1

EEEN30047 Power System Analysis

2020-21

Power Flow Analysis & Frequency Stability Coursework

Introduction

These are the instructions for the Power System Analysis software-based coursework on power flow analysis

and frequency stability.

You should have already completed the pre-lab exercise (to work out loading values). If you have not

completed that then stop reading this, and go and complete that.

Definitely done the pre-lab? OK, let’s keep going.

This coursework will use a (very) simplified equivalent model of the GB power system to study some of the

concepts you have learned about during the power system analysis course. This coursework has two parts:

Part 1 asks you to analyse the steady state behaviour of the GB power system model based on power flow

results. You will identify some problems that might occur and look at how to solve these using grid

reinforcement strategies.

Part 2 will use the same network in order to analyse the frequency stability of the system and analyse how

this is affected by the integration of non-synchronous sources of energy such as large wind power plants.

The objectives of the coursework are:

• To be able to explain the need for power flow analysis, the processes involved in completing it, and

the information it can provide.

• To learn how to use PowerWorld to conduct power flow analysis and frequency stability analysis.

• To be able to describe and explain different problems that power system operators may face when

integrating large amounts of renewable generation into power systems as well as how different

mitigation strategies work.

Your report must not exceed a total length of 10 pages (excluding any cover or contents pages, but including

references).

You can download a free educational version of the software (PowerWorld Simulator – Version 21) to

complete this coursework on your own computer at this link: https://www.powerworld.com/download-

purchase/demo-software

All parts of the coursework should be completed by Friday 4 December 2020 at 13:00. Late submissions will

be awarded zero marks (as detailed in your course student handbook). If you have a genuine need for an

extension, get in touch before the submission deadline, not after.

If you have an automatic one week extension, there will be a separate upload option for you (it will be

obvious).

Avoid any form of plagiarism. The system Turnitin will be used to check this – you will be caught.

2

Tips for analysing your results

The following tips are recommended as a general guide on how to analyse the results you get during this

coursework. You might find these general tips useful for other coursework or reports you have to write.

Pick the best plot type

• Sometimes that might mean using a table instead of a plot. That is okay if it gets the

right information across. Plots can be good for showing trends but not precise

values. If you need precise values then a table is very useful.

• Think about your data type, is it continuous or discontinuous (e.g. categories are

discontinuous, you could reorder them and it would not matter). If it is

discontinuous, then do not draw lines between the points. If it is continuous, make

sure you use consistent axis scaling (i.e. do not have different intervals between the

tick marks).

• Avoid 3D plots unless you need them. They do not look cool. They distort shapes

and make it hard to see relationships. Search for “chartjunk” to see what to avoid.

• Make sure that any lines or points can easily be seen and distinguished. If they

cannot (and that might be the point) then make sure you clearly state this in the

next step!

Describe your results

• Do not expect the reader to look at your table or plot and work out what the

numbers show. Tell them. Explicitly refer to your results and tell them what to look

for.

• Be specific and make sure you state the directions of any changes (e.g. increases or

decreases) as you vary controlled parameters.

• Is the relationship linear or non-linear? Does it have a threshold where things

change? Describe all of this.

Quantify things

• You are an engineer so put numbers on things. If something increased by 10% as the

control parameter doubles then state this.

• Make sure you work to a suitable precision. You cannot report things to eight

decimal places if the methods or software does not output to that degree of

accuracy.

Put it into context

• So it changed by 10%, does this actually matter? What are the acceptable limits for

that output? If it can vary by +/- 50% under normal operation then maybe this 10%

change is not that significant.

• This is your chance to refer to the wider engineering context of your results and

explain the broader relevance. What impact might this have on a real system? Could

this cause other issues.

3

Explain why this is happening

• This is the hardest step but the most useful. If you can explain why this result is

happening, then you can apply this knowledge to other systems and scenarios to

generalise the results.

• Can you refer back to what you learned in class or in books? Are there any equations

or theories which explain this result?

So what?

• It is time to make some conclusions. Does this result tell you something new? Or

maybe it confirms some existing knowledge or theory.

• This is your chance to explain the significance of the result and how this ties in with

the overall aims and objectives of the study.

• It might be that this result is going to lead onto another study that will help you

answer some outstanding questions. You should state this and make those links

clear.

It is okay if you cannot do all of these steps every time you produce some results. In fact, if you did, your

report would be far too long and would often be quite repetitive. But try to include these different aspects in

your report writing and think about these steps when analysing your results.

4

Consider the following examples of discussion about some results you might get when testing a voltage

divider circuit. See how the analysis improves as the different aspects are incorporated into the discussion.

As the load resistor changes, the voltage drop across it changes.

This is very vague and not good at all.

This is not worth any marks – don’t write statements like this.

As the value of the load resistor increases, the voltage drop across it also increases…

This description is better (but not by much) and at least describes the direction of change.

…An increase in resistance of 10% from the initial value results in a 4.76% increase in the

voltage drop…

Now this adds some quantification to these variations.

…This relationship is non-linear but saturates. Initial increases in the value of load resistor

cause significant increases in voltage. Once the load resistor exceeds 190 Ω, the voltage drop

has reached 9.5 V. Doubling this load resistance to 380 Ω results in a small increase in voltage

drop to 9.744 V (just a 2.56% increase despite a 100% increase in resistance)…

This adds more detail about the type of relationship.

…This saturation can be used to determine a range of load resistor values that will provide

consistent voltage outputs. For example, If it is important to maintain the load voltage drop

above 9.5 V then a load resistance greater than 190 Ω should be used…

Now we are starting to add a little bit of broader context to the discussion to try to relate this

to practical systems. This is a little bit weak as I am just describing a voltage divider. It will be

easier to contextualise results from models of real systems.

…This load resistance and voltage relationship can be explained by considering the voltage

divider equation: = [2 (1 + 2)⁄ ]. By taking the test values obtained (i.e. = 9.5 V

when 2 = 190 Ω, and = 9.744 V when 2 = 380 Ω), it is possible to form simultaneous

equations and solve to find the value of the input voltage () and other resistance (1) in

the circuit. In this case, these values can be calculated as 10 V and 10 Ω respectively…

Now the relationship has been explained by referring to a known equation. This result can now

be generalised to other cases with other values.

…The results obtained have validated the performance of the circuit against the expected

behaviour and the results have been used to reveal unknown circuit component values.

This is a short concluding statement to summarise the discussion and the findings. In your

work, some conclusions will be brief like this and some may be much lengthier.

IM

P

R

O

V

IN

G

Q

U

A

LI

T

Y

O

F

D

IS

C

U

S

SI

O

N

5

Scripting in PowerWorld Simulator

This coursework requires you to run quite a lot of simulations. This is necessary in order to generate the

results you need to analyse the system and draw some meaningful conclusions. As a power system engineer,

you may often be asked to do similar tasks – running lots of simulations to identify conditions or situations

which might have issues or to find the limits of system operation.

The software you are using – PowerWorld Simulator – has some basic scripting capability. Unfortunately it is

not very user friendly (and does not support for or while loops), but it does have enough functionality to

use if you want to.

Please note, you do not have to do any scripting in order to complete this coursework. You can do the

whole thing using the graphical interface by clicking on parameters and changing values. You will have to

choose if it is quicker and easier to develop, debug, and run scripts or to click and type.

There are no marks awarded for scripting.

You can find the script command execution window on the Tools ribbon by clicking the