conditions, we
can express the active current and power accountable
to the load in each phase as:
p2 3
U 1
ialn = Gn u ⇒ Pln = u , ialn = Pn
p
fn
p
fn
U
f
2
I al = p
Uf
∑ ln
P 2
n =1
=
n
Similarly, if the equivalent phase reactivity remains the

same irrespective of supply conditions, we can
express the reactive current and power accountable to
the load in each phase are:
) p2 3
Uf 1
irln
)p )p
= Bnu fn ⇒ Wln = u fn , irln = Wn ) 2
Un
I rl = p
Uf
∑ ln
Q 2
n =1
90
Balanced current and power terms
The total power terms accountable to the load are:
3 3
Pl Wl Ql
Pl = ∑ Pln ⇒ G = b
l Wl = ∑ Wln ⇒ B = ) 2 = ω
b
l
p2 p p2
n =1 3U f n =1 3U f 3U f
The balanced current terms accountable to the load

are:
1 Pl b)p 1 Wl 1 Ql
i =G u ⇒ I = i rl = Bl u fn ⇒ I rl = )p =
b b p b b b
al l f pal p
3Uf 3Uf 3Uf
91
Unbalanced current and power terms
The unbalanced active current and power
accountable to the load are :
(
iauln = ialn − iabln = Gn − Glb u fp n )
∑ (G )
3 3 2
1 P
I aul =
n =1
n
b 2
−G U f
l
p2
= p
Uf
∑
n =1
P 2
ln −
3
l
The unbalanced reactive current and power

accountable to the load are :
u
i
rln = irln − i b
rln ( )p
= Bn − B u fn
b
l )
∑ (B )
3 ) p2 3 2
1 Q
I rul =
n =1
n −Bb 2
l Uf = p
Uf
∑
n =1
Q 2
ln −
3
l
92
Void current and power
The void currents satisfy the condition:
u , i v = 0 ⇒ u pf , i v + u nf + u zf + u h , i v = 0
) )p )n ) z )
u,iv = 0 ⇒ u f ,iv + u f + u f + u h ,iv = 0
The void current terms which can be accounted

to the load are therefore given by:
p )p
u , iv
f u f , iv ) p
i vl = i v − uf − ) 2 uf
p
p2
3U f 3U fp
In fact, the fundamental component of the void
current has been already accounted for in the
active and reactive current terms. 93
Currents terms accountable to the load
i l = i a l + i r l + i v l = i ba l + i brl + i ua l + i url + i v l
All current terms are orthogonal, thus:
I l = I ba 2l + I br l2 + I ua l2 + I ur l2 + I v2l
Apparent power accountable to the load

Al = U fp I l = Pl2 + Ql2 + N a2l + N r2l + Vl2
14243
N l2
94
Application Examples: 3-
3-phase 3
3--wire
Case I: Symmetrical sinusoidal voltages
Case II: Asymmetrical sinusoidal voltages
Case III: Symmetrical non-
non-sinusoidal voltages
Case IV: Asymmetrical non-
Case I Case II
∠0º Vrms
U1 = 127∠ ∠0º Vrms
U1 = 127∠
∠-120º Vrms U2 = 113∠
U2 = 127∠ ∠-104,4º Vrms
∠120º Vrms U3 = 147,49∠
U3 = 127∠ ∠144º Vrms
cases III and IV are the same of
Balanced Load cases I and II with the addition
SOURCE LINE PCC LOAD of 10% of 5th and 7th harmonics
3
i3
u32 L3
2
R C
uSn L2
u21
1
i1
RLn LLn L1
n {1,2,3}
Distorting Load
Ω and LL0 = LL1= LL2= LL3= 10 µH.
The line parameters are : RL1= RL2= RL3= 1mΩ
95
3-phase 3
3--wire
Example # 1: Balanced Load
CASE I CASE II CASE III CASE IV

PCC LOAD PCC LOAD PCC LOAD PCC LOAD
A 1,0000 1,0000 1,0000 0,9634 1,0000 0,9840 1,0000 0,9538
P 0,7985 0,7985 0,7985 0,7693 0,7913 0,7758 0,7945 0,7556
Q 0,6020 0,6020 0,6020 0,5800 0,6023 0,5962 0,6022 0,5770
N 0,0000 0,0000 0,0000 0,0000 0,0002 0,0002 0,0056 0,0061
V 0,0000 0,0000 0,0000 0,0000 0,1054 0,1038 0,0782 0,0760
λ 0,7985 0,7985 0,7985 0,7985 0,7913 0,7885 0,7945 0,7922
The load is accounted for less active, reactive and void

power than the PCC
96
3-phase 3
3--wire
Example # 2: Distorting load
load
CASE I CASE II CASE III CASE IV

