Coupled DEM–CFD simulation of drying wood chips in a rotary drum – Baffle design and model reduction

Published Date
Fuel
15 November 2016, Vol.184:896–904, doi:10.1016/j.fuel.2016.05.054

Author 

Viktor Scherer ^a

Martin Mönnigmann ^b

Marc Oliver Berner ^b,,

Florian Sudbrock ^a

aRuhr-Universität Bochum, Department of Energy Plant Technology (LEAT), Universitätsstr. 150, D-44780 Bochum, Germany

bRuhr-Universität Bochum, Automatic Control and Systems Theory (RUS), Universitätsstr. 150, D-44780 Bochum, Germany

Received 15 February 2016. Revised 6 May 2016. Accepted 10 May 2016. Available online 24 May 2016. 

Abstract

In this contribution coupled DEM–CFD simulations of convective drying of wood chips in a baffled laboratory rotary dryer are presented. Due to the anisotropy of the biogenous (fibrous) material a three-dimensional spatial resolution of inner particle transport processes within the DEM code has been incorporated. The drying law is based on a diffusion approach. The simulations show that L-shaped baffles lead to higher drying rates than straight baffles. L-shaped baffles lead to a more even distribution of the particles across the cross-section of the drum where a larger amount of the wood chips are located in regions of hot, unsaturated air. However, the assumption of anisotropic transport properties within the wood chips and the subsequent solving of the associated differential equation of the three dimensionally resolved particle requires high computational effort. Therefore, the second purpose of the paper is to propose a model reduction method for single particle DEM-models based on proper orthogonal decomposition to reduce computation times. Specifically, we assess the impact of these methods on the computational complexity of the single particle models. The results show that even with a basic implementation a considerable reduction can be achieved on the single particle level. While our results only apply to the specific example treated here, it is evident that the effect of model reduction grows with grid size.

Keywords

DEM–CFD

Rotary dryer

Wood chips

POD

Reduced order model

---

Nomenclature

specific heat capacity [J kg⁻¹ K⁻¹]

d

diameter [m]

correction factor

source term for conservation equation of momentum [N m⁻³]

F

force vector [N]

g

acceleration of gravity vector [m s⁻²]

heat transfer coefficient [W m⁻² K⁻¹]

mass transfer coefficient [m s⁻¹]

h

enthalpy [J kg⁻¹]

heat flux density [W m⁻²]

mass flux density [kg s⁻¹ m⁻²]

k

stiffness of linear spring [kg s⁻¹]

K

effective coefficient of transport [W m⁻¹ K⁻¹ m² s⁻¹]

M

moment vector [N m]

M

number of time points

m

mass [kg]

N

number grid points

n

normal vector

np

number of particles in CFD cell

number of surface segments of ith particle in CFD cell

P

pressure [N m⁻²]

Nu

Nusselt number

Pr

Prandtl number

volumetric heat source [W m⁻³]

volumetric mass source [kg s⁻¹ m⁻³]

Sh

Sherwood number

t

time [s]

t

tangential vector

T

temperature [K]

T

stress tensor [N m⁻²]

v

velocity [m s⁻¹]

V

volume [m³]

x

position vector [m]

X

moisture content [kg_w  ]

mass fraction of the kth species

Greek letters

resistance coefficient [kg m⁻³ s⁻¹]

damping coefficient [kg s⁻¹]

transport coefficient [m² s⁻¹]

area of surface element [m²]

porosity

moment of inertia [kg m²]

thermal conductivity [W m⁻¹ K⁻¹]

friction coefficient

dynamic viscosity [kg s⁻¹ m⁻¹]

overlap [m]

normal displacement rate [m s⁻¹]

mass concentration, density [kg m⁻³]

angle of revolution [rad]

Subscripts

diss

dissipative

db

dry basis

dyn

dynamic

eff

effective

el

elastic

f

fluid

FC

CFD cell

h

heat

i

integral indices

j

integral indices

k

integral indices

lam

laminar

m

mass

q

energy

p

particle

S

surface

turb

turbulent

ambient conditions

Superscripts

n

normal

r

rolling friction

t

tangential

 Table 1

Table 1.

Fig. 1.

 Table 2

Table 2.

 Table 3

Table 3.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

 Table 4

Table 4.

Fig. 6.

Fig. 7.

Fig. 8.

 Table 5

Table 5.

• ⁎ 
  Corresponding author.

For further details log on website :
http://www.sciencedirect.com/science/