Page: 1 of 13
Course:
ELEC ENG 3110 Electric Power Systems
ELEC ENG 7074 Power Systems PG
(Semester 2, 2025)
Assignment 1:
Power System Frequency Control
(Due 23:00 Fri. 19 September 2025
myUni Submission)
Electric Power Systems
Page: 2 of 13
Investigation of power system frequency control using simplified
models.
The objective of this assignment is to investigate some aspects of power system frequency con-
trol using simplified models. The investigation is intended to provide conceptual insights only.
The models are not necessarily optimized to yield the best frequency control performance. The
Mathworks Simulink program, which is a companion of Matlab, will be employed to build and
simulate dynamic models of the systems. Thus, the assignment will also provide an opportunity
for you to become familiar with a very widely used tool for simulating dynamic systems.
The deliverable outcome is to be an engineering technical report that clearly and concisely de-
tails the conduct and findings of your investigation and clear and pointed discussion of the tech-
nical and engineering significance of your findings. The report should address each of the
matters and questions listed in the scope of work in Section 3. Credit will be given for innova-
tive studies and analysis that either reveal other aspects of system performance or which im-
prove the performance of the system.
It is recommended that you follow the guidelines for writing technical engineering reports pro-
duced by The Institute of Engineering Technology and which are available at the following
web-site: https://www.theiet.org/media/5182/technical-report-writing.pdf
With reference to the above guidelines your report is expected to convey information to other
engineers about key aspects of the performance of the system and it is intended for selective
reading. The latter point means that you should organize your report into numbered sections
with informative headings.
It is strongly recommend that you approach this assignment in the same way as you would
as a professional engineer conducting the project for an employer or client.
Introductory analysis is presented in Section 1 to assist in the systematic formulation of the fre-
quency control model.
1 Introductory Analysis
The objective is to explore the performance of the system frequency control system as the mix
of generation sources is varied. We are interested in assessing performance for different propor-
tions of synchronous and asynchronous generation and how the performance of the overall fre-
quency control system performs as different proportions of generation capacity are equipped
with primary frequency control. The overall structure of the system model is shown in Figure
1. In this model the equivalent generator model represents the effective inertial response of the
system taking into consideration that a proportion of the generation is asynchronous. A propor-
tion of both the synchronous and asynchronous generation sources are not equipped with fre-
quency controls and therefore do not contribute to the control of system frequency. A proportion
of synchronous sources are equipped with governors which control the speed of their generators
and a proportion of asynchronous sources are also equipped with frequency controllers that are
used to regulate their power output so as to control system frequency.
Electric Power Systems © The University of Adelaide, 2025
Page: 3 of 13
Turbine P msp Pm
P msp 0
Governor
Equivalent
Synchronous P msu Generator
P msu f
Ungoverned 0 P L0
Generation P
Asynchronous P mau0 mau
P map0
Frequency P map
Control
Figure 1: Simplified model of system frequency control containing synchronous and
asynchronous sources. (Note that blocks with no input signal represent a fixed or constant
input to the model with the value shown in the block.)
1.1 Per-unit scaling and specification of generation capacities
We will use the total online generation capacity ( S b in MW) as the base value of power. Thus,
a power system load, P l , of 0.8 per-unit means that the system is consuming 80% of the rated
on-line generation capacity. (Note: We will see below that it is not necessary to specify the ac-
tual value of S b ).
To facilitate this exploration the total online generation capacity is divided into a number of
components.
S b = S sb + S ab (1)
where S sb and S ab are respectively the online synchronous and online non-synchronous (or
asynchronous) generation capacity.
S sb = sb S b and S ab = 1 – sb S b (2)
where sb is the specified proportion of the total online generation capacity which is synchro-
nous.
The online synchronous generation capacity is then partitioned into the following components:
S sb = S spb + S sub where (3)
S spb = spb S sb = spb sb S b and S sub = 1 – spb S sb = 1 – spb sb S b (4)
are respectively the proportion of online synchronous generation capacity with primary govern-
ing control and the balance of online synchronous generation capacity is ungoverned.
