Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al...

14
Validation of Computational Fluid Dynamics Codes using Radiotracer Residence Time Distribution Analysis of Stirred vessels. C.P.K. Dagadu, H.A. Affum, S.Y. Adzaklo, I.I. Mumuni, G.K. Appiah, A. Coleman (Ghana)

Transcript of Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al...

Page 1: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Validation of Computational Fluid Dynamics Codes using Radiotracer Residence Time

Distribution Analysis of Stirred vessels.

C.P.K. Dagadu, H.A. Affum, S.Y. Adzaklo,

I.I. Mumuni, G.K. Appiah, A. Coleman

(Ghana)

Page 2: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Introduction

• Radiotracer residence time distribution (RTD): an experimental tool study the performance of industrial process reactors.

• RTD is used for:

Troubleshooting,

Check the validity of design data.

• The main concern during RTD investigation:

accurate formulation of the experimental distribution curve.

2 1

Page 3: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Objectives of study

• The final RTD results depends on the accuracy of the experimental RTD formulation.

• Hence the need to generated the RTD curve d using CFD and compare simulation and experimental results.

• In other words, to validate CFD codes using results of experimental RTD formulation.

3 2

Page 4: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Methodology : Experimental RTD

• Experimental RTD was conducted in a gold leaching tank to determine the MRT

Dagadu et. al (2014)

• Radiotracer used: I-131 (mixed with Na2S2O3)

• MRT Calculated from the method of moments

• MRTs: – Theoretical: 257mins.

– Experimental: 212 mins.

4 3

Page 5: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Description of Leaching Tanks

26/04/2017 5 4

Page 6: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Tracer Injection and Monitoring

6 5

I-131

Page 7: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Methodology: CFD Simulation

• Flowfield in the leaching tank calculated to describe the flow structure and mixing.

• Models used:

–Mixture model (recommended for slurry flows)

– Standard k- ϵ model

–MRF approach

• Dagadu et. al (2014) , (2015)

-

7 6

Page 8: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Computational Model

26/04/2017 8 7

Page 9: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

RTD Simulation • Lagrangian particle tracking method was used to

calculate the trajectories of massless tracer particles through a calculated flowfield.

• Dispersion of tracers due to turbulent eddies was calculated with the discrete random walk model under the stochastic tracking technique.

• Particle were tracked from the inlet to the outlet of the tank at a step time of 2 minutes.

9 8

Page 10: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

RTD Simulation (cont’d)

• No of particles tracked =5760

(required= 5000–10,000 , Bai et al., 2008)

• Particle concentration versus time curve generated.

–Histogram of particle concentration against recorded particle RT.

• The curve was then normalized to obtain the RTD curve.

10 9

Page 11: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Results and Discussions

Figure 1. Particle trajectories coloured by particle residence time [s]:

11 10

Page 12: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

Simulation vs Experimental RTDs

Exp. MRT = 212 min., Sim. MRT=205 min

12 12

0

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0 200 400 600 800 1000

E(t) Experimental

E(t) Simulation

t, min

E(t)

min

-1

Page 13: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

CONCLUSIONS

• The simulated RTD is in good agreement with experimental results.

• Particle tracking method also describes the flow structure in mixing vessels similar to plots of velocity vectors and contours of velocity as in previous works.

13 13

Page 14: Validation of Computational Fluid Dynamics Codes using ... · 4/26/2017  · •Dagadu et. al (2014) , (2015) - 6 7 . Computational Model 26/04/2017 7 8 . RTD Simulation •Lagrangian

THANK YOU FOR YOUR ATTENTION!

14 14