Você está na página 1de 19

7-1

1234567898




7.1


In the absence of more accurate data, use a first-order transfer function as


'( )
'( ) 1
s
i
T s Ke
Q s s

+


o
( ) (0) (124.7 120) F
0.118
540 500 gal/min
i
T T
K
q




= 3:09 am 3:05 am = 4 min

Assuming that the operator logs a 99% complete system response as no
change after 3:34 am, 5 time constants elapse between 3:09 and 3:34 am.

5 = 3:34 min 3:09 min = 25 min
= 25/5 min = 5 min

Therefore,


4
'( ) 0.188
'( ) 5 1
s
i
T s e
Q s s

+


To obtain a better estimate of the transfer function, the operator should log
more data between the first change in T and the new steady state.


7.2

Process gain,
2
(5.0) (0) (6.52 5.50) min
0.336
30.4 0.1 ft
i
h h
K
q




a) Output at 63.2% of the total change

= 5.50 + 0.632(6.52-5.50) = 6.145 ft

Interpolating between h = 6.07 ft and h = 6.18 ft

Solution Manual for Process Dynamics and Control, 2
nd
edition,
Copyright 2004 by Dale E. Seborg, Thomas F. Edgar and Duncan A. Mellichamp.
7-2

(0.8 0.6)
0.6 (6.145 6.07) min 0.74min
(6.18 6.07)


b)


0
(0.2) (0) 5.75 5.50 ft ft
1.25
0.2 0 0.2 min min
t
dh h h
dt




Using Eq. 7-15,


0 t
KM
dh
dt


_

,
min 84 . 0
25 . 1
) 1 . 0 4 . 30 ( 347 . 0



c) The slope of the linear fit between t
i
and
1
]
1


) 0 ( ) (
) 0 ( ) (
1 ln
h h
h t h
z
i
i
gives an
approximation of (-1/) according to Eq. 7-13.

Using h() = h(5.0) = 6 .52, the values of z
i
are

t
i
z
i
t
i
z
i

0.0 0.00 1.4 -1.92
0.2 -0.28 1.6 -2.14
0.4 -0.55 1.8 -2.43
0.6 -0.82 2.0 -2.68
0.8 -1.10 3.0 -3.93
1.0 -1.37 4.0 -4.62
1.2 -1.63 5.0 -

Then the slope of the best-fit line, using Eq. 7-6 is

2
13 1
13 ( )
tz t z
tt t
S S S
slope
S S

_



,
(1)

where the datum at t
i
= 5.0 has been ignored.

Using definitions,

0 . 18
t
S 4 . 40
tt
S
5 . 23
z
S 1 . 51
tz
S

Substituting in (1),


1
1.213
_

,
0.82min
7-3
d)

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
5.4
5.6
5.8
6
6.2
6.4
6.6
6.8
Experimental data
Model a)
Model b)
Model c)

Figure S7.2. Comparison between models a), b) and c) for step response.



7.3


a)

1 1
1
( )
( ) 1
T s K
Q s s

+

2 2
1 2
( )
( ) 1
T s K
T s s

+


2
2 1 2 1 2
1 2 1
( )
( ) ( 1)( 1) ( 1)
s
T s K K K K e
Q s s s s


+ + +
(1)

where the approximation follows from Eq. 6-58 and the fact that
1
>
2
as
revealed by an inspection of the data.

667 . 2
82 85
0 . 10 0 . 18 ) 0 ( ) 50 (
1 1
1

q
T T
K
75 . 0
0 . 10 0 . 18
0 . 20 0 . 26
) 0 ( ) 50 (
) 0 ( ) 50 (
1 1
2 2
2

T T
T T
K

Let z
1
, z
2
be the natural log of the fraction incomplete response for T
1
,T
2
,
respectively. Then,
7-4
1
]
1

1
]
1

8
) ( 18
ln
) 0 ( ) 50 (
) ( ) 50 (
ln ) (
1
1 1
1 1
1
t T
T T
t T T
t z
1
]
1

1
]
1

6
) ( 26
ln
) 0 ( ) 50 (
) ( ) 50 (
ln ) (
2
2 2
2 2
2
t T
T T
t T T
t z
A graph of z
1
and z
2
versus t is shown below. The slope of z
1
versus t line
is 0.333 ; hence (1/-
1
)=-0.333 and
1
=3.0

From the best-fit line for z
2
versus t, the projection intersects z
2
= 0 at
t1.15. Hence
2
=1.15.