Electric Power Systems ©
Page: 4 of 13
The online asynchronous generation capacity is similarly partitioned into a fraction with prima-
ry governing control, S apb , and the balance of asynchronous generation which is ungoverned,
S aub , where:
S ab = S apb + S aub and (5)
S apb = apb S ab = apb 1 – sb S b
(6)
S aub = 1 – apb S ab = 1 – apb 1 – sb S b
Thus, to define the online capacities of the different types of generation represented in our mod-
el the parameters sb , spb and apb must be specified.
For example, if sb = 0.8 then 80% of all online generation capacity is synchronous and the
balance 1 – sb = 0.2 (i.e. 20%) is asynchronous. If spb = 0.4 then 40% of the online syn-
chronous generation capacity is under primary frequency control and the balance of such gen-
eration (i.e. 60%) is not under frequency control. In this example it means that
sb spb = 0.8 0.4 = 0.32 or just 32% of all online capacity is under primary frequency
control.
1.2 Specification of the initial generation output
In this model we specify the initial power output from each of the generation sources. On the
basis of this specification the initial system load is determined, on the assumption that the sys-
tem is lossless.
The initial power output from the online synchronous generation with primary speed control is
specified as a fraction sp of the specified online capacity S spb of this type of generation. The
initial power output is expressed in per-unit of S b .
S spb spb S sb
P msp0 = sp ---------- = sp ------------------ = sp spb sb in pu of S b (7)
Sb Sb
The initial output from the other types of generation are similarly specified as follows:
P msu0 = su 1 – spb sb
P map0 = ap apb 1 – sb pu on S b . (8)
P mau0 = au 1 – apb 1 – sb
The initial power output from the synchronous generation sources is:
P ms0 = P msp0 + P msu 0 (9)
and the initial power output from the asynchronous generation sources is:
P ma0 = P map 0 + P mau 0 (10)
Electric Power Systems ©T
Page: 5 of 13
and finally, the initial output from all generation sources is,
P m0 = P ms0 + P ma 0 . (11)
Thus, to define the initial steady-state operating point of the system it is necessary to specify
sp su ap and au .
It is important to distinguish between online capacity of the various generation categories and
the actual (initial) power output from each of these generation categories. Thus, if, for example,
sp = 0.5 then it means that initially the power output of the online synchronous generation
capacity that is under primary frequency control is 50%. If, furthermore, b = 0.8 and
spb = 0.4 it follows that the initial power output from all online synchronous generation ca-
pacity that is under primary frequency control is:
P msp0 = sp spb b = 0.5 0.8 0.4 = 0.16 pu of S b .
1.3 Equivalent generator rotor dynamics
The transfer-function block diagram of the equivalent generator of the system is shown in Fig-
ure 2.
Pm 1 f
----------
2Hs
Pl
DL
X
P l0
1
P l0
Figure 2: Equivalent generator rotor dynamics for system
load / frequency analysis.
1.3.1 Specification of the inertia constant
The inertia constant of online synchronous generation limits the rate of change of frequency. In
the model we specify the inertia constant, H s , in per-unit of the on-line synchronous generation
capacity. Typically, H s = 3.5 pu on S sb .
For use in the model this inertia constant must be converted to per-unit on S b . Thus,
S sb
H = H s ------- = sb H s (12)
Sb
Electric Power Systems ©
Page: 6 of 13
1.3.2 Frequency dependent load
The system load is linearly dependent on the system-frequency perturbation as follows
P l f = P l0 + P l0 1 + Df (13)
where f = 1 + f pu is the system frequency in per-unit of the system nominal synchronous
frequency, f 0 = 50 Hz, f is the per-unit perturbation of the system frequency and P l0 is the
initial steady-state value of the load. Note that under initial steady-state conditions the system
is operating at synchronous frequency, i.e. f 0 = 0 pu. P l0 is the initial steady-state load on
the system assuming that the system is initially operating at 50 Hz; and P l0 is the applied
change in system load at the nominal system frequency of 50 Hz
1.3.3 Composite rotor equation of motion for the analysis of system-frequency controls
The acceleration equation of the system is, as detailed in the lectures,
df
2H --------- = P m – P l in pu of S b (14)
dt
where P m and P l are respectively the generation and load, f is the per-unit system-frequency
perturbation and H is as defined in (12). The block diagram of this equation, including the fre-
quency dependence of the load, is shown in Figure 2.
