Topic 2 Power Flow

Topic 2
Power Flow
EE 605 Power Network Analysis
Dr. Muhyaddin Jamal Rawa

Contents
1. Introduction.
2. Bus Admittance Matrix.
3. Power Flow Solution.
4. Gauss-Seidel Power Flow Solution.
5. Line Flows and Losses.
6. Tap Changing Transformers.
7. Newton-Raphson Power Flow Solution.
8. Fast Decoupled Power Flow Solution.
10/01/2019 Dr. Muhyaddin Jamal Rawa 2
1. Introduction
• This chapter deals with steady-state analysis
of interconnected power systems during
normal operating conditions.
• The three-phase system is assumed to be
balanced, so it can be represented and
analyzed by a single-phase network.
• The node-voltage method is the basis of load
flow analysis discussed in this section.

• If node currents are known, a set of linear
equations can be solved for the node voltages.
• However, in power systems, powers are
known rather than currents. Therefore, the set
of power flow equations become nonlinear
and must be solved in iterative techniques.

• Power flow studies, also knowns as load flow,
are the backbone of power system analysis
and design as they are necessary for
 Planning, operation, economic scheduling and
control of an existing system.
 Planning for future expansion.
 Exchange of power between utilities.
 Conducting transient stability and contingency
analysis.

2. Bus Admittance Matrix
• Consider a simple power system shown in the
figure.
• The impedances are
expressed in a common
MVA base.
• Resistances are neglected.

• As node-voltage method is used, the
impedances are converted to admittances.
• Voltage sources are also transformed to
current sources.

• Node 0 is the reference node (usually the
ground).
• Applying KCL to each node results in

• Rearranging the equations to get
• By introducing the following admittances

• The node equations reduce to
• As there is no connection between bus 1 and

bus 4 in the above network
• Similarly

• Extending the above relation to n bus system
result in a matrix form as follows.
• Ibus is the vector of the injected bus currents

(external current sources).

• The current is positive when it flows towards
the bus and negative when it flows away from
the bus.
• Vbus is the vector of bus voltages with respect
to the reference node (node voltages).
• Ybus is known as the bus admittance matrix.
• The diagonal element of each node is the sum
of admittances connected to it.

• It is known as self-admittance or driving-point
admittance.
• The off-diagonal element equals to the

negative of the admittance between the
nodes.
• It is known as the mutual admittance or
transfer admittance.

• If bus currents are known, bus voltages can be
calculated from
• The inverse of the bus admittance matrix is

known as the bus impedance matrix Zbus.

Example 6.1
• Obtain the bus admittance matrix for the
following network. If E1=1.1L0° and E2=1L0°,
calculate the bus voltages
E1 E2

3. Power Flow Solution
• Power flow problem consists of determining:
 magnitudes and phase angles of voltages
at each bus.
 active and reactive power flow in each
line.
• The system is assumed to be balanced and
therefore a single phase model is used.

• Four different quantities are associated with
each bus:
 voltage magnitude |V|.
 voltage phase angle δ.
 real power P.
 reactive power Q.

• Generally, system buses are classified into
three types.
 Slack bus (swing bus).
 Load buses (P-Q buses).
 Regulated buses (generator buses), (P-V buses),
(voltage-controlled buses).

• Slack bus.
 It is the reference bus.
 |V| and δ are specified.
 Makes up the difference between the scheduled
loads and generated power (caused by network
losses).

• Load buses.
 P and Q are specified.
 |V| and δ are unknown.

• Voltage-controlled (regulated) buses.
 P and |V| are specified.
 δ and Q are unknown.
 The limits on the value of Q are also specified.

3.1 Power Flow Equation
• Consider a typical bus of a power system
network shown in Fig. 6.7
• Transmission lines are represented by their
equivalent π models and the impedances are
converted to per unit admittance on a
common MVA base.

• Eq. 6.27 is a system of algebraic nonlinear
equation that must be solved by iterative
techniques.

4. Gauss-Seidel Power Flow Solution
• In Gauss-Seidel method, eq. (6.27) is solved
for Vi and the iterative sequence becomes:
Where:
yij is the actual admittance in per unit.
Pisch is the real power in per unit.
Qisch is the reactive power in per unit.
• Current entering a bus is assumed to be
positive.
• Current leaving a bus is assumed to be
negative.
• Pisch and Qisch are positive if they are injected
into a bus (generator buses).
• Pisch and Qisch are negative if they are flowing
away from a bus (load buses).

