CFD Lagrangian Multiphase Simulation Applied to Dust Explosion Characterization

Transcription

1 CFD Lagrangian Multiphase Simulation Applied to Dust Explosion Characterization Speaker: Hugo Pineda Authors: Daniel Vizcaya, Carlos Murillo, Nathalie Bardin-Monnier, Olivier Dufaud, Laurent Perrin, Nicolás Ratkovich & Felipe Muñoz Contact: Daniel Vizcaya 1

2 Content 1. Introduction to dust explosions 2. Explosivity Parameters and Equipment 3. Simulation 4. Results 5. Conclusiones and Future Work 2

3 1. Introduction to dust explosions: Imperial Sugar Case Place: Port Wentworth, Georgia, USA Date: February 7, Process: Sugar refinery. Combustible: Sugar. Mean Particle size ( m): 23 P max (bar): 7.5 MEC (g/m 3 ): 95 Kst (bar m/s): 139 Impact: 14 fatalities. 36 injured. Process plant total destruction. Vorderbrueggen (2011) 3

4 1. Introduction to dust explosions Eckhoff (2009) 4

5 1. Introduction to dust explosions 5

6 2. Explosivity parameters: Maximum pressure and maximum rate of pressure rise Typical pressure profile of an explosion on the 20 L Sphere Dahoe et al. (2001) 6

7 2. Explosivity parameters: Minimum Explosivity Concentration (MEC) Maize Starch Characterization Eckhoff (2009) 7

8 2. Explosivity parameters: Particle Size Aluminium Dust in Air Eckhoff (2009) 8

9 2. Explosivity parameters: Turbulence level Lycopodium in Air Eckhoff (2009) 9

10 2. Explosivity parameters: Deflagration Index K st = dp dt V 1/3 Is a severity index developed from the maximum rate of pressure rise and volume-invariant. Table 1. Risk level associated to Deflagration Index Risk Level Kst (bar m/s) Explosion Description ST Weak ST Strong ST 3 > 300 Very Strong 10

11 2. Explosivity parameters: Deflagration Index K st = dp dt V 1/3 Is a severity index developed from the maximum rate of pressure rise and volume-invariant. Material Mean Particle Size Concentration Kst (bar m/s) Starch Polyethylene Phenolic Resin

12 2. Dust explosion characterization equipment 20 L Sphere 1 m 3 Vessel MEC from nominal concentration. Maximum pressure (P max ). Maximum rate of pressure rise (dp/dt max ). 12

13 2. Deflagration Index K st = dp dt V 1/3 Dahoe et al. (2001) 13

14 2. Dust explosion characterization equipment: Comparative Table Table 2. Comparison between dust explosion characterization equipments Dust container pressure (barg) 20 L Sphere 1 m 3 Vessel Ignition Time (ms) Ignition Energy (kj) Type of nozzle Rebound Annular Type of flammable dust High densities All densities [13] 14

15 2. Experimental Studies Dahoe et al. (2001) 15

16 3. Objectives To develop a CFD multiphase flow to evaluate the dispersion phenomena of flammable dust within the 20 L Sphere under standard test conditions. To evaluate the dust concentration at ignition point on the moments previous to the explosion. To evaluate the turbulence level during the dispersion of the dust. To evaluate the particle distribution within the domain taking into account the particle size. 16

17 3. Simulation: General Geometry Igniters Sphere Nozzle Canister Nozzle Vertical Cut 17

18 3. Simulation: Mesh Models: Polyhedral mesher Advancing Layer Mesher Surface Remesher Principal Reference Values: Number prism layers: 2 Prism Layer Thickness: mm Minimum Surface Size: 0.2 mm Mesh Results: Cells: 7,316,852 Faces: 51,224,435 Verts: 43,619,275 18

19 3. Simulation: Mesh Vel ref > 400 m/s 19

20 3. Simulation: Principal Models Implicit Unsteady. Time Step: 1 ms Temporal Discretization: 2 nd Order Compressible Gas: Peng-Robinson Lagrangian Multiphase Detached Eddy Simulation 20

21 Mass Fraction 3. Simulation: Principal Models Lagrangian Multiphase: Material: Starch Models: Constant density: = 610 kg/m 3 Drag Force Material Particles Pressure Gradient Force Spherical Particles Injector: Mass Flow Rate: 6 kg/s for t < 0.2 ms Number of Parcels: 2.2e+06 Particle Size Distribution from laser diffraction (Table 3) Table 3. Particle Size Distribution from laser diffraction Particle Size Distribution Particle Size (μm) 21