1.4 Synchronous generator turbine-governor model
The block diagram of the synchronous generator turbine governor system employed in the as-
signment is shown in Figure 3. This block diagram shows the conversion from per-unit quanti-
ties on the system MVA base, S b , at the input and output from the model to per-unit quantities
on the turbine rated capacity, S spb , within the turbine-governor model. The block diagrams of
the governor and turbine models are shown respectively in Figures 4 and 5. The turbine model
is representative of a steam turbine which is presently the most commonly deployed turbine in
Australia.
f
1 R – Droop – pu on Sspb
---
R
P msp 0 P0
G P mt P msp
SYS2SYN
ug GOV TURB SYN2SYS
pu on Sb pu on Sspb pu on Sspb pu on Sb
Figure 3: Synchronous generator turbine-governor model including base conversions at the
turbine / governor model interfaces. (Note: Pref = Pmsp0 is the turbine load reference and is
represented as an external input. In Simulink such an external input can be represented as a
constant block).
Electric Power Systems ©
Page: 7 of 13
LR
ug 1 1 G
----- ---
Tg s
LL
Figure 4: Governor model (i.e. the block GOV in Figure 3).
F HP
P
G 1 1 1 mt
------------ ------------ ------------ F LP
sT CH sT RH sT CO
HP flow IP flow LP flow
F IP
FHP, FIP, FLP
turbine power
fractions.
Figure 5: Turbine model (i.e. the block TURB in Figure 3).
1.5 Simulink Model
A Simulink model and associated Matlab driver software of the system in Figure 1 is provided
as an input to the assignment as listed in Table 1. The provided model does not include the asyn-
chronous-generation frequency control system. A key task in this assignment is to implement
and assess the performance of an asynchronous control system.
Table 1: Simulink and Matlab files
File Description
FrequencyControlSystem.slx Simulink implementation of the frequency control sys-
tem in Figure 1 except that the asynchronous-genera-
tion frequency control system is omitted.
FrequencyControlSystem_SimDriver.m This Matlab m-file contains a Matlab function that can
be used and adapted to drive the Simulink model pro-
grammatically.
FCS_01.m This Matlab script demonstrates how to set up the fre-
quency control system model and to run a simulation
using the above driver.
2 Getting started with Simulink
The University makes Matlab and Simulink licences available to students as described at https:/
/www.adelaide.edu.au/technology/your-services/software/software-for-students#matlab
Electric Power Systems ©
Page: 8 of 13
Mathworks, the company that provides Matlab and Simulink, provide a number of getting start-
ed resources. The following may be useful:
• A listing of getting started resources is available at https://au.mathworks.com/help/sim-
ulink/getting-started-with-simulink.html
The assignment includes a Simulink model and associated Matlab driver software which can be
used as a good basis for developing skills in using Matlab and Simulink. Executing the matlab
script FCS_01.m from the matlab command prompt will run the base case simulation.
3 Scope of work
3.1 Base case parameters
The system parameters for the base case model are listed in Table 2. In subsequent sections pa-
rameters are specified as variations of this set of base case parameters. In this base case all on-
line generation is synchronous (i.e. there is no asynchronous generation). All online
synchronous generation capacity is governing and is loaded to 80% of its capacity.
Table 2: Base case model parameters
Parameter Value Parameter Value Parameter Value
Capacity fractions
sb 1.0 spb 1.0 apb 0.0
Load fractions
sp 0.8 ap 0.8
su 0.8 au 0.8
Equivalent generator parameters
Hs 3.5 D 1.0
Turbine / governor model parameters
R 0.05 T g (s) 0.3 LR (pu/s) 0.1
LL (pu/s) -1.0 T CH (s) 0.3 T RH (s) 7.0
T CO (s) 0.5 F HP 0.3 F IP 0.3
F LP 0.4
3.2 Ungoverned frequency response
Suppose the system is not equipped with any frequency control system. The change in frequen-
cy is then limited only by the frequency sensitivity of the loads.
Electric Power Systems ©
Page: 9 of 13
Setup the model with its base case parameters in Table 2 and then set spb = 0 so that there
will be no frequency control. Calculate the response of the system to a step increase in the load
of P L0 = 0.02 pu.
Plot and discuss the response of the system frequency. Conduct mathematical analysis to verify
that the initial rate of change of frequency and the final frequency obtained from the model re-
sponse is correct.
3.3 Synchronous only system with governors.
The objective in this section is to use the Simulink model provided to analyse the frequency con-
trol performance of the base case system and verify that its performance accords with theoretical
expectations.