• If eq. (6.27) is solved for Pi and Qi, we get
• The power flow equations are usually

expressed in terms of the elements of the bus
admittance matrix.

• Let
then eq. (6.28) becomes

• Eqs. (6.29) and (6.30) become
• Note that Yii includes the admittance to

ground of line charging susceptance and any
other fixed admittance to ground.

• Note that under normal operating conditions:
 The voltage magnitude of the buses are about 1
pu or closer to the voltage magnitude of the slack
bus.
 Voltage magnitudes at load buses are somewhat
lower than the voltage of the slack bus depending
on the reactive power demand.
 The schedule voltage at the generator buses are
somewhat higher than the voltage of the slack
bus.

 The phase angles of the load buses are below the
reference angle depending on the real power
demand.
 The phase angles of the generator buses may be
higher than the reference angle depending on the
amount of real power flowing into the bus.
• For the Gauss-Seidel method, an initial
estimate of 1.0+j0.0 for unknown voltages is
satisfactory.

• For P-Q buses, the real and reactive powers
(Pisch and Qisch) are known.
• Starting with an initial value, eq. (6.31) can be
used to calculate the real and imaginary
components of the voltage Vi(k+1).

• For P-V buses, Pisch and |Vi| are known.
• Eq. (3.33) is first solved for Qi(k+1).
• Then eq. (3.31) to solve for the voltage Vi(k+1).

• Since |Vi| is specified, only the imaginary part
of Vi(k+1) is retained.
• The real part of Vi(k+1) is calculated to satisfy:

• ei(k+1) and fi(k+1) are the real and imaginary
components of the voltage Vi(k+1) in the
iterative sequence.
• The rate of converge is increased by applying
an acceleration factor to the approximate
solution obtained from each iteration.
• For typical systems, a satisfactory value for the

acceleration factor (α) is in the range of 1.3 to
1.7.

• The updated voltages immediately replace the
previous values in the solution of the
subsequent equations.
• The process is continued until the changes in
the real and imaginary components of the bus
voltages between success iterations are within
specified accuracy.

• A voltage accuracy in the range of 0.00001 to
0.00005 pu is satisfactory.
• In practice, the method for determining the
completion of a solution is based on an
accuracy index set up on the power mismatch.
• The iteration continues until the magnitude of
the largest component in the ΔP and ΔQ
columns is less than the specified value.
• A typical power mismatch accuracy is 0.001
pu.

• Once a solution is converged, the net real and
reactive powers at the slack bus are calculated
using eqs. (6.32) and (6.33).

5. Line Flows and Losses
• Consider the line connecting the two buses i
and j shown in Fig. 3.8.

• Current iij, measured at bus i and defined
positive in the direction from i to j is given by:
• Similarly, current Iji, measured at bus j and

defined positive in the direction from j to i is
given by:

• The complex power Sij from bus i to j and Sji
from j to i are given by
• Therefore, the power loss in the line i-j is

Example 6.7
• Fig. 6.9 shows the one-line diagram of a
simple three-bus power system with
generation at bus 1. The magnitude of the
voltage at bus 1 is adjusted to 1.05 pu.
• The schedule loads at buses 2 and 3 are as
marked on the diagram. Line impedances are
marked in pu on a 100-MVA base and the line
charging susceptances are neglected.

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

a. Line impedances are converted to admittances
The admittance diagram is shown in Fig. 6.10.

At the P-Q buses, the complex loads expressed
in pu are
Starting from an initial estimate of
V2 and V3 can be calculated using (6.28) as

follows.

For the second iteration
The process is continued and the solution is

converged with an accuracy of 5X10-5 pu in
seven iterations as given below.

The final solution is

b. The slack bus power is obtained using eq.
(6.27).
Therefore,

c. To find the line flows, line currents must first
be calculated.
Therefore,

The line losses are
The power flow diagram is shown in Fig. 6.11.

Real Power is indicated by
Reactive power is indicated by
The values within parentheses are the real and
reactive losses in the line.

Example 6.8
• Fig. 6.12 shows the one-line diagram of a
simple three-bus power system with
generators at buses 1 and 3. The magnitude of
voltage at bus 1 is adjusted to 1.05 pu. Voltage
magnitude at bus 3 is fixed at 1.04 pu with a
real power generation of 200 MW. A load
consisting of 400 MW and 250 Mvar is taken
from bus 2. Line impedances are marked in
per unit on a 100 MVA base. The line charging
susceptances are neglected.
Obtain the power flow solution by the Gauss-
Seidel method including line flows and line
losses.

