CSTR
CSTR
CSTR
Where:
The dependence of to the concentration c and temperature T is often neglected so that the
density is assumed as constant.
The mass balance equation can be rewritten as follow:
State eq. (1)
Mass Balance on Component A
f
Where:
The state variables in this reactor are c
A
and V. It is defined that
. Thus mass
balance on component A
f
is given as follow:
or
Total Energy Balance
Assume that the work done by stirring machine is neglected so that the total energy in
reactor can be represented below:
where U is the internal energy, K the kinetic energy, and P the potential energy of reacting
mixture. Assuming that the reactor does not move so the total energy balance yields
For liquid system, an approximation below:
can be used to simplified the equation of total energy balance. Consider the input of total energy
balance per unit time is given as
where Q is the amount of heat released by the fluid in the jacket per unit time. The enthalpy is a
function of temperature and its composition so that:
State eq. (2)
where n
Af
and n
A
are the moles of A
f
and A in the CSTR. Thus the differential expression of the
enthalpy is given as:
Note that
where c
p
is the specific heat capacity of the reacting mixture and
and
Substitute the mass balance eq. of A to differential expression of enthalpy hence we obtain:
]
Subtititute dH/dt into the total energy balance equation then we get:
Note that
+
and
[
]
Consequently the equation of energy balance becomes:
Or,
)
Since (
, the
equation above can be rewritten as follow:
State eq. (3)
>> Linearization of Mathematical model of jacketed adiabatic CSTR
The presence of term
)
(
Substitute the linear approximation above to the state eq. (2) and (3) so that we obtain the new
form of state eq. as below:
For ) ( T T UA Q
j
= ,
)
) ( T T UA
j
If the T
0
and c
Af0
are steady-state conditions of CSTR and input value are c
Afi,0
, T
i,0
and T
j,0
. The
steady state form is given as:
and
State eq. (2a)
State eq. (3a)
State eq. (2b)
State eq. (3b)
Eliminate eq. (2a) and (3a) from (2b) and (3b) respectively and we obtain the new form:
*(
) (
)+
[(
)]
Lets define new dummy variable:
0
'
Af Af Af
c c c = and
0 ,
'
Afi Afi Afi
c c c =
0
' T T T = ,
0 ,
'
i i i
T T T = , and
0 ,
'
j j j
T T T =
Hence it is get the final linear model of state eq. (2) and (3)
and
)
>> State Space and Transfer Function of CSTR
Case 1
Dynamics equation of mass balance on component A (state eq. (2c)):
State Eq. (2c)
State Eq. (3c)
i Af
Af
Af
c
T e c e
dt
dc
,
) 16 . 298 )( 314 , 8 (
) 846 . 238 )( 843 , 11 (
2
) 16 . 298 )( 314 , 8 (
) 846 . 238 )( 843 , 11 (
' 0003 , 0
' .
) 16 . 298 )( 314 , 8 (
) 846 . 238 )( 843 , 11 (
825 , 14 ' . 825 , 14
3600
1
'
=
|
|
.
|
\
|
+
(
(
+ +
i Af Af
Af
c T c
dt
dc
,
' 0003 , 0 ' . 1813 , 0 ' . 7369 , 4
'
= + + (1.1)
Dynamics equation of derived total energy balance (state eq. (3c)):
( )
j i
Af
T T
c T
dt
dT
'
) 500 )( 3600 (
250
. 0003 , 0
' ) 7366 , 4 )( 43 , 10 ( '
) 500 )( 3600 (
250
) 1813 . 0 ).( 825 , 14 (
3600
1 '
=
+
(
+
j i Af
T T c T
dt
dT
' . 00014 , 0 . 0003 , 0 ' 4027 , 49 ' . 8914 , 1
'
= (1.2)
Based on eq. (1.1) and (1.2) above, we can derive the state space model as given below:
(
+
(
=
(
i
i Af Af Af
T
c
T
c
T
c
' 0003 , 0 0
0 0003 , 0
' 8914 , 1 4027 , 49
1813 , 0 7369 , 4
'
,
+
(
=
(
j
Af
T T
c
y
y
'
0
0 00014 , 0
0 0
' 1 0
0 1
2
1
Note that the input T
j
is considered as disturbance so that the input term is c
Af,i
and T
i
and the output is c
Af
and T.