PCC LOAD PCC LOAD PCC LOAD PCC LOAD
A 1,0000 1,0009 1,0000 0,9050 1,0000 0,9832 1,0000 0,8973
P 0,8275 0,8280 0,8841 0,7668 0,8254 0,8098 0,8808 0,7570
Q 0,2432 0,2451 0,2135 0,2245 0,2422 0,2417 0,2135 0,2232
N 0,5060 0,5060 0,4156 0,4250 0,5055 0,4980 0,4185 0,4231
V 0,0000 0,0000 0,0000 0,0000 0,0674 0,0666 0,0581 0,0566
λ 0,8275 0,8273 0,8841 0,8473 0,8254 0,8237 0,8808 0,8437
Again the apparent and active power accounted to the

load are always lower than those computed at PCC due
to the depuration of the effects of voltage asymmetry
and distortion
97
Defects of proposed accountability
approach
The equivalent phase conductance and reactivity are
computed by considering the effect of fundamental and
harmonic supply voltages as a whole.
whole. In practice, the
load can have different response to fundamental and
harmonic voltages.
The void currents are not orthogonal to supply voltages
if these latter become sinusoidal.
Only the load impedance is modeled, while supply
impedance is not estimated and enters the computation
process indirectly, in a way that does not allow to fully
analyze its effect.
98
Seminar Outline
3. Instantaneous and average power & energy terms in
poly-phase networks
conditions
7. Measurement & accountability issues (extended
approach)
99
Extended approach to accountability
Both load and supply are modeled based on
measurements made at PCC
Load modeling is done under sinusoidal conditions
o This makes the load model more reliable, since harmonic effects
are depurated
o Moreover, the harmonic currents generated by the load are
represented separately and their effect can directly be accounted
for accountability purposes
Supply modeling is made for three-phase symmetrical
systems and allows estimation of no-load supply
voltages and line impedances
The extended accountability approach is more reliable
than the basic one, and possibly avoids under- and over-
penalization of the loads.
Of course, better results can be achieved if a more
accurate modeling of the load is available. 100
Load modeling (3-
(3-phase 4-
4-wire) - 1
Single-phase equivalent
circuit of 3-phase load seen
from PCC
The passive parameters of the equivalent circuit are computed to suit

the circuit performance at fundamental frequency, i.e.:
 f U mf 2 U mf 2
 Pm = u m ,im = Rm =
f f
⇒
u mf û mf  Rm Pmf
imf = + ⇒  f2
Rm Lm W f = û f ,i f = Û m Û mf 2
⇒ Lm =
 m m m
Lm Wmf

With this assumption current jm is purely harmonic. In fact:
) ) f )h )h
um um u mf+ u mh um + um u mh um
j m = im − − = imf + imh − − = imh − −
Rmf Lmf Rmf Lmf Rmf Lmf
101
Load modeling (3-
(3-phase 4-
4-wire) - 2
from PCC
For the validity of the model we must assume that the

equivalent circuit parameters remain the same within
reasonable variations of the voltage supply, both in terms
of asymmetry and distortion.
This is only approximately true in real networks, but it
makes possible an accountability approach based on
measurement at the load terminals, without requiring a
precise knowledge of the load itself.
102
Supply modeling (3-
(3-phase 4-
4-wire) - 1
Single-phase equivalent circuit

of 3-phase supply seen from
PCC
The passive parameters of the equivalent circuit are the same for all
phases, due to supply lines symmetry. The circuit equations are:
 p di p
e = u p
+ R i p
+ LS
di  S
e = u + R S i + LS ⇒  dt
dt  en = un + R in + L d in


S S
dt
Due to its linearity, this equation can be applied separately to the
fundamental positive-sequence voltage and current terms (index p)
and the remaining terms (index n), which represent the unwanted
current and voltage components. 103
Supply modeling (3-
(3-phase 4-
4-wire) - 2