Line impedances are converted to admittances
as follows.
The load and generation expressed in per unit

are

Bus 1 is taken as the reference bus (slack bus).
Starting from an initial estimate of
V2 and V3 can be calculated from eq. (6.28) as

Bus 3 is a regulated bus for which voltage
magnitude and real power are specified.
For the voltage controlled bus, the reactive
power is first computed using eq. (6.30)

The value of Q3(1) is used as Q3sch for the
computation of the voltage at bus 3. The
complex voltage at bus 3, denoted by Vc3(1) is
calculated as follows.

Since |V3| is held constant at 1.04 pu, only the
imaginary part of Vc3(1) is retained. Hence,
f3(1)=-0.005170.
Its real part is obtained from
Therefore,

For the second iteration

Since |V3| is held constant at 1.04 pu, only the
imaginary part of Vc3(2) is retained. Hence,
f3(1)=-0.00730. Its real part is obtained from

The process is continued and a solution is
converged with an accuracy of 5X10-5 pu in
seven iteration as given below
f3(1)=-0.00730. Its real part is obtained from

The final solution is
Line flows and line losses in MW and Mvar are

7. Tap Changing Transformers
• Consider the simple transmission line shown
in Fig. 2.9.

• The flow of real power along a transmission
line is determine by the angle difference of
the terminal voltages.
• The flow of reactive power is determine

mainly by the magnitude difference of
terminal voltages.

• Real and reactive powers can be controlled by
the use of tap changing transformers and
regulating transformers.
• In a tap changing transformer, when the ratio
is at the nominal vale, the transformer is
represented by a series admittance yt in per
unit.

• With off-nominal ratio, the per unit
admittance is different from both sides of the
transformer, and the admittance must be
modified to include the effect of the off-
nominal ratio.

• Consider a transformer with admittance yt in
series with an ideal transformer representing
the off-nominal tap ratio 1:a as shown in Fig.
6.14.
• yt is the admittance in per unit based on the

nominal turn ratio and a is the per unit off-
nominal tap position allowing for small
adjustment in voltage of usually ±10 percent.
• In the case of phase shifting transformers, a is
a complex number.
• Consider a fictitious bus x between the turn
ratio and admittance of the transformer.
• Since the complex power on either side of the
ideal transformer is the same, it follows that if
the voltage goes through a positive phase
angle shift, the current will go through a
negative phase angle shift.

• Thus, for the assumed direction of currents
• Where the current Ii is given by
• Substituting for Vx to get

• Substituting for Ii from (6.45) to get
• Writing (6.45) and (6.46) in a matrix form

results in

• For the case when a is real, the π model
shown in Fig. (6.15) represents the admittance
matrix in (6.47).

• In the π model, the left side corresponds to
the non-tap side and the right side
corresponds to the tap side of the
transformer.

8. Newton-Raphson Power Flow Solution
• Newton-Raphson method is mathematically
superior to the Gauss-Seidel method.
• It is less prone to diverge with ill-conditioned
problems.
• For large power systems, it is found to be
more efficient and practical.
• The number of iteration required to obtain a
solution is independent of the system size.

• However, more functional evaluations are
required at each iteration.
• Since in the power flow problem real power
and voltage magnitude are specified for the
voltage-controlled buses, the power flow
equation is formulated in polar form.

• For the typical bus of the power system shown
in Fig. 6.7, the current entering bus i is given by
• This equation can be rewritten

in terms of the bus admittance
matrix as
• In eq. (6.48), j includes bus i.

• Expressing eq. (6.48) in polar form gives
• The complex power at bus i is
• Substitute from (6.49) for Ii in (6.50) results in

• Separating the real and imaginary parts give
• Equations (6.52) and (6.53) constitute a set of

nonlinear algebraic equations in terms of the
independent variables
 voltage magnitude in per unit, and
 phase angle in radians.

• There are two equations for each load bus,
given by (6.52) and (6.53).
• There is one equation for each voltage-
controlled bus, given by (6.52).

• Expanding (6.52) and (6.53) in Taylor’s series
about the initial estimate and neglecting all
higher order terms results in the following set
of linear equations.

• Bus 1 is assumed to be the slack bus.
• The Jacobian matrix gives the linearized
relationship between small changes in voltage
angle Δδi(k) and voltage magnitude Δ|Vi(k)|
with the small changes in real and reactive
power ΔPi(k) and ΔQi(k).

