Optimal Power Flow in PSS/E

Tao Cui (tcui@ece.cmu.edu)

In this tutorial, we will learn the basics of using PSS/E to solve optimal power flow problems (OPF). The
goal of this tutorial is to introduce: 1) what OPF problem PSS/E can solve; 2) how OPF problem is
formulated in PSS/E; 3) how the problem is organized (input, output and related files) and 4) how to
solve the problem in PSS/E.

1. General about Optimal Power Flow Case:

What is OPF: Given the network structures and parameters, the optimal power flow (OPF) is a
computation process to determine the value of the control variables such that the power system is
operating at a point that minimizes certain cost or maximizes certain profit (objective function),
meanwhile satisfying the security and operational constraints.

The objective function is the objective of the optimization, usually it would be the cost or loss that need
to be minimize or the benefit that need to be maximize.

The constraints are usually the physical or security limits of certain variables that must be satisfied. For
example, the generator has power output limits.

The control variables are the variables that need to be optimally computed to achieve the minimal or
maximal objective. These control variables are usually the control actions that can affect the power
system states, such as: the scheduled active or reactive power output of a generator, the scheduled
voltage amplitude of a generator, and sometime, the load that can be controlled, etc. So the Control
Variables are related to the actual control actions in power system. There are also some other variables
called state variables in the optimization, these variables represents the system states and are usually
depended on the control variables and given system parameters.

2. Optimal Power Flow in PSS/E:

The optimal power problem consists three parts:

1) Power flow case: PSS/E OPF uses “.sav” or “.raw” for the power flow case, the power flow
configuration in OPF is the same as in power flow analysis.
2) Constraints: PSS/E OPF uses “.rop” file to describe all constraints. It is an ASCII file, you can use
any editor to read and write. You can also load and modify the “.rop” file using PSS/E GUI.
3) Objective functions: PSS/E provides various standard cost functions. You can use menu or check
box to select the cost functions. You can also input your own cost functions.

We are already familiar with the power flow analysis. Now let’s look at the two new components:
constraints and objective functions.
2. Configure Optimal Power Flow Data in PSS/E:
We use the "savnw" example. First click "File->Open", open the "savnw.sav" file to load the power flow
base case, then click "File -> Open", change the file type to ".rop" file, select "savenw.rop" to load the
optimal load flow constraints file as showed in Figure 1.

Figure 1

In the ".rop" file, there are two types of data related to OPF: 1) The bus oriented data and table oriented
data. These two types of data are displayed in the different way in PSS/E.

Bus oriented data: we can view and change these data from the "OPF data" spread sheet Figure 2:

OPF data
Tree View spread sheet

Select to view different group of OPF data

Click this first or select
"OPF->Data"

Figure 2
Table oriented data: Select "OPF->Data tables..." to view the table oriented data as showed in Figure 3:

Figure 3

We can view and modify all these data in PSS/E GUI. Also, all these data are stored in ".rop" file in ACSII
format, so we can open to see and modify the ".rop" file using any editor as showed in Figure 4. All the
formats in "rop" file are fixed, you can modify the data in the line, add or remove lines. But you should
not change the format.

Figure 4
example case are:

1) The topology and all impedance data showed in Error: Reference source not found.
2) Some data show in Error: Reference source not found are given:

Table 1

Bus No Bus Type V θ P Q

1 Slack 1.040 0 ? ?
2 Generation 1.025 ? 163 ?
3 Generation 1.025 ? 85 ?
4 Load (Zero Injection) ? ? 0 0
5 Load ? ? -125 -50
6 Load ? ? -90 -30
7 Load(Zero Injection) ? ? 0 0
8 Load ? ? -100 -35
9 Load(Zero Injection) ? ? 0 0
As shown in Table 1, the Power Flow computation will be able to compute all “?” as well as other result
data in Error: Reference source not found.

3. Create Power Flow Case by Single Line Diagram of GUI

We will now input the system impedance data and the given power flow data into PSS/E.

3.1 Placing all components

Open PSSE, Select File->New, in Figure 5, select Network Case and Diagram
Figure 5

Then, in Figure 6, PSS/E asks us to input the global parameters of our power flow case. We know from
Error: Reference source not found, the example system’s Base MVA is 100 MVA, the Heading line 1 and
2 are used for reference purpose, they will appears on the first two lines of the raw file. Units for ratings
of non-transformer branches means what unit are used for the ratings of line etc, we use the default
one.
Figure 6

Figure 7
In Figure 7, the “Network Data” is the spread sheet that contains all the data of the power flow case,
“Diagram1” is the Single Line Diagram. We will draw buses and line on the Single Line Diagram and
associate the element of diagram with the Network Data spread sheet.

Click anywhere in the “Diagram1”, the Diagram toolbar will be activated:

In this toolbar, you can select the elements and place them on the diagram to draw the Single Line
Diagram for the power network.

We may need to place all the buses first, before we place buses, we can select “Edit -> Preferences” to
let PSS/E automatically numbering the buses we placed in the diagram, as show in Figure 8.

Make sure this

Change this to “1”
is selected
may be better for
our example

Figure 8

Now we click to place buses on the diagram.

Figure 9

Note in the “Network Data” window, on the “Bus” tab, these buses are automatically added.

