Residence time behaviour

theoretical mean residence time:

VL
tR 
Q

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 1
tracer experiment to estimate the residence time behaviour

tracer: statistic approach:

 good soluble  
 easy to detect A G  Sdt A G   Si  t
 not adsorbal 0 0
 innert t
S( t)
E( t)= F (t)=∫ E(t) dt
for example: AG −∞

 salt   t Si t
 colour t R  tE( t)dt tR 
 isotops 0
 Si t
 trace elements
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 2
 statistic approach:
∞ 2
2 ∑ S t Δt
∫Ett 2
σ =
i
- tR
2

σ 2 =∫ E  t  t −t R 2 =
0
−t 2R
∑ S i Δt
AG 2 2
N 
2
tR

CSTR-cascade: real tube reactor:

1 2 2
σ 2N =
2
σ N= − [ 1−exp−Bo  ]
n Bo Bo

comparison cascade and tube reactor:

Bo
n=
1

2 1−
Bo
[ 1−exp −Bo  ] 
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 3
 step input signal

St− S B
F=
S 0− S B

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 4
 pulse input signal

S t  SB
E
S0  SB

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 5
 Balancing of a continuous stirred tank reactor cascade

Q Sout
Q Sin

dS 1
V1 = Q  S in - S1  t = 0 ⇒ S1 = S 0
dt
dS 2
V2 = Q  S 1 - S 2  t = 0 ⇒ S 2 = S0
dt
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 6
Balancing of a continuous stirred tank reactor cascade
1
dS dt in 1 t
= ⇔ ln ( S - S )= - + a1
(S in
-S
1
) t R1 t R1
in 0
ln ( S - S )= a 1
( S in -S )1
t
( S in 0
-S )
= exp
( )
−
tR1

S *1= 1 - exp (−τ 1 )

2 1 2 *2
dS S -S dS *2
= ⇔ = 1 - exp (−τ1 ) - S
dt t R2 dτ2
substitution: Y = 1 - S*2 and τ1 =τ2 = n τ
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 7
Balancing of a continuous stirred tank reactor cascade
dY 1 dY
− = Y - exp−τ 1  ⇒  Y = exp  -n τ 
dτ 2 n dτ
dY
⇔  n Y = n exp  -n τ 
dτ
m= exp [∫ n d τ ] = exp  n τ 
Y = exp  -n τ  ∫ n exp  -n τ  exp  n τ  d τ  a 2
[ ]
⇔ Y = exp -n τ  [ n τ  a 2 ] with a 2 = 1 : initial condition
⇒ S*2 = 1 - exp-n τ  [ 1n τ ] with n = 2
2 3 n−1
 nτ   nτ   nτ 
*n
[
⇒ S = 1 - exp -n τ  1 n τ 
2

6
.. .
 n−1 !
Joachim Behrendt: Process Modelling of Wastewater Treatment
]
Tracer: 8
Cumulative frequency F of a CSTR-cascade
Response of a step input (dimensionless)
1,2

c u m m u la tiv e fre q u e n c y F 1,0

0,8 N=1
N=2
0,6 N=5
N=10
N=50
0,4

0,2

0,0
0,0 0,5 1,0 1,5 2,0 2,5 3,0
dimensionless retention time (t/t_R)

2 3 n−1
 nτ   nτ   nτ 
F = 1 - exp  -n τ  1n τ  [ 2

6
. ..
 n−1 ! ]
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 9
 Frequency E of a CSTR-cascade
2.0

fre q u e n c y E 1.5 n= 1
n= 2
1.0 n= 5
n = 10
0.5 n= 50

0.0
0.0 0.5 1.0 1.5 2.0
dimensionless time t/tR

dF  n  n  1
E n exp(-n )
d (n  1)!
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 10
 Balancing of a continuous tube reactor

Sin vx

∂ S ∂ S ∂2 S
= - vx  DS i nitial condition: t = 0 ⇒ S = 0
∂ t ∂ x ∂x
2

in dS
boundary conditions  t  0 : x = 0 ⇒ v x S = v x S - DS
dx
dS
x= L ⇒ =0
dx

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 11
 Balancing of a continuous tube reactor
transfor min g in the dim ensionless form :