• Elements of the Jacobian matrix are the
partial derivatives of (6.52) and (6.53),
evaluated at Δδi(k) and Δ|Vi(k)|.
• The Jacobian matrix can be written as

• For voltage-controlled buses, the voltage
magnitudes are known.
• Therefore, if m buses of the system are
voltage-controlled, m equations involve ΔQ
and ΔV and the corresponding columns of the
Jacobian matrix are eliminated.
• Accordingly, there are n-1 real power
constraints and n-1-m reactive power
constraints.

Matrix Order
Jacobian matrix (2n-2-m)X(2n-2-m)
J1 (n-1)X(n-1)
J2 (n-1)X(n-1-m)
J3 (n-1-m)X(n-1)
J4 (n-1-m)X(n-1-m)

• The diagonal and off-diagonal elements of J1
are

are

are

are

• The terms ΔPi(k) and ΔQi(k) are the differences
between the schedule and calculated values,
known as the power residuals, given by
• The new estimates for bus voltages are

Example 6.10
• Obtain the power flow solution by the
Newton-Raphson method for the system of
Example 6.8.

9. Fast Decoupled Power Flow Solution
• Power transmission lines have a very high X/R
ratio.
• For such a system, real power changes ΔP are
less sensitive to changes in the voltage
magnitude and most sensitive to changes in
phase angle Δδ.
• However, reactive power changes ΔQ are less
sensitive to changes in the phase angle Δδ and
most sensitive to changes in the voltage
magnitude.
• Therefore, it is reasonable to set the elements
J2 and J3 of the Jacobian matrix to zero.
• Hence,

• (6.69) and (6.70) show that the matrix
equation is separated into two decoupled
equations requiring considerably less time to
solve compared with the time required to
solve (6.54).

• Furthermore, considerable simplification can
be made to eliminate the need for
recomposing J1 and J4 during each iteration.
• This procedure results in the decoupled power
flow equations developed by Stott and Alsac.
• The diagonal elements of J1 described by
(6.55) may be written as

• Replacing the first term of the above equation
with –Qi, as given by (6.53), results in
where Bii=|Yii|sinθii is the imaginary part of

the diagonal elements of the bus admittance
matrix.
Bii is the sum of susceptances of all the
elements incident to bus i.

• In a typical power system, the self-
susceptance Bii>>Qi, so Qi may be neglected.
• Further simplification is obtained by assuming
|Vi|2≈|Vi|.
• Under normal operating conditions, δj-δi is

quit small. Thus, by assuming θii-δi+δj ≈ θii, the
off-diagonal elements of J1 becomes.

• Further simplification is obtained by assuming
|Vj|≈1.
• Similarly, the diagonal elements of J4

described by (6.61) may be written as

• Replacing the second terms of the above
equation with –Qi as given by (6.53) results in
• Again, since Bii(=Yiisinθii)>>Qi, Qi may be

neglected and (6.61) reduces to
• Similarly in (6.62), assuming θij-δi+ δj≈ θij gives

• Therefore, (6.69) and (6.70) take the following
form
B’ and B’’ are the imaginary part of the bus

admittance matrix Ybus.
• Since the elements of this matrix are constant,
they need to be triangularized and inverted
only once at the beginning of the iteration.
• B’ is of order (n-1).
• For voltage-controlled-buses where |Vi| and Pi
are specified and Qi is not specified, the
corresponding row and column of Ybus are
eliminated.
• Hence, B’’ is of order of (n-1-m), where m is
the number of voltage-controlled buses.

• Therefore, in fast decoupled power flow
algorithm, the successive voltage magnitude
and phase angle changes are

• The solution of fast decoupled power flow
 requires more iterations than Newton-Raphson
method,
 but requires considerably less time per iteration,
and a power flow solution can be obtained rapidly.
• This technique is very useful
 in contingency analysis where numerous outages
are to be simulated, or
 when a power flow solution is required for on-line
control.

Example 6.12
• Obtain the power flow solution by the fast
decoupled method for the system of Example
6.8.

• In this example, the fast decoupled power
flow takes 14 iterations with the maximum
power mismatch of 2.5X10-4 pu compared to
the Newton-Raphson method which takes
only three iterations.
• The highest X/R ratio of the transmission lines
in this example is 3.
• For systems with higher X/R ratio, the fast
decoupled power flow method converges in
relatively fewer iterations.