PCC
The passive parameters of the equivalent circuit are selected so as to

minimize the unwanted components of the supply voltages en, i.e., the
function:
2 M
ϕ = en = ∑ emn , emn =
m =1
2
n
2 2 di d in
= u n
+ RS2 i n
+ L2S + 2 RS u , i
n n
+ 2 LS n
u , LS
dt dt
104
Supply modeling (3-
(3-phase 4-
4-wire) - 3

PCC
The result is expressed as a function of the quantities measured at

PCC in the form:
∂ϕ 2
=0 ⇒ RS = − u , i
n n
i n
∂ RS
2
∂ϕ di n
di n
=0 ⇒ LS = − u n ,
∂ LS dt dt
Obviously, only positive solutions are acceptable for RS and LS. In

case of negative solution, the corresponding parameter is set to
zero. 105
Supply modeling (3-
(3-phase 4-
4-wire) - 4

PCC
Given RS and LS, we may compute the positive-sequence supply

voltages to be included in the equivalent circuit :
di p
e = u + R S i + LS
p p p
dt
Note that ep, up, ip are the fundamental positive sequence

components of the related voltages and currents, while en, un, in are
calculated by difference from the original voltages and currents.
106
Accountability – Procedure (1)
from PCC
1. From the voltages and currents measured at PCC we estimate the

phase parameters Rm and Lm and the current source jm of the
equivalent circuit.
U mf 2 Û mf 2
Rm = Lm =
Pmf Wmf
)h
u mh um
jm = imh − −
Rmf Lmf
107

PCC
2. From the voltages and currents measured at PCC we estimate the

supply line parameters RS and LS and the fundamental positive-
sequence supply voltages ep
2
n n
2 di di
RS = − u n , i n in LS = − u n ,
dt dt
di p
e = u + R S i + LS
p p p
dt
108
Equivalent circuit for the

computation of fundamental
voltages at PCC
4. Applying now the positive-sequence supply voltages at the input

terminals of the equivalent circuit, we may determine the
f
fundamental phase currents i l absorbed by the load under these
supply conditions and the corresponding fundamental phase
f
voltages u l appearing at the PCC terminals.
Note that the currents and voltages at PCC may result
asymmetrical due to load unbalance. This non-ideality must
obviously be ascribed to the load, since the voltage supply and
distribution lines are symmetrical
109
5. Finally, the load voltages and currents at PCC, which are
accountable to the load are given by:
u l = u lf + u lh
i l = i lf + i lh
We can now compute all power terms accountable to the load and
the corresponding performance factors.
)
Plm = u lm ,ilm Wlm = u lm ,ilm
Glm , Blm , I la , I lr
M M
Pl = ∑ Plm Wl = ∑ Wlm Glb , Blb , I bla , I blr
m =1 m =1
⇒ u u
I l a , I lr
N la = K N lr = K
I lv
Vl
110
Performance factors
V2 P2 + Q2 + N 2
Distortion factor: λ D = 1 − 2
=
A A2
N2 P2 + Q2
Unbalance factor: λ N = 1 − =
P +Q + N
2 2 2
P2 + Q2 + N 2
Q2
Reactivity factor: λ Q = 1 −
P2 + Q2
P P
Power factor: λ= = = λQ λ N λ D
A P 2 + Q 2 + N a2 + N r2 + D 2
111
3-phase 4
4--wire
Load circuits
A. Unbalanced
linear load
B. Unbalanced
nonlinear load
Ω - LL1= LL2= LL3= 38.5 µH

Line parameters: RL1= RL2= RL3= 10.9 mΩ 112
3-phase 4
4--wire
Supply conditions
Case 1: Symmetrical sinusoidal voltages
Case 2: Symmetrical non-
Case 1 Case 2
1 = 127∠0 Vrms 1 =
+ ∑ 1 Vrms
2 = 127∠ − 120 Vrms 2 =
+ ∑ 2 Vrms
3 = 127∠120 Vrms 3 =
+ ∑ 3 Vrms
In case 2 the terms called ΣH represent the harmonic contents of the

phase voltages.
Each phase voltage includes 2% of 3rd harmonic, 2% of 5th harmonic.
The phase angle of each harmonic term is the phase angle of the
fundamental voltage (as in Case 1) multiplied by the harmonic order.
113
3-phase 3
3--wire
Case A:
A: Unbalanced
Unbalanced linear load
Case A.1 Case A.2