S x
S*  and x * 
Sin L

Sin  S * Sin  S Sin  2 S

- v x  DS i nitial condition : t 0  S * 0
*
t Lx L2  x * 2

 S* S 1 2 S v L
 -  with Bo  x as Bodenstein number
t  x* Bo  x * 2 DS

* in in * Sin d S
boundary conditions (t  0) : x 0  v x S v x S S - DS
Ldx

1 d S*
 1 S -
Bo d x *

* d S*
x 1  0
*
dx

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 12
 revealing adverse flow behavior

prolongation

shortcut flow

internal
shortcut flow
circulation
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 13
 Shortcut flow

1.6
freq u en cy E

1.2

0.8

0.4

0
0.0 0.5 1.0 1.5 2.0
dimensionless time t/tR

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 14
 Stagnant volume

1.6
fre q u en cy E

1.2

0.8

0.4

0
0.0 0.5 1.0 1.5 2.0
dimensionless time t/tR

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 15
 Delay

1.2
1
fre q u e n cy E

0.8
0.6
0.4
0.2
0
0.0 0.5 1.0 1.5 2.0
dimensionless time t/tR

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 16
 Internal circulation

2
fre q u e n cy E

1.5

0.5

0
0.0 0.5 1.0 1.5 2.0
dimensionless time t/tR

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 17
Example for Modeling with OpenModelica

Source R1 R2 R3 Sink

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 18
Example for Modeling with OpenModelica

Source R1 R2 R3 Sink

connector con. con. connector

model model model

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 19
 Connector in Modelica

connector

output input

model model

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 20
 Connector in Modelica

connector OutPipe Statement for the object connector

(here for an output)
Real T; Potential variable like pressure,
temperature, voltage..

flow Real Q; Flow variable

output Real S_I; Components that are transported

with the flow out of the model

end OutPipe;
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 21
 Connector in Modelica

connector InPipe Statement for the object connector

(here for an input)
Real T; Potential variable like pressure,
temperature, voltage..

flow Real Q; Flow variable

input Real S_I; Components that are transported

with the flow into the model

end InPipe;
Joachim Behrendt: Process Modelling of Wastewater Treatment
Tracer: 22
Inflow Model

model Inflow
parameter Real InS_I = 1;
parameter Real InQ = 1.0;
parameter Real T = 20.0;
Real t;
Real Q;
OutPipe Out1;
equation
t = time;
Q = InQ;
Out1.S_I = InS_I;
Out1.Q = -abs(Q);
Out1.T = T;
end Inflow;

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 23
Modelling a Sink with Modelica

model SinkWater
Real S_I;
Real Q;
Real T;
InPipe In1;
equation
S_I = In1.S_I;
Q = In1.Q;
T = In1.T;
end SinkWater;

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 24
Modelling a reactor with Modelica
model CSTR
parameter Real V_R = 1;
output Real S_I;
InPipe In1;
OutPipe Out1;
initial equation
S_I = 0;
equation
der(S_I) = In1.Q / V_R * (In1.S_I - S_I);
Out1.Q + In1.Q = 0;
Out1.S_I = S_I;
Out1.T = In1.T;
end CSTR;

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 25
 Modelling a system with Modelica
model cascade
Inflow inflow;
SinkWater outflow;
CSTR R1(V_R = 1.0);
CSTR R2(V_R = 1.0);
CSTR R3(V_R = 1.0);
CSTR R4(V_R = 1.0);
equation
connect(inflow.Out1, R1.In1);
connect(R1.Out1, R2.In1);
connect(R2.Out1, R3.In1);
connect(R3.Out1, R4.In1);
connect(R4.Out1, outflow.In1);
end cascade;

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 26
Homework until November 11 of 2019
Problem discussion at 04.11.2019

The tracer model should operate with a step input

signal (ΔS = 2 g/m3) and additional with a dirac input,
where m = 15 g of trace is introduced within 6 min. The
inflow is keept constant at 2.5 m³/d while the total
Reaktor volumes is 4 m³ (R1=R2=R3=R4=1.0 m³). As
recirculation flow rate choose 0, 1, 2 and 5 m³/d. The
simulation time should be excees three hydraulic
retention times.

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 27
Homework until November 11 of 2019

Source R1 R2 R3 R4 Sink

Mixer Split

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 28
 If-Statement

algorithm
if ((D_t >= time) and not Step) then if expression then
S_Iin := m_I / InQ / D_t; statement ";"}
elseif Step then elseif expression then
S_Iin := InS_I; statement ";"}
else else
S_Iin := 0; statement ";"}
end if; end if;

Joachim Behrendt: Process Modelling of Wastewater Treatment

Tracer: 29