(
=
(
=
0003 , 0 0
0 0003 , 0
8914 , 1 4027 , 49
1813 , 0 7369 , 4
B A
(
=
(
=
0 00014 , 0
0 0
1 0
0 1
D C
The transfer function of the system can be obtained by using MATLAB as shown
above. The transfer function of each pair input-output is given as follow:
) (
0027 , 0 8455 , 2
0006 , 0 0003 , 0
) ( ' ) (
1
2
1
s u
s s
s
s c s y
Af
+
= =
) (
0027 , 0 8455 , 2
0148 , 0 0004 , 0 0001 , 0
) ( ' ) (
2
2
2
2
s u
s s
s s
s T s y
+
+
= =
Output Component of A
Controllability
Lets transform the transfer function of y
1
(s)/u
1
(s) into state space model using
MATLAB then we obtain the state space matrix:
(
=
(
=
0
1
0 1
0027 , 0 8455 , 2
B A
| | 0 0006 ., 0 0003 , 0 = = D C
Controllability matrix then derived as below:
| |
(
=
1 0
8455 , 2 1
AB B
Since the determinant of matrix above isnt zero and the matrix is non-singular matrix.
Hence the system is controllable. The rank matrix is 2 as the order of matrix so that the
system is completely controllable.
Observability
The matrix of observability is given as follow:
| |
(
=
0 0006 . 0
0015 . 0 0003 . 0
* * * C A C
The determinant of controllability matrix is -9 x 10
-7
so that the matrix isnt singular and
the rank matrix is same to order of matrix. Thus, the system is observable.
Evaluation of pole zero of transfer function and Step Response
Based on transfer function y
1
(s)/u
1
(s), we get the zero is 2 and poles are -2,8464 and
0,0009. Plot zero pole in Argand diagram as follow:
There is pole that located in right half plane. It indicates that the response will be
unstable or grow/decay rapidly. The presence of zero in right half plane considers
causing the inverse response.
Step response for transfer function y
1
(s)/u
1
(s) is shown as figure in above that the
response is decay into negative amplitude so that it shows the inverse response since
the input is step signal having positive magnitude.
Output Temperature of fluid T
Controllability
Lets transform the transfer function of y
2
(s)/u
2
(s) into state space model using
MATLAB then we obtain the state space matrix:
-3 -2 -1 0 1 2
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
Real Part
I
m
a
g
i
n
a
r
y
P
a
r
t
0 500 1000 1500
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
Step Response
Time (sec)
A
m
p
l
i
t
u
d
e
(
=
(
=
0
1
0 1
0027 , 0 8455 , 2
B A
| | 0001 . 0 0148 . 0 0001 , 0 = = D C
Controllability matrix then derived as below:
| |
(
=
1 0
8455 , 2 1
AB B
Since the determinant of matrix above isnt zero and the matrix is non-singular matrix.
Hence the system is controllable. The rank matrix is 2 as the order of matrix so that the
system is completely controllable.
Observability
The matrix of observability is given as follow:
| |
(
=
0 0148 . 0
0151 . 0 00001 . 0
* * * C A C
The determinant of controllability matrix is -2.2348 x 10
-4
so that the matrix isnt
singular and the rank matrix is same to order of matrix. Thus, the system is observable.
Evaluation of pole zero of transfer function and Step Response
Based on transfer function y
1
(s)/u
1
(s), we get the zero are -14.3288 and 10.3288 also
poles are -2,8464 and 0,0009. Plot zero pole in Argand diagram as follow:
-15 -10 -5 0 5 10
-8
-6
-4
-2
0
2
4
6
8
Real Part
I
m
a
g
i
n
a
r
y
P
a
r
t
There is pole that located in right half plane. It indicates that the response will be
unstable or grow/decay as in the mass of component A. The presence of zero in right
half plane also considers causing the inverse response.
Step response for transfer function y
2
(s)/u
2
(s) is shown as figure in above that the
response is growing slowly. The inverse response due to the term of zero in right half
plane isnt seen in the step response as the slowly response of system.
0 500 1000 1500
-500
-400
-300
-200
-100
0
100
200
300
400
500
Step Response
Time (sec)
A
m
p
l
i
t
u
d
e