Using the base case model perform a simulation in which a step increase in the load of
P L = 0.02 pu is applied.
Based on your knowledge of the steady-state behaviour of governors with a steady-state droop
characteristic what do you expect the final frequency deviation f f to be after the application
of the step-change in load. Ensure that you also include the frequency dependence of the system
load. How does your expectation compare with the final steady-state frequency from your Sim-
ulink study?
What is the initial rate-of-change-of-frequency (RoCoF) from your Simulink study? How does
this compare with what you expect from a mathematical analysis of the model? (Note that the
initial RoCoF due to a step-change in load is not influenced by the governor and turbine. Why?)
From the instant that the step is applied how long does it take for the system-frequency to settle
to within 5% of its final value. This time is referred to as the 5% settling time of the response.
If the final frequency deviation from nominal frequency is f f then the 5% settling time is de-
termined by the time taken for the frequency deviation to first fall within, and then remain with-
in, the frequency band f = 1 0.05 f f .
What is the maximum deviation in frequency f max and at what time t fmax does it occur after
the step is applied? (Note, f max may be positive or negative).
Plot the responses of the turbine power output (P mt ), the frequency deviation ( f ) and the gov-
ernor valve position ( G ).
3.4 Investigate factors that influence system frequency response – Governing by syn-
chronous generators only.
Analysis in this section is based on the simplified model of the primary frequency control sys-
tem of a power system in which frequency control is only performed with governors fitted to
synchronous generators. The factors that influence the system-frequency response are investi-
gated. Use, as a minimum, the following measures to assess the system-frequency response.
Electric Power Systems ©
Page: 10 of 13
1. The final value of the frequency deviation, f f .
2. The initial RoCoF.
3. The 5% settling-time of the system-frequency response.
4. The maximum frequency deviation ( f max ) and the time ( t fmax ), following the step, at
which it occurs.
In this investigation the emphasis is on comparing these measures as certain parameters are var-
ied.
The above performance measures are meaningful only if the system is stable. It is conceivable
that the system will be unstable for some combinations of model parameters. If so you should
clearly indicate in your report if a scenario is unstable and suggest reasons why.
At a minimum investigate the influence of the following factors on the system-frequency re-
sponse to step-changes in the system load.
1. Starting with the base case, investigate the effect of varying the synchronous machine
inertia constant, H s , within the range from 1.0 to 6.0 pu.s on S sb . Report your findings
with emphasis on those performance factors most influenced by the variation in inertia.
Explain the reasons for and engineering significance of your findings.
2. Starting with the base case, investigate the effect of varying the governor droop, R , within
the range from 1.0 to 8.0 % on S spb . Report your findings with emphasis on those perfor-
mance factors most influenced by the variation in droop. Explain the reasons for and engi-
neering significance of your findings.
3. Starting with the base case investigate the effect of reducing the proportion of online syn-
chronous generation capacity from 100% (base case) in steps of 25% to 25%. Report on
how this variation in capacity affects the system inertia constant H and the droop R when
the latter parameters are expressed in per-unit on the system MVA base S b . Adjust other
capacity and loading fractions to keep the total system load equal to the base case value of
0.8 pu on S b . Report your findings with emphasis on those performance factors most
influenced by the variation in online synchronous generation capacity. Relate your find-
ings in this section to your findings concerning the variation in inertia and the variation in
droop. Explain the reasons for and engineering significance of your findings.
Carefully consider how you present the results of your studies. It is important that you use tables
and figures that succinctly summarize the key findings of your studies. Importantly, overlaying
responses from multiple studies helps to visualize the sensitivity of the system responses to pa-
rameter variations. Similarly, summarizing in a single table the variation in performance met-
rics due to variation in system parameters aids interpretation.
You are encouraged to similarly investigate the effect on the system-frequency response of var-
ying other model parameters such as the following.
1. The size and direction of the step change in load.
2. The proportion of synchronous generation that is under primary frequency control.
3. The proportion of load supplied from synchronous sources.
Electric Power Systems ©
Page: 11 of 13
3.5 Modelling of asynchronous frequency control.
In this section you will extend the Simulink model to include a model of an asynchronous source
(SRC) that is equipped with a frequency controller (FC) as shown in Figure 6. The asynchro-
nous source is assumed to have a much faster response than the steam turbines assumed to sup-
ply the synchronous generators. The asynchronous source could be, for example, an
electrochemical battery.