1 3
667 . 2
) ( '
) ( '
1
+

s s Q
s T
(2)

1 15 . 1
75 . 0
) ( '
) ( '
1
2
+

s s T
s T
(3)
Figure S7.3a. z
1
and z
2
versus t

b) By means of Simulink-MATLAB, the following simulations are obtained

0 2 4 6 8 10 12 14 16 18 20 22
10
12
14
16
18
20
22
24
26
28
time
T
1


,



T
2
T
1
T
2
T
1
(experimental)
T
2
(experimental)

Figure S7.3b. Comparison of experimental data and models for step change
-8.0
-7.0
-6.0
-5.0
-4.0
-3.0
-2.0
-1.0
0.0
0 5 10 15 20
time,t
z
1
,
z
2
7-5
7.4


s s s s
s X s G s Y
5 . 1
) 1 )( 1 3 )( 1 5 (
2
) ( ) ( ) (
+ + +


Taking the inverse Laplace transform

( ) -75/8*exp(-1/5*t)+27/4*exp(-1/3*t)-3/8*exp(-t)+3 y t (1)

a) Fraction incomplete response

1
]
1


3
) (
1 ln ) (
t y
t z

Figure S7.4a. Fraction incomplete response; linear regression

From the graph, slope = -0.179 and intercept 3.2

Hence,

-1/ = -0.179 and = 5.6

= 3.2

3.2
2
( )
5.6 1
s
e
G s
s

+


b) In order to use Smiths method, find t
20
and t
60


y(t
20
)= 0.2 3 =0.6
y(t
60
)= 0.6 3 =1.8

Using either Eq. 1 or the plot of this equation, t
20
= 4.2 , t
60
= 9.0
Using Fig. 7.7 for t
20/
t
60
= 0.47
= 0.65 , t
60
/= 1.75, and = 5.14
z(t) = -0.1791 t + 0.5734
-9.0
-8.0
-7.0
-6.0
-5.0
-4.0
-3.0
-2.0
-1.0
0.0
0 10 20 30 40 50
time,t
z
(
t
)
7-6

1 68 . 6 4 . 26
2
) (
2
+ +

s s
s G

The models are compared in the following graph:

0 5 10 15 20 25 30 35 40
0
0.5
1
1.5
2
2.5
time,t
y
(
t
) Third-order model
First order model
Second order model

Figure S7.4b. Comparison of three models for step input


7.5


The integrator plus time delay model is
G(s)
s
K
e
s


In the time domain,
y(t) = 0 t < 0
y(t)= K (t-) t 0

Thus a straight line tangent to the point of inflection will approximate the
step response. Two parameters must be found: K and (See Fig. S7.5 a)

1.- The process gain K is found by calculating the slope of the straight
line.
K = 074 . 0
5 . 13
1


2.- The time delay is evaluated from the intersection of the straight line
and the time axis (where y = 0).
= 1.5
7-7
Therefore the model is G(s) =
s
e
s
5 . 1
074 . 0




Figure S7.5a. Integrator plus time delay model; parameter evaluation

From Fig. E7.5, we can read these values (approximate):

Time Data Model
0 0 -0.111
2 0.1 0.037
4 0.2 0.185
5 0.3 0.259
7 0.4 0.407
8 0.5 0.481
9 0.6 0.555
11 0.7 0.703
14 0.8 0.925
16.5 0.9 1.184
30 1 2.109

Table.- Output values from Fig. E7.5 and predicted values by model

A graphical comparison is shown in Fig. S7.5 b

Figure S7.5b. Comparison between experimental data and integrator plus time
delay model.
Slope = KM
y(t)

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20 25 30
Time
O
u
t
p
u
t
Experimental data
Integrator plus time delay model
7-8
7.6


a) Drawing a tangent at the inflection point which is roughly at t 5, the
intersection with y(t)=0 line is at t 1 and with the y(t)=1 line at t 14.
Hence =1 , = 141=13

