Load Flow

Load Flow Analysis
• A load flow study is done on a power system to

ensure that
– Generation supplies the demand (load) plus
losses.
– Bus voltage magnitudes remain close to rated
values
 Generation operates within specified real and
reactive power limits
 Transmission lines and transformers are not
overloaded.
A Load Flow Study Specifically Investigates
the Following
• Bus-bar voltages
• Effect of rearranging circuits and
incorporating new circuits on system
loading.
• Effect of injecting in-phase and quadrature
boost voltages on system loading.
• Optimum system running conditions and
load distribution.
• Optimum system losses.
• Optimum rating and tap range of
transformers.
Load Flow Problem
• The starting point of a load flow problem is a
single line diagram of the power system, from
which input data for computer solutions can be
obtained. Input data consist of bus data,
transmission line data and transformer data.
• The following four variables are associated with
each bus k - voltage magnitude Vk, phase angle
dk, net real power Pk and reactive power Qk
supplied to the bus.
• The starting point of a load flow problem is a
single line diagram of the power system, from
which input data for computer solutions can be
obtained. Input data consist of bus data,
transmission line data and transformer data.
• The following four variables are associated with
each bus k - voltage magnitude Vk, phase angle
dk, net real power Pk and reactive power Qk
supplied to the bus.
• Voltage Controlled bus or PV bus - These are
generally generator buses where Pk and Vk
are specified and Qk and dk are computed.
• There are three methods of solving the load
flow problem.
– A) The Gauss Seidel Method
– B) The Newton Raphson Method
– C) Fast Decoupled Load Flow (FDLF)
Load Flow Analysis
• is the calculation of magnitude and phase
angle of voltages of all the buses.
• It gives P , Q through tr.lines and losses.
• Load flow analysis solves the steady
operation state with node voltages and branch
power flow in the power system. The
mathematical model of load flow problem is a
nonlinear algebraic equation system without
differential equations.
• The power flow solution is used to evaluate
the bus voltage, branch current, real power
flow, reactive power flow for the specified
generation and load conditions.
• To solve the power flow equations, two of the
four variables must be known at each bus.
Specified unknown
1) Slack Bus V ,δ P,Q
2) Load Bus P,Q V ,δ
3) Voltage controlled Bus P, V Q,δ

Gauss-Seidel Method
• This method solves, by an iterative process,
the following equation that represents a
power system having N buses
• The relationship between node current and
voltage in the linear network can be described
by the following node equation: I = YV
Or
• Ii= ∑Yij *Vj ; i,j= 1,2,3,…..n
• (Pi - jQi )/Vi * = ∑Yij *Vj ; i,j= 1,2,3,…..n -----(1)
• There are n nonlinear complex equations. They are
the principal equations in load flow calculation.
Based on different methods to solve this equn.
various load flow algorithms can be formed.
• In the power system load flow problem, the
variables are nodal complex voltages and complex
powers: V, δ, P, Q. If there are n nodes in a power
system, then the total number of variables is 4 n.
• As mentioned above, there are n complex
equations or 2n real equations defined in
principal by (1), thus only 2n variables can be
solved from these equations, while the other
2n variables should be specified as original
data.
• Usually, two variables at each node are
assumed known, while the other two variables
are treated as state variables to be resolved.
• Pp - jQp = Ep * Ip
Therefore Ip = (Pp - jQp ) / Ep *
Ibus = [Ybus ] Ebus
• Ii= ∑Yij *Vj ; i,j= 1,2,3,…..n

• Ip= Yp1 E1 + Yp2 E2 ……+ Ypp Ep + …..+ Ypn En
Therefore Ip = ∑ Ypq Eq ; q = 1,….,n
= Ypp Ep + ∑ Ypq Eq ; q = 1,….,n
≠p
Ep = (1/Ypp)[Ip -∑ Ypq Eq ] q = 1,….,n
≠p
Ep = 1/Ypp[(Pp-jQp/Ep* )- ∑ Ypq Eq ]
Newton – Raphson method
We first need to rewrite complex power equations
as equations with real coefficients (we've seen this earlier):
n * n
 