PCC Load PCC Load
A [KVA] 95.522 98.722 95.263 96.571
P [KW] 65.224 67.862 65.231 65.089
Q [KVA] 67.698 69.813 67.729 69.597
N [KVA] 15.158 16.334 15.163 15.582
D [KVA] 0.017 0.074 1.627 1.649
λ 0.6828 0.6874 0.6847 0.6740
λQ 0.6938 0.6970 0.6937 0.6831
λN 0.9872 0.9862 0.9872 0.9869
λD 1.0000 0.9999 0.9999 0.9999
The load is penalized for its unbalance, especially in case A.1

(sinusoidal and symmetrical supply voltages)
114
3-phase 3
3--wire
Case B:
B: Unbalanced
Unbalanced nonlinear load
Case B.1 Case B.2
PCC Load PCC Load
A [KVA] 93.267 94.007 89.494 91.207
P [KW] 63.909 63.334 62.738 62.763
Q [KVA] 33.274 35.376 33.599 36.159
N [KVA] 20.873 21.143 20.619 21.296
D [KVA] 55.421 55.916 50.190 51.171
λ 0.6852 0.6737 0.7010 0.6881
λQ 0.8870 0.8730 0.8815 0.8665
λN 0.9605 0.9601 0.9605 0.9594
λD 0.8043 0.8039 0.8279 0.8278
The apparent, reactive and unbalance power

accounted to the load are higher than those computed
at PCC
115

Conservative Power Theory - Seminario PDF

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Conservative Power Theory - Seminario PDF

Enviado por

Direitos autorais:

Formatos disponíveis

UNICAMP – UNESP Sorocaba

No one of these theories was able to target all goals (characterization

• Norm (rms value) X = x = x, x

• The vector norm is also called collective rms value

The above operators have the following properties:

• Both quantities do not depend on the voltage reference

The equal sign is u ∝ i ⇒ u, i = u i ⇒ λ =1

We select the voltage reference so as to ensure unity

Thus, the voltage reference must be selected to comply

This choice minimizes the norm of the voltage vector

Derivation of phase voltages

Unity power factor may occur only if the neutral current is

• ia active current • isa scattered active current

Active current: the minimum current (i.e., with minimum

Pa = u, ia = Ge u, u = Ge U 2 = P Active current conveys full

Reactive current: the minimum current needed to

) Active and reactive current

Void current: is the remaining current component

Void current is not conveying active power or reactive energy

Void current is orthogonal to active and reactive terms

The void current reflects the presence of scattered active,

Scattered current terms: Load-generated current

Scattered and load-generated

Scattered active current

Total harmonic reactive current

Scattered reactive current

Note that, unlike reactive energy W, REACTIVE POWER Q

Void Power: V = U I v = S a2 + S r2 + Vg2

Scattered active power:

Scattered reactive power:

Sinusoidal voltage Non – sinusoidal voltage

Sinusoidal voltage Non – sinusoidal voltage

This example shows the correspondences between the

Sinusoidal voltage Non – sinusoidal voltage

Sinusoidal voltage Non – sinusoidal voltage

p(t) = u(t)i(t) P = p = UIcosϕ p(t) = u(t)i(t) P = p

q(t) = ω û(t)i(t) Q = ω w = UIsinϕ q(t) = ω û(t)i(t) Q

This example shows the correspondence between CPT

Sinusoidal voltage Non – sinusoidal voltage

0 Scattered active current

Non – sinusoidal voltage

Non – sinusoidal voltage

Non – sinusoidal voltage

Non – sinusoidal voltage

• Orthogonal current decomposition into active, reactive,

Balanced Reactive currents: minimum collective

Unbalanced Active currents

Scattered current terms: Load-generated harmonic

Each current component has a precise

Moreover, all current terms defined in the above equation

Void power: V = U I v = u i v = S a2 + S r2 + Vg2

Unbalance Active Power

Unbalance Reactive Power

Unbalance active and reactive power vanish

Voltage and Current Conservative Power Terms

p (t) & P ; w (t) & W [p u ]

0 Unbalanced active currents

Voltage and Current Conservative Power Terms

p (t) & P ; w (t) & W [p u ]

0 Balanced active currents

0 Unbalanced active currents

0 Balanced active currents