1 13
) (
1
+

s
e
s G
s


b) Smiths method

From the graph, t
20
= 3.9 , t
60
= 9.6 ; using Fig 7.7 for t
20/
t
60
= 0.41

= 1.0 , t
60
/= 2.0 , hence = 4.8 and
1
=
2
= = 4.8


2
) 1 8 . 4 (
1
) (
+

s
s G

Nonlinear regression

From Figure E7.5, we can read these values (approximated):









Table.- Output values from Figure E7.5

In accounting for Eq. 5-48, the time constants were selected to minimize
the sum of the squares of the errors between data and model predictions.
Use Excel Solver for this Optimization problem:

1
=6.76 and
2
= 6.95

)) 1 76 . 6 )( 1 95 . 6 (
1
) (
+ +

s s
s G

The models are compared in the following graph:

Time Output
0.0 0.0
2.0 0.1
4.0 0.2
5.0 0.3
7.0 0.4
8.0 0.5
9.0 0.6
11.0 0.7
14.0 0.8
17.5 0.9
30.0 1.0
7-9
0 5 10 15 20 25 30 35 40 45 50
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
time
O
u
t
p
u
t
Non linear regression model
First-order plus time delay model
Second order model (Smith's method)

Figure S7.6. Comparison of three models for unit step input



7.7


a) From the graph, time delay = 4.0 min

Using Smiths method,

from the graph,
20
5.6 t + ,
60
9.1 t +

6 . 1
20
t , 1 . 5
60
t , 314 . 0 1 . 5 / 6 . 1 /
60 20
t t

From Fig.7.7 , 1.63 ,
60
/ 3.10 t , 1.645

Using Eqs. 5-45, 5-46,
1
4.81 ,
2
0.56


b) Overall transfer function


4
1 2
10
( )
( 1)( 1)
s
e
G s
s s

+ +
,
1 2
>

Assuming plug-flow in the pipe with constant-velocity,

7-10
( )
p
s
pipe
G s e

,
3 1
0.1min
0.5 60
p


Assuming that the thermocouple has unit gain and no time delay


2
1
( )
( 1)
TC
G s
s

+
since
2 1
<<

Then


3
1
10
( )
( 1)
s
HE
e
G s
s

+
, so that

3
0.1
1 2
10 1
( ) ( ) ( ) ( ) ( )
1 1
s
s
HE pipe TC
e
G s G s G s G s e
s s

_ _


+ +
, ,




7.8


a) To find the form of the process response, we can see that

2
( ) ( )
( 1) ( 1) ( 1)
K K M K M
Y s U s
s s s s s s s

+ + +


Hence the response of this system is similar to a first-order system with a
ramp input: the ramp input yields a ramp output that will ultimately cause
some process component to saturate.

b) By applying partial fraction expansion technique, the domain response for
this system is

Y(s) =
2
1
A B C
s s s
+ +
+
hence y(t) = -KM + KMt KMe
-t/

In order to evaluate the parameters K and , important properties of the
above expression are noted:

1.- For large values of time (t>>) , y(t) ( ) y t = KM (t-)
2.- For t = 0, (0) y = KM

These equations imply that after an initial transient period, the ramp input
yields a ramp output with slope equal to KM. That way, the gain K is
7-11
obtained. Moreover, the time constant is obtained from the intercept in
Fig. S7.8


Figure S7.8. Time domain response and parameter evaluation




7.9


For underdamped responses,

'

1
1
]
1

,
_

,
_




t t e KM t y
t
2
2
2
/
1
sin
1
1
cos 1 ) ( (5-51)

a) At the response peaks,

2 2
/
2
1 1
cos sin
1
t
dy
KM e t t
dt

1 _ _

1 +
'


1

, , ]



2 2 2
/
1 1 1
sin cos 0
t
e t t

1 _ _

1 +
;


1

, , ]


Since KM 0 and 0
/

t
e

,
_

,
_

,
_

,
_

t t
2 2
2
2 2
1
sin
1
1
1
cos 0

Slope = KM
y(t)
7-12

,
_


n t sin
1
sin 0
2
, t
2
1

n
where n is the number of peak.

