4 Dynamic Simulation of Technical Precipitation Processes
135
Fig. 22 Median of
volume-based PSD (L 50,3 )
versus impeller speed for
experiments and Setup C
simulation
50
100
150
200
0.4
0.5
0.6
0.7
0.8
0.9
Q prim = 0.2 L/min
L
50,3 (μm)
N (rpm)
Experiment
Simulation
illustrates the massive improvement of numerical efficiency if the PFR is not recalculated in every timestep. Without the approximation method ( f rec = 1), this simulation
case requires approximately τ ct ≈ 58 min on a single core of an Intel Core i-7 7700
CPU with 3.60 GHz. As shown in Fig. 23, the calculation speed can be improved by
several orders of magnitude by using a lower value for f rec . Interestingly, even for
f rec = 1/30, which requires τ ct ≈ 2.1 min instead of τ ct ≈ 58 min, the accuracy of
the model is only insignificantly diminished.
Fig. 23 Median of
volume-based PSD (L 50,3 )
for recalculation frequencies
of f rec =
1
2 to f rec =
1
80 .
Simulations performed at
Q prim = 0.2 L/min,
N = 100 rpm and,
= 0.5 s with Setup C
simulation. Computational
time τ ct referred to a single
core of an Intel Core i-7
7700 CPU with 3.60 GHz
0
5
10
15
20
25
30
0.50
0.55
0.60
0.65
0.70
0.75
0.80
L
50,3 (μm)
t (min)
f rec = 1/2, τ ct = 29 min
f rec = 1/10, τ ct = 5.7 min
f rec = 1/30, τ ct = 2.1 min
f rec = 1/80, τ ct = 0.9 min