Extend the Simulink model of the frequency control system to include the model shown in Fig-
ure 6. The base case parameters of the frequency control system following the inclusion of asyn-
chronous frequency control are listed in Table 3.
f
1
------ Ra – Droop – pu on Sapb
Ra
P map0 P a0 P ord Pa P map
S2A FC SRC A2S
uf
pu on Sb pu on Sapb pu on Sapb pu on Sb
Figure 6: Asynchronous generator frequency-control model including base conversions at the
frequency-control model interfaces. (Note: (i) S2A is the conversion factor to convert power
from a power base of S b to a base of S apb ; (ii) (ii) A2S is the inverse conversion factor; and
(iii) Pref = Pmap0 is the load reference of the asynchronous source and is represented as an
external input. In Simulink such an external input can be represented as a constant block).
Pmx
Rmx
uf 1 1 P ord
---- ---
Tf s
Rmn
Pmn
Figure 7: Frequency controller model (i.e. the block FC in Figure 6).
P ord 1 1 Pa
----- ---
Ts s
Figure 8: Asynchronous source model (i.e. the block SRC in Figure 6).
Electric Power Systems ©
Page: 12 of 13
Table 3: Base case model parameters with asynchronous-generation frequency control
included.
Parameter Value Parameter Value Parameter Value
Capacity fractions
sb 0.5 spb 1.0 apb 0.5
Load fractions
sp 0.8 ap 0.8
su 0.8 au 0.8
Equivalent generator parameters
Hs 3.5 D 1.0
Turbine / governor model parameters
R 0.05 T g (s) 0.3 LR (pu/s) 0.1
LL (pu/s) -1.0 T CH (s) 0.3 T RH (s) 7.0
T CO (s) 0.5 F HP 0.3 F IP 0.3
F LP 0.4
Asynchronous source / frequency controller model parameters
Ra 0.05 T f (s) 0.1 R mx (pu/s) 0.1
R mn (pu/s) -0.1 P mx (pu) 1.0 P mn (pu) 0.0
T s (s) 0.08
3.6 Investigate system frequency response of system with a mixture of synchronous
and asynchronous frequency controls.
The objective is to investigate the system frequency response of the system with a mixture of
synchronous and asynchronous frequency controls. The performance of the frequency response
is to be assessed using the measures that were employed in Section 3.4. You should design a set
of simulation tests that you expect to reveal interesting aspects of the frequency control perfor-
mance of the system. To assist you the following table lists an initial set of tests which you
should extend to consider additional factors. Discuss the results of each test with reference to
theoretical considerations and compare the responses obtained with different combinations of
parameters.
Electric Power Systems ©
Page: 13 of 13
Table 4: Initial tests for investigating frequency control performance.
(The Parameters column lists those parameters that are changed from Table 3)
Test Parameters Purpose
C01 sb = 1 , apb = 0 Base case with no asynchronous generation.
Should yield same performance as in Section
3.3
C02 Base case with mixture of synchronous and
asynchronous frequency control. Compare fre-
quency responses from C01 and C02 and relate
differences in behaviour to theory.
C03 spb = 0.5 Reduce fraction of synchronous generation
with governors enabled from 1.0 to 0.5. Com-
pare with C01 and C02.
C04 spb = 0.5 Starting with the case in C03 investigate how
varying the droop setting, Ra, of the asynchro-
nous governors over the range from 0.01 to
0.05 affects the final power output from the
synchronous an asynchronous sources.
C05 T f = 1.0 s Assess effect of increasing frequency controller
time constant from 0.1 to 1.0 s.
4 Factors considered in assessment
This assignment is structured as an engineering investigation and the assignment report will be
assessed in that context.
It is strongly recommend that you approach this assignment in the same way as you would
as a professional engineer conducting the project for an employer or client.
Thus assessment will consider:
(a) Completion of the scope of work.
(b) Report organization, brevity and clarity.
(c) Clarity and sophistication in explaining the engineering significance of the findings.
(d) Accurate, relevant and clear application of power system analysis principles to the
scope of work.
(e) Accuracy and correctness of results.
(f) Skill in application of the Matlab and Simulink programs.
Electric Power Systems ©