Time to the first peak,
2
1

p
t

b) Graphical approach:

Process gain,


( ) (0) 9890 9650 lb
80
hr
95 92
psig
D D
w w
K
Ps





Overshoot = 333 . 0
9650 9890
9890 9970

b
a


From Fig. 5.11, 0.33


t
p
can be calculated by interpolating Fig. 5.8

For 0.33 , t
p
3.25

Since t
p
is known to be 1.75 hr , = 0.54



2 2 2
80
( )
2 1 0.29 0.36 1
K
G s
s s s s

+ + + +



Analytical approach

The gain K doesnt change:
lb
80
hr
psig
K
To obtain the and values, Eqs. 5-52 and 5-53 are used:

Overshoot = 333 . 0
9650 9890
9890 9970

b
a
= exp(-/(1-
2
)
1/2
)

Resolving, = 0.33

7-13

2
1.754 hence 0.527 hr
1
p
t



2 2 2
80
( )
2 1 0.278 0.35 1
K
G s
s s s s

+ + + +



c) Graphical approach

From Fig. 5.8, t
s
/ = 13 so t
s
= 2 hr (very crude estimation)


Analytical approach

From settling time definition,

y = t 5% KM so 9395.5 < y < 10384.5


(KM t 5% KM) = KM[ 1-e
(-0.633)
[cos(1.793t
s
)+0.353sin(1.793t
s
)]]


1 t 0.05 = 1 e
(0.633 ts)
cos(1.793 t
s
) + 0.353e
(-0.633 ts)
sin(1.7973 t
s
)



Solve by trial and error t
s
6.9 hrs



7.10


a)
2 2
'( )
'( ) 2 1
T s K
W s s

+ +


o
( ) (0) 156 140 C
0.2
80 Kg/min
T T
K
w



From Eqs. 5-53 and 5-55,

Overshoot = 344 . 0
140 156
156 5 . 161

b
a
= exp(- (1-
2
)
1/2

By either solving the previous equation or from Figure 5.11, = 0.322
(dimensionless)
7-14
There are two alternatives to find the time constant :

1.- From the time of the first peak, t
p
33 min.

One could find an expression for t
p
by differentiating Eq. 5-51 and
solving for t at the first zero. However, a method that should work
(within required engineering accuracy) is to interpolate a value of
=0.35 in Figure 5.8 and note that t
p
/ 3

Hence min 10 5 . 9
5 . 3
33


2.- From the plot of the output,

Period =
2
2
1
P

= 67 min and hence =10 min



Therefore the transfer function is


1 44 . 6 100
2 . 0
) ( '
) ( '
) (
2
+ +

s s s W
s T
s G

b) After an initial period of oscillation, the ramp input yields a ramp output
with slope equal to KB. The MATLAB simulation is shown below:

0 10 20 30 40 50 60 70 80 90 100
140
142
144
146
148
150
152
154
156
158
160
time
O
u
t
p
u
t

Figure S7.10. Process output for a ramp input

7-15
We know the response will come from product of G(s) and X
ramp
= B/s
2

Then
2 2 2
( )
( 2 1)
KB
Y s
s s s

+ +

From the ramp response of a first-order system we know that the response
will asymptotically approach a straight line with slope = KB. Need to find
the intercept. By using partial fraction expansion:

3 4 1 2
2 2 2 2 2 2
( )
( 2 1) 2 1
s KB
Y s
s s s s s s s
+
+ +
+ + + +


Again by analogy to the first-order system, we need to find only
1
and

2
. Multiply both sides by s
2
and let s 0,
2
= KB (as expected)

Cant use Heaviside for
1
, so equate coefficients

2 2 2 2 3 2
1 2 3 4
( 2 1) ( 2 1) KB s s s s s s s + + + + + + +

We can get an expression for
1
in terms of
2
by looking at terms
containing s.

s: 0 =
1
+
2
2
1
= -KB2

and we see that the intercept with the time axis is at t = 2. Finally,
presuming that there must be some oscillatory behavior in the response,
we sketch the probable response (See Fig. S7.10)



7.11


a) Replacing by 5, and K by 6 in Eq. 7-34

/ 5 / 5
( ) ( 1) [1 ]6 ( 1)
t t
y k e y k e u k

+

b) Replacing by 5, and K by 6 in Eq. 7-32

( ) (1 ) ( 1) 6 ( 1)
5 5
t t
y k y k u k

+

In the integrated results tabulated below, the values for t = 0.1 are shown
only at integer values of t, for comparison.