Figure 10

Then, we click to place all branches (non-transformer) onto the diagram. In our example, the
branches: 4-5, 5-7, 4-6, 6-9, 7-8, 8-9 can be placed in this way.

The branches: 1-4 2-7 3-9 are actually two-winding transformers, we need click to place these
branches. After we place all branches, the diagram looks like Figure 11.
Figure 11

Then we click to place generators on bus 1, 2, 3. And click to place load on bus 5, 6, 8. Now the
diagram looks like this in Figure 12:

Figure 12
 Now we have all the elements on the diagram, we need to properly configure them and assign correct
parameters. Navigating through the “Tree View” on the left, you can check the element we have in our
case. The diagram is also bounded to the system data. And each element on the diagram has an entry in
the spread sheet. So “Tree View”, “Diagram”, and “Network Data” spread sheet are all bounded to the
power flow case now(Figure 13) , you can click the items in either one of them to modify the
parameters.

Network Data
Spread Sheet

Tree
View
Diagram

Figure 13

3.2 Configuring Elements and Input Parameters:

By placing the elements on diagram, we have all the power flow elements in place. By connecting the
buses using branches, we also have input the topology information of the network. Now we need to
configure these elements and input the parameters for these elements. We will configure the elements
in this order: Plants, Buses, Generators, Branches (Lines and Transformers) and Loads.

Plants:

Figure 14
Once we place the generator on the bus, we actually put a power plant on that bus. So in the spread
sheet’s plant tab, there are new plants entry added in. The parameter “VSched” for the “Plant” is used
as scheduled voltage amplitude in the power flow computation. So we put the scheduled voltage
amplitudes into these cells in spread sheet. As shown in Figure 14.

Buses:
In order to solve the power flow, we need a swing bus, several generator buses and several load buses.

Click bus 1 will popup Figure 15. In this example, set the bus 1’s type code to Swing bus (also called Slack
bus). We have already specified the voltage amplitude to 1.040 in “Plants” section in Figure 14. Here we
need to make sure the Angle is specified to 0 (as show in Error: Reference source not found). If you also
want to check the actual value, you need to input the Base kV for the voltage on this bus. Otherwise,
you can leave it to 0. The Grouping Data are for the management of the system into groups, we don’t
need it right now.

Figure 15 Swing Bus

For other generator buses which are not swing buses, we need to specify the bus type code to
Generator Bus. As shown in Figure 16 for bus 2 and bus 3. We need to specify the parameters in red
box.
Figure 16 Generator Bus

The rest of the buses are load buses, we can just use the default settings. For example, in Figure 17, it is
the “Non-Generator Bus” by default. We don’t need to modify anything.

Figure 17

Generators:
We now need to specify the generators. In this example, for each plant (or generator bus), we only have
one generator connected to it. For bus 1, which is the swing bus, we don’t need to specify any power
output value. Starting from Bus 2, the generator parameter is showed in Figure 18. For our example, we
need to specify the active power output of each generator, which is the Pgen in red box in Figure 18.
Figure 18 Generator Data

Branches:
Then, we input the non-transformer branch data:

Non-transformer branch: 8-9, 7-8, 7-5, 5-4, 4-6, 6-9.

We use Branch 8-9 as an example: as shown in Error: Reference source not found, the branch
impedance is: 0.0119+j0.1008, the charging B is: B/2 = j0.1045 (B = 0.209). Modify the values in the red
box in Figure 19. The other non-transformer branch can be input in the same way.

Figure 19
There will be a warning saying the line is no longer zero impedance, just click OK.

Transformer Branch: 1-4, 2-7, 3-9

We need to specify the reactance X (the data is shown in Error: Reference source not found). We use
Branch 1-4 as example as showed in Figure 20:

Figure 20
Loads:
The default load model is a combination of impedance, current, and power, this is also called ZIP load
model. For our example, we only consider the load as constant power, so we need to modify the Pload
and Qload parts as shown in red box in Figure 21.

Figure 21

3.3 Save and Solve the Power Flow:

Save Our Data:
Now we finish input all the necessary data for the 9 bus power flow case. We need to save our work: this
includes the data we just input, and the diagram we just draw.

Figure 22
As shown in Figure 22, click anywhere on the Network Data spreadsheet, select “Save”, we can save our
data into “.sav” PSS/E binary file and “.raw” ASCII raw power flow format file as well as other power flow
exchange data format. We should make sure to save it to “.sav” since it is the native PSS/E format for
power flow case.

Also, make sure you have saved the Diagram into Single Line Diagram file (.sld) file: click anywhere on
the diagram and click “File->Save”, select the file type as .sld file.

Solve Power Flow:

Select “File->Power Flow-> Solution-> Solve(NSOL/FNSL/FDNS/SOLV/MSLV)”, the following window
popup:

Figure 23

Just use the default settings, click Solve, the power flow will be solved, and all the data in spread sheet
and single line diagram will be updated as showed in following figures:
You can compare the results to the power flow solution in Error: Reference source not found, if you can
obtain the same results as shown in P.M Anderson’s example, then Congratulations, you have your first
PSS/E power flow case solved. You can save these results to the files for further usage.

4. Summary
In this tutorial, we learned how to create a power flow case via PSS/E GUI, you can use diagram
schematic combined with Network Data spread sheet to create a simple system. The data needed for
computation are stored in the spread sheet. The diagram are used for better visualization, it is not
necessary but can be helpful for our studies in the future.