0% found this document useful (0 votes)
17 views10 pages

QP 2014 Exam

Engineering Exam

Uploaded by

brightbunny99
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
17 views10 pages

QP 2014 Exam

Engineering Exam

Uploaded by

brightbunny99
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

SCHOOL OF ENGINEERING

EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC

DURATION: 3 HOURS
TOTAL MARKS: 100

INTERNAL EXAMINER(S): PROF M. MULHOLLAND


DR D. LOKHAT

INTERNAL MODERATOR(S): DR S. MOODLEY

EXTERNAL MODERATOR: DR D. LOVE

INSTRUCTIONS:

1. ALL questions may be attempted.

2. Calculations must be shown in sufficient detail to illustrate your understanding of


the method.

3. Calculators must be cleared of any information that would subvert the purpose of
the examination.

4. Automatic solutions and matrix manipulations by calculator may not be used – all
steps of such calculations must me shown.

5. Write your student number on this question paper, which must be handed in
before you leave the examination room.

6. QUESTIONS 1 to 6:
(a) Your answer must be written in the space provided with each question.
(b) Only answers in the appropriate space will be considered.
(c) It is recommended that you work in dark pencil, using an eraser for
corrections.

7. QUESTIONS 7 to 10 are to be answered in the blank answer book provided. The


plot for QUESTION 9 (see page 9) is to be removed from the question paper
and returned within your answer book

Seat Number:
Student Number:
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 1 of 9

1 The fixed bed catalytic reactor shown below


has gaseous reagents passing through it by plug
flow. In terms of modelling it is thus a
distributed system.

As indicated, you are going to do a crude


lumped model approximation by dividing the
bed into two equal mixed-flow portions.

You need to account for the following:


* The original bed gas volume is V
* The reaction is 2A→B with the rate of
consumption of A given by
( −rA ) =
kC A2 per unit volume of gas,
with CA the gas-phase concentration of A.
* Both the gas feed flow-rate F(t) and the feed
concentration CA0(t) are time-variant.
* The system operates isothermally
* You do not need to account for flow
variation through the reactor due to
consumption of A

(a) Derive the system of non-linear ODE’s 3


required to describe the exit concentration
of A, as dependent on the feed rate and
feed concentration.

(b) Linearise your equations from (a) in order


to represent the system in the state-space 5
vector-matrix form
dx
= Ax + Bu
dt

[8]
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 2 of 9

2 In its bid to reduce water leakage losses in the


water distribution system, Durban Metro have
installed pressure reduction valves on a number
of trunk mains.

PC
100 m

consumers

Pat

PC

from reservoir to consumers


As seen above, the pressure reduction valve
uses its downstream pressure to close the valve
as this pressure rises, ie. x = 1 − α PC where PC
is obviously a gauge pressure

A particular reservoir is at a height of 100 m


(providing 10 barg), and supplies through a
long pipeline to the pressure reduction valve
and consumers. The pressure reduction valve
has CV=300 [m3h-1][bar]-1/2, and is set to shut
when the consumer pressure PC reaches 6 barg.

At 12.00 midday the flow through the pipeline


to consumers is 200 m3h-1, giving a PC value of
4 barg.

Twelve hours later, at midnight, the consumer


demand has dropped to 100 m3h-1.

Required: Find the consumer supply pressure


PC at midnight (obtain a numerical equation in
which PC is the only unknown, but do not
attempt to solve it)

[Treat the valve as linear]

[10]
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 3 of 9

3 A sulphuric acid plant with a 3-bed SO2 converter is shown. Sulphur is burned in an air stream and the
resultant SO2, O2, and N2 mixture is fed to the first bed. An inter-changer cools gas proceeding to the
second bed. After the second bed, some of the product SO3 is absorbed into a circulating sulphuric acid
solution (dashed lines) in an inter-absorber. This gas stream is then re-heated before entering the third
bed. Gas from the third bed passes through the final absorber before proceeding to the stack.

Using the valves shown on the diagram, add schemes to achieve the following:

(a) Control the feed temperature to each bed {11/2}


(b) Control the temperatures of the cooled gas streams fed to the absorbers {1}
(c) Control plant pressure using the stack valve {1/2}
(d) Control level and composition of the acid sump {2}
(e) Control the flow-rate of recirculating acid to each absorber {1}
(f) Control the flow-rate of molten sulphur feed to the burner {1/2}
(g) Ratio the air flow-rate to the molten sulphur flow-rate {1}
(h) Keep the first bed exit temperature below TMAX by overriding the upstream BFW valve {1}
(i) Use valve position control to minimise the plant pressure {11/2}

inter-
absorber SO2 O2 N2

to
bed 1 st.
stack
BFW
burner

bed 2
steam

final BFW
steam
absorber bed 3

BFW air

make-up water

conc. sulphuric acid


acid sump
molten sulphur

[10]
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 4 of 9

4 (a) Consider the following objective function


3
used in DMC:
J ( ∆m =
) ( eOL + B∆m ) W ( eOL + B∆m ) + ∆mT Λ ∆m
T

Explain the meaning/purpose of each term and


each variable

(b) Draw a block flow diagram to represent the


functioning of a Dynamic Matrix Controller. 3
Annotate the signal lines with appropriate
variable names, and show the relevant
calculations in the boxes.