22 3. Simulation: Initial Conditions 22

23 Pressure (bar) 4. Results: Influence of iterations Pressure profile Time Step 250 Iter 1000 Iter Experimental results 23

24 Simulation time (h) 4. Results: Computational cost 200 Computational Cost iter 1000 iter Time Step Windows Server bit Processor: Intel Xeon 2.67GHz RAM; 40 Gb 24

25 4. Results: Particle flow 25

26 4. Results: Particle flow 26

27 4. Results: Particle diameter for 1000 iter 27

28 5. Conclusions STAR-CCM+ was used to developed a multiphase lagrangian flow in order to simulate the flammable dust dispersion within the 20 L sphere. The number of Maximum Inner Iterations has a significant impact on the results as well as on the computational cost. The nozzle disperses correctly the dust within the domain. However, the smaller particles tends to the walls and the bigger ones to the center of the sphere. In order to model correctly the behavior of the flow, and due to the geometry high complexity, it was necessary to implement the DES turbulence model. 28

29 5. Future Work The Maximum Inner Iterations will be increased to The simulation time will be 120 ms in order to evaluate a better ignition time for the standard test. The use of cohesion models to represent particle fragmentation and agglomeration will be added. Another types of nozzle will be simulated in order to evaluate the dust homogeneity within the sphere. 29

30 CFD Lagrangian Multiphase Simulation Applied to Dust Explosion Characterization Speaker: Hugo Pineda Authors: Daniel Vizcaya, Carlos Murillo, Nathalie Bardin-Monnier, Olivier Dufaud, Laurent Perrin, Nicolás Ratkovich & Felipe Muñoz Contact: Daniel Vizcaya 30

31 7. BIBLIOGRAFHY "ISO 6184/1," Explosion Protection Systems - Part 1: Determination of Explosion Indices of Combustible Dusts in Air, ISO 6184/1, W. Bartknecht, "Explosions and how they may be prevented," W. Bartknecht, "Explosions: Course, Prevention, Protection," W. Bartknecht, "Ignition capabilities of hot surfaces and mechanically generated sparks in flammable gas and dust/air mixtures," A. E. Dahoe, R. S. Cant, M. J. Pegg, and B. Scarlett, "On the transient flow in the 20-liter explosion sphere," Journal of Loss Prevention in the Process Industries, vol. 14, pp , V. Di Sarli, P. Russo, R. Sanchirico, and A. Di Benedetto, "CFD simulations of dust dispersion in the 20 L vessel: Effect of nominal dust concentration," Journal of Loss Prevention in the Process Industries, vol. 27, pp. 8-12, R. K. Eckhoff, "Understanding dust explosions. The role of powder science and technology," Journal of Loss Prevention in the Process Industries, vol. 22, pp , M. Hertzberg, I. A. Zlochower, R. S. Conti, and K. L. Cashdollar, "Thermokinetic transport control and structural microscopic realities in coal and polymer pyrolysis and devolatilization: their dominant role in dust explosions," 1987, pp

32 7. BIBLIOGRAPHY O. Kalejaiye, P. R. Amyotte, M. J. Pegg, and K. L. Cashdollar, "Effectiveness of dust dispersion in the 20-L Siwek chamber," Journal of Loss Prevention in the Process Industries, vol. 23, pp , N. Kalkert and H. G. Schecker, "Determination of explosion limits of ammonia in mixtures with simple hydrocarbons and air," German Chemical Engineering, vol. 3, pp , C. Murillo, O. Dufaud, N. Bardin-Monnier, O. López, F. Munoz, and L. Perrin, "Dust explosions: CFD modeling as a tool to characterize the relevant parameters of the dust dispersion," Chemical Engineering Science, vol. 104, pp , C. Proust, A. Accorsi, and L. Dupont, "Measuring the violence of dust explosions with the "20 l sphere" and with the standard "ISO 1 m3 vessel". Systematic comparison and analysis of the discrepancies," Journal of Loss Prevention in the Process Industries, vol. 20, pp , R. Siwek, "20-L Laborapparatur für die Bestimmung der Explosionskenngrößen brennbarer Stäube," Thesis, J. B. Vorderbrueggen, "Imperial sugar refinery combustible dust explosion investigation," Process Safety Progress, vol. 30, pp ,