7-16
t
y(k)
(exact)
y(k)
(1t=1)
y(k)
(1t=0.1)
0 3 3 3
1 2.456 2.400 2.451
2 5.274 5.520 5.296
3 6.493 6.816 6.522
4 6.404 6.653 6.427
5 5.243 5.322 5.251
6 4.293 4.258 4.290
7 3.514 3.408 3.505
8 2.877 2.725 2.864
9 2.356 2.180 2.340
10 1.929 1.744 1.912

Table S7.11. Integrated results for the first order differential equation

Thus t = 0.1 does improve the finite difference model bringing it closer
to the exact model.



7.12

To find
1
a and
1
b , use the given first order model to minimize


10
2
1 1
1
( ( ) ( 1) ( 1))
n
J y k a y k b x k


0 ) 1 ( ))( 1 ( ) 1 ( ) ( ( 2
1 1
10
1 1

k y k x b k y a k y
a
J
n


10
1 1
1
1
2( ( ) ( 1) ( 1))( ( 1)) 0
n
J
y k a y k b x k x k
b



Solving simultaneously for
1
a and
1
b gives


10 10
1
1 1
1 10
2
1
( ) ( 1) ( 1) ( 1)
( 1)
n n
n
y k y k b y k x k
a
y k



10 10 10 10
2
1 1 1 1
1 2
10 10 10
2 2
1 1 1
( 1) ( ) ( 1) ( 1) ( 1) ( 1) ( )
( 1) ( 1) ( 1) ( 1)
n n n n
n n n
x k y k y k y k x k y k y k
b
x k y k y k x k


_


,




7-17
Using the given data,

212 . 35 ) ( ) 1 (
10
1

k y k x
n
, 749 . 188 ) ( ) 1 (
10
1

n
k y k y

14 ) 1 (
10
1
2

n
k x , 112 . 198 ) 1 (
10
1
2

n
k y

409 . 24 ) 1 ( ) 1 (
10
1

n
k x k y

Substituting into expressions for
1
a and
1
b

gives


1
a = 0.8187 ,
1
b = 1.0876

Fitted model is ) ( 0876 . 1 ) ( 8187 . 0 ) 1 ( k x k y k y + +

or ) 1 ( 0876 . 1 ) 1 ( 8187 . 0 ) ( + k x k y k y (1)

Let the first-order continuous transfer function be


( )
( ) 1
Y s K
X s s

+


From Eq. 7-34, the discrete model should be


/ /
( ) ( 1) [1 ] ( 1)
t t
y k e y k e Kx k

+ (2)


Comparing Eqs. 1 and 2, for t=1, gives

= 5 and K = 6

Hence the continuous transfer function is 6/(5s+1)
7-18

0 1 2 3 4 5 6 7 8 9 10
2
3
4
5
6
7
8
time,t
y
(
t
)
actual data
fitted model

Figure S7.12. Response of the fitted model and the actual data



7.13


To fit a first-order discrete model

) 1 ( ) 1 ( ) (
1 1
+ k x b k y a k y

Using the expressions for
1
a and b
1
from the solutions to Exercise 7.12,
with the data in Table E7.12 gives

918 . 0
1
a , 133 . 0
1
b

Using the graphical (tangent) method of Fig.7.5 .

1 K , 0.68 , and 6.8

The response to unit step change for the first-order model given by

0.68
6.8 1
s
e
s

+
is
8 . 6 / ) 68 . 0 (
1 ) (


t
e t y
7-19

Figure S7.13- Response of the fitted model, actual data and graphical method




0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0 2 4 6 8 10 time,t
y
(
t
)
actual data
fitted model
graphical method