(c) Draw a similar block flow diagram to


represent a Smith Predictor used in a control 4
loop, and briefly discuss similarities to (a) [2
lines]

[10]
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 5 of 9

5 A “cross-limiting” furnace control scheme is


shown for the boiler below:

(a) Explain in sequence all of the events /


actions which will occur in this control scheme
5
if the steam users suddenly start to draw more
steam from the drum.

(b) What is the benefit of the scheme?


1
[6]

6 A stirred vessel is used to correct the


concentrations CA and CB of a water stream, by
adding water flows F1 and F2:
F1 F2
CA1= 3 CA2= 0
CB1= 0 CB2= 3

F0 = 1
CA0= 0 F
CB0= 0 V=1 CA
CB
at steady-state:
CASS = 1
CBSS = 1

This process can be represented by

d  C A′   −3 0   C A′   +2 −1  F1′
=    +  −1 +2   F ′ 
dt  CB′   0 −3  CB′    2 

where CA’,CB’, F1’,F2’ are deviations from the


normal steady-state.

Obtain the steady-state relationship between the


input and output deviations. Now explain
briefly (DO NOT CALCULATE) how to
obtain the
(i) RELATIVE GAIN ARRAY and
(ii) MORARI RESILIENCY INDEX
Indicate what these results can be used for, and
note any shortcomings.
[6]
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 6 of 9

QUESTION 7

A toxic gas sensor is installed in a duct through which gas at a rate of 1 m3·s-1 flows. The sensor
exhibits first-order dynamics with a time constant of 5 s and a steady-state gain of 1. The alarm
level on the sensor is 4% (on a volume basis). The concentration of toxic gas in the duct is 1% for
a long period of time. At a particular time, a step change in the toxic gas concentration in the duct
from 1% to 7% occurs. Determine the total volume of gas that passes the sensor before the alarm
activates.

L[dx/dt]=sX(s)
L [u(t)] = 1/s
L [e-bt] = 1/(s+b)
TOTAL /10/

QUESTION 8

A particular system consists of three non-interacting first-order processes connected in series. The
processes have the following transfer functions:
1
GP1( s ) =
s+4
1
GP 2( s ) =
s+3
1
GP 3( s ) =
s+2
A PI controller with an integral time constant of 0.2 time units is to be used to control this system.

a) Draw the appropriate root locus diagram (NOTE: for parts of the method that might require
an iterative solution, simply give the necessary equations and assume a result).
(11)

b) Determine the controller gain Kc above which the system will become closed-loop unstable
(use a value directly from your own diagram).
(3)
TOTAL /14/
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 7 of 9

QUESTION 9

A process consists of two mixing vessels with a function to ensure thorough mixing of an input
stream and a makeup stream which is added in order to keep the output flow at a constant
composition. It is known that the input stream will always be in error toward the lean side in
component 2.

Make-up
(component 2) Composition
controller

AC

Input (component
1 + component 2)

Composition
analyzer
AT

Output
Mixer 1 Mixer 2

Each mixer exhibits first-order dynamics with unity gains and different time constants. The
arrangement is non-interacting. The time constants are τp1 = 5 min and τp2 = 10 min. A time delay
of 0.1 min is associated with the connecting piping. A proportional feedback controller with a
gain of 3 has been specified for the control loop. The dynamics of the control valve and
measurement transmitter may be ignored. Plot the openloop Bode diagram for this system using
the specified controller gain (on the blank graph provided). Will the closed loop be stable? What
is the gain margin and phase margin of the system?
TOTAL /14/
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 8 of 9

QUESTION 10

Consider a process with the following transfer function:

1
GP ( s ) =
s( s + 1)( s + 2)

A proportional controller with Kc = 1 is used to control the process.

a) Using directed line segments, sketch the openloop Nyquist diagram for the system.
(10)

b) Show on your diagram how you would determine the phase margin and gain margin for this
system (numerical values are not required).
(2)
TOTAL /12/
UNIVERSITY OF KWAZULU-NATAL, EXAMINATIONS: JUNE 2014

COURSE AND CODE: PROCESS DYNAMICS & CONTROL ENCH4DC


PAGE: 9 of 9

BODE PLOT FOR QUESTION 9

(i) COMPLETE: STUDENT NUMBER :

(ii) REMOVE THIS PAGE FROM THE QUESTION PAPER AND RETURN IT INSIDE
YOUR ANSWER BOOK

Bode Plot
20

0
Magnitude (dB)

-20

-40

-60

-80
-3 -2 -1 0 1
10 10 10 10 10

-45
Phase (degrees)

-90

-135

-180

-225

-270
-3 -2 -1 0 1
10 10 10 10 10
Frequency - log scale

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy