EE4218 / EE4216
Faculty of Science and Engineering
Department of Electronic and Computer
Engineering
End of Semester Assessment Paper
Module Code: EE4218 / EE4216
Module Title: Control 2
Semester: Spring 2018
Duration of Exam: 21
Grading Scheme: Final Exam : 80%
Coursework : 20%
Instructions to Candidates:
Answer any FOUR questions. All questions carry equal marks.
A University Standard Calculator may be used.
Module Code & Title: EE4218 / EE4216 Control 2 Page 2 of 6
MARKS
- (a) Consider the system
You may assume that β is a parameter that is allowed to vary in the range
β ∈ [0.25, . . . , 1.5]
i. Determine the natural frequency of oscillation ωn for this system. 2
ii. Write a MATLAB script (.m file) that will illustrate the effect of variation
in the parameter β in the region [0.25, . . . , 1.5] on a system step
response.
Your answer should briefly sketch the typical step responses that you
would expect to observe. 7
(b) This question concerns the locus of the roots of the characteristic equation
for the closed loop system described by Figure 1.
i. Determine equations for the asymptotes to the locus as K → ∞. 2
ii. What are the departure angles for the locus as it leaves the complex
open loop poles? 3
iii. Assume that one feasible multiple root (accurate to 1 decimal place)
exists for this locus at s = ?4.8 and that two infeasible multiple roots
exist at s = ?1.8 ± ?1.3.
Determine the other feasible multiple root location for this locus. 8
iv. Sketch the root locus for this system. 3
Figure 1:
Module Code & Title: EE4218 / EE4216 Control 2 Page 3 of 6
MARKS - This question considers the D.C Motor
System performance is to be improved using the following controller.
K(s) = 50(s + 3)(s + 7)
(a) Derive a bound on the Phase Margin of a system in terms of its worst case
system sensitivity kSk∞. 6
(b) Draw a system sensitivity plot, S(?ω), for this combination of plant and
controller.
Your answer should compute S(?ω) at
ω = [0, 14, 23, 50,∞] rad\s
You may assume that kSk∞ = ks(?23)k. 8
(c) Describe how bounds on Phase Margin, Steady state error, 5% settling
time and Bandwidth can be read from the S(?ω) you have computed in
Part (b) of this question. 6
(d) Estimate the 5% settling time for this design to a step input demand.
Briefly describe how you would use MATLAB to construct S(?ω) and to
observe the step response for this particular design.
Your answer should sketch an indicative step response for this system. 5
Module Code & Title: EE4218 / EE4216 Control 2 Page 4 of 6
MARKS - This questions considers Nyquist based tuning for the following system :-
2s + 8
(s + 1)(s + 3)((s + 2)2 + 5)
(a) Using the following frequency vectorω ∈ [0, 1.25, 3.3, . . . ,∞]
draw a Nyquist diagram for this system. 10
(b) Use your diagram to estimate the system phase and gain margins. 3
(c) Using your Nyquist diagram, design a PID controller, K(s), for this system
that is tuned for good low frequency performance. You may assume that
the following tuning rules are appropriate :-
KP = 0.4 × Ku
where Ku is the limit cycle gain with the application of proportional only
control and
TI = 1.0 × Tu
TD = 0.25 × Tu
where Tu is the corresponding limit cycle period. 7
(d) How would you improve the speed of response of your design using Phase
Lead methods?
Please note that an exact design is NOT required, it is sufficient to provide
the design steps that you would take. 5
Module Code & Title: EE4218 / EE4216 Control 2 Page 5 of 6
MARKS - This question considers the system shown in Figure 3. Denote the state variables
for this system as x1 = VC(t) x2 = iL(t). The following describing equations
for the system may be useful.
Vin(t) = u(t) = L
diL(t)dt + VC(t) ; Vout(t) = Vin(t) ?VC(t)
VC(t)R1+ CdVC(t)
dt = iL(t) + L
R2diL(t)
Figure 2:
(a) Briefly list TWO reasons for a state space approach to Control Systems
Design 2
(b) For the case where
L = 2H, R1 = 1?, R2 = 3?, C = 0.5F
Complete the continuous A state matrix for this system and hence write
the continuous state and output equation. 8
(c) Design a State feedback controller that will place the closed loop poles for
this system at ?2 ± 0.2?.
Your answer should provide a diagram illustrating how this design can be
implemented in practice. 8
(d) Determine the feasibility of an Observer based controller design strategy
for this system. 2
(e) Describe one advantage and one disadvantage of such an Observer based
strategy over a more traditional state feedback approach. 2
(f) Explain how such an Observer based strategy might be implemented in
practice 3
Module Code & Title: EE4218 / EE4216 Control 2 Page 6 of 6
MARKS - This question concerns the following system that you have studied extensively
during your Laboratory work this semester
(a) What type of problem does this system model and why is it such a difficult
problem to control? 3
(b) Determine the values of Proportional gain K that will stabilise this system.
Why is the proportional control that is suggested by Figure 3 unsuitable
for this system? 8
(c) Discuss the various approaches that you used to control this system over
the course of the semester. For each different strategy that you consider
your answer should describe
i. How you implemented the design in MATLAB.
ii. The frequency domain analysis tools that you used to assess the design.
iii. The steps that you took to improve the time domain performance of
your design. 14
WX：codehelp