• However, the number of iterations is a
function of system size.
• Finally, the line flows are calculated in the
same manner as the line flow calculations in
the Gauss-Seidel method described in
Example 6.7, and the power flow diagram is as
shown in Fig. 6.13.

Reference
• H. Saadat, Power System Analysis, 3rd ed., PSA
Publishing, 2010
• J. Grainger and W.D. Stevenson, Power System
Analysis, Mc Graw-Hill, 1994.

Topic 2 Power Flow

Uploaded by

Copyright:

Available Formats

Topic 2 Power Flow

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Topic 2 Power Flow

Uploaded by

Copyright:

Available Formats

Topic 2

EE 605 Power Network Analysis

Dr. Muhyaddin Jamal Rawa

10/01/2019 Dr. Muhyaddin Jamal Rawa 3

10/01/2019 Dr. Muhyaddin Jamal Rawa 4

10/01/2019 Dr. Muhyaddin Jamal Rawa 5

10/01/2019 Dr. Muhyaddin Jamal Rawa 6

10/01/2019 Dr. Muhyaddin Jamal Rawa 7

10/01/2019 Dr. Muhyaddin Jamal Rawa 8

• By introducing the following admittances

10/01/2019 Dr. Muhyaddin Jamal Rawa 9

• As there is no connection between bus 1 and

10/01/2019 Dr. Muhyaddin Jamal Rawa 10

• Ibus is the vector of the injected bus currents

10/01/2019 Dr. Muhyaddin Jamal Rawa 11

10/01/2019 Dr. Muhyaddin Jamal Rawa 12

• The off-diagonal element equals to the

10/01/2019 Dr. Muhyaddin Jamal Rawa 13

• The inverse of the bus admittance matrix is

10/01/2019 Dr. Muhyaddin Jamal Rawa 14

10/01/2019 Dr. Muhyaddin Jamal Rawa 15

10/01/2019 Dr. Muhyaddin Jamal Rawa 18

10/01/2019 Dr. Muhyaddin Jamal Rawa 19

10/01/2019 Dr. Muhyaddin Jamal Rawa 20

10/01/2019 Dr. Muhyaddin Jamal Rawa 21

10/01/2019 Dr. Muhyaddin Jamal Rawa 22

10/01/2019 Dr. Muhyaddin Jamal Rawa 23

10/01/2019 Dr. Muhyaddin Jamal Rawa 24

10/01/2019 Dr. Muhyaddin Jamal Rawa 26

10/01/2019 Dr. Muhyaddin Jamal Rawa 28

• The power flow equations are usually

10/01/2019 Dr. Muhyaddin Jamal Rawa 29

then eq. (6.28) becomes

10/01/2019 Dr. Muhyaddin Jamal Rawa 30

• Note that Yii includes the admittance to

10/01/2019 Dr. Muhyaddin Jamal Rawa 31

10/01/2019 Dr. Muhyaddin Jamal Rawa 32

10/01/2019 Dr. Muhyaddin Jamal Rawa 33

10/01/2019 Dr. Muhyaddin Jamal Rawa 34

• Then eq. (3.31) to solve for the voltage Vi(k+1).

10/01/2019 Dr. Muhyaddin Jamal Rawa 35

10/01/2019 Dr. Muhyaddin Jamal Rawa 36

• For typical systems, a satisfactory value for the

10/01/2019 Dr. Muhyaddin Jamal Rawa 37

10/01/2019 Dr. Muhyaddin Jamal Rawa 38

10/01/2019 Dr. Muhyaddin Jamal Rawa 39

10/01/2019 Dr. Muhyaddin Jamal Rawa 40

10/01/2019 Dr. Muhyaddin Jamal Rawa 41

• Similarly, current Iji, measured at bus j and

10/01/2019 Dr. Muhyaddin Jamal Rawa 42

• Therefore, the power loss in the line i-j is

10/01/2019 Dr. Muhyaddin Jamal Rawa 43

10/01/2019 Dr. Muhyaddin Jamal Rawa 44

10/01/2019 Dr. Muhyaddin Jamal Rawa 46

The admittance diagram is shown in Fig. 6.10.

10/01/2019 Dr. Muhyaddin Jamal Rawa 47

Starting from an initial estimate of

V2 and V3 can be calculated using (6.28) as

10/01/2019 Dr. Muhyaddin Jamal Rawa 48