Si  Vi I i*
 Vi   YikVk   Vi  Yik*Vk*
 k 1  k 1
These can be derived by defining
Yik  Gik  jBik
Vi  Vi e ji  Vi i
ik  i   k
Recall e j  cos   j sin 
• To solve the load flow equation, the relation of
node power with current should be used
I = (P i - jQi )/Vi * ;i = 1,2,…..,n
• where Pi, Qi are the injected active and
reactive power at node i, respectively.
• If node “i" is a load node, then Pi and Qi
should take negative values.
Ii= ∑Yij Vj *; where i,j= 1,2,3,…..n
Pi - jQi = Vi * Ii
= Vi ∠-δi ∑Yij Vj ∠(𝜽ij +δj )
where j = 1,2,….,n
Separating the real and imaginary parts
Pi = Vi ∑Vj Yij cos(𝜽ij-δi+δj)
Qi = Vi ∑Vj Yij sin(𝜽ij-δi+δj)
The Jacobian matrix gives the linearized
relationship between small changes in Δδi(k)
and voltage magnitude Δ[Vik] with the small
changes in real and reactive power ΔPi(k) and
ΔQi(k)
When k=i;and bus 1 is the slack;
∂Pi/ ∂δi = ∑ Vi VkYik sin(𝜽ik+ δk-δi)
k= 1,2,3,….,n ---------(1)
≠i
When k ≠ i
∂Pi/ ∂δk = Vi VkYik sin(𝜽ik+ δk-δi)
k=1, 2,3,….,n ---------(2)
≠i
When k = i; ∂Pi/ ∂Vi = 2 Vi Yii cos(𝜽ii ) + ∑VkYik
cos(𝜽ik+ δk-δi))
k= 1,2,3,….,n ---------(3)
≠i
When k ≠ i ; ∂Pi/ ∂Vk = Vi Yik cos(𝜽ik+ δk-δi)
k= 1,2,3,….,n ---------(4)
≠i
When k = i; ∂Qi/ ∂δi = -∑ Vi VkYik cos(𝜽ik+ δk-δi)
k= 1,2,3,….,n ---------(5)
≠i
When k ≠ i ; ∂Qi/ ∂δk = Vi VkYik cos(𝜽ik+ δk-δi)
k= 1,2,3,….,n ---------(6)
≠i
When k = i; ∂Qi/ ∂Vi = 2 Vi Yii sin(𝜽ii ) + ∑VkYik
sin(𝜽ik+ δk-δi))
k= 1,2,3,….,n ---------(3)
≠i
When k ≠ i ; ∂Qi/ ∂Vk = Vi Yik sin(𝜽ik+ δk-δi)
k= 1,2,3,….,n ---------(4)
≠i
Algorithm
1. Assume for PQ buses; V∠δ = 1 ∠00
For PV bus ; δ = ∠00
2.In the rth iteration
Pi r= Vi r ∑Vj r Yij cos(𝜽ij-δi r +δj r)
Qi r = Vi r ∑Vj r Yij sin(𝜽ij-δi r+δj r )
•
3) Compute
∆ Pi r = Pi (scheduled) – Pi r for PV & PQ buses
∆ Qi r = Qi (scheduled) – Qi r for PV & PQ buses

If all ∆ Pi r , ∆ Qi r < tolerance ; stop the iteration.
Calculate slack bus P & Q and line flows.
3) If the convergence criterion is not satisfied,
evaluate the Jacobian matrix elements using
equns. (1) – (6).
4) Solve equn. (A) for convergence of voltage
magnitude ∆V r and angle ∆δ r.
5) Next update voltage magnitudes and angles
V r+1 = V r + ∆ V r
δ r+1 = δ r + ∆δ r
If the system is having PV buses
In step 2, Q is computed each time and check its
Q limits and find whether it is going to operate
as a PV bus or PQ bus.
If the Q limit is violated, it will operate as a PQ bus
If not it will operate as a PV bus.
PQ bus ----- corrections for V and δ are needed
PV bus ------ correction for δ is needed
Advantages and disadvantages of NR method
 Advantages
– fast convergence as long as initial guess is close to
solution
– large region of convergence
 Disadvantages
– each iteration takes much longer than a Gauss-
Seidel iteration
– more complicated to code, particularly when
implementing sparse matrix algorithms
• Newton-Raphson algorithm is very common in
power flow analysis.
Fast Decoupled Load Flow (FDLF)
• The transmission lines have high X/R ratio.
• For such system the real power changes P are
less sensitive to changes in the voltage
magnitude and are most sensitive to change in
phase angle δ.
• Similarly the reactive power changes Q is less
sensitive to change in angles δ and is mainly
dependent on changes in voltage magnitude .
Therefore J2 = J3=0
∂Pi/ ∂δi = -Qi-Vi2 Bii
Vi2  Vi & Bii>> Qi
∂Pi/ ∂δi = -Vi . Bii
δk - δi  0;
∂Pi/ ∂δk = -Vi . Bik
∂Qi/ ∂Vi = -Vi2 Bii + Qi
Vi2  Vi & Bii>> Qi
∂Qi/ ∂Vi = -Vi Bii
δk - δi  0 & Vk = 1p.u.
∂Qi/ ∂Vk = -Vi. Bik
With these assumptions;
P/ Vi = -B’.  δ
Q/ Vi = -B”.  Vi
Above two equns. are solved for  δ and  Vi.
Then  δk+1 = δk +  δk
and  Vk+1 = Vk +  Vk
These values are used for the bus power
mismatches for the next iteration.
Prob: B= -23.508 11.764 11.764
11.764 -23.508 11.764
11.764 11.764 -23.508
[P2/ V2
P3/ V3] = [ -B22 -B23 ; -B32 -B33] * [  δ2 ;
 δ3]
Q2/ V2 = -B22.  V2
[0.73/1 ; -1.62/1.04] = [23.508 -11.764; -11.764
23.508] [  δ2 ;  δ3]
 δ2= -0.003 rad;  δ3 = -.068rad
δ2 = 0 -0.003 = - 0.003 rad
δ3 = 0 – 0.068 = -0.068 rad
Q/ Vi = -B”.  Vi
2.125 = (23.508)( V2)
 V2 = 0.09
V2= 1 + 0.09 =1.09

Load Flow

Uploaded by

Copyright:

Available Formats

Load Flow

Uploaded by

Document Information

Original Title

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

Load Flow

Uploaded by

Copyright:

Available Formats

Load Flow Analysis

• A load flow study is done on a power system to

3) Voltage controlled Bus P, V Q,δ

• Ii= ∑Yij *Vj ; i,j= 1,2,3,…..n

∆ Qi r = Qi (scheduled) – Qi r for PV & PQ buses

You might also like