Very Important: Question 1 - Simulation of Systems Using MATLAB

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Very Important

This coursework requires you to develop code comprising scripts, functions as well as Simulink files.
There should be a main file for the whole of question 1 and another for the whole of question 2.
Each main file should be calling functions to carry out the simulation of the system, the plotting and the
various calculations required.
Ideally the program should work for any second order system.

Question 1 Simulation of systems using MATLAB

1 a) (i) Give the general expression of a second order transfer function model
in the continuous-time domain in terms of natural frequency, =
2 , damping factor, = 1.25, and steady state gain, = 1 unit.
(7 marks)
(ii) Write down the corresponding differential equation. (3 marks)

b) (i) Simulate this system from 1a) (i) with a unit step input (using the step
function) and plot the system response in a well-documented figure
including legend, title, axis labels and units. (4 marks)
(ii) Determine the poles of the system mathematically and create a
diagram in the s-plane (e.g. pzmap function). (2 marks)
(iii) Using the s-plane pole zero map, explain how the poles affect the
transient response of the system. (9 marks)
Question 2 Control system design using MATLAB and Simulink

2 a) A second order system is represented by the differential equation


2 () ()
+5 + 4() = () (2.1)
2

(i) Write down the state equations and hence the vector-matrix representation
= + (2.2a)
= (2.2b)
(9 marks)
(ii) For the system represented in eq. (2.1), write down the transfer function:
()
= () (2.3)
()

(5 marks)
(iii) Show that the poles of the transfer function eq. (2.3) are identical to the
eigenvalues of the system matrix, , eq. (2.2a). (6 marks)

b) Implement the state space model from eq. (2.2) in Simulink, making use of the
phase variable state space form. Create an m-file script in MATLAB that: (9 marks)
(i) Initialises the Simulink diagram with the appropriate model parameters. (2 marks)
(ii) Simulates the system with a unit step input. (2 marks)
(iii) Creates suitably annotated gures with relevant information e.g. legend, title,
axis labels and units. (2 marks)

c) (i) Using the reaction curve method, determine a first order + lag approximation
of the system simulated in 2b). Use the step response to determine the first
order system parameters. This approximate model of the system will be used
in 2e).
(10 marks)
2 d) (i) Implement a PID control system, as depicted in Figure 2.1, of the plant
model () described in eq. (2.1). Use a unit step as the input. Due to a
constraint associated with the actuator in this system, the rising and falling
()
slew rate, = | |, of the control signal cannot exceed 100, where ()

is the control action.


(5 marks)

+ Actuator
SR<100
-

Figure 2.1 PID control loop

(ii) Tune the PID in order to minimise the rise time of the system with an
overshoot not exceeding 20% of steady state. Produce a table of PID gains
and provide a rationale for your choices, commenting on the effect of the
actuation constraint on the control action and the system response.
(10 marks)

e) This question builds on the closed loop control system implemented in 2d)
and the first order approximation of the system identified in 2c).
(i) Add a delay = 2 after the plant () in the open loop path of the
system implemented in 2d). Simulate the system and make an appraisal of
the PID closed loop response. (6 marks)
(ii) Implement and tune a Smith predictor using the approximated model to
represent the system model with the plant being represented by the
second order system (). Critically evaluate the performance of the Smith
predictor compared to that of the PID.
(9 marks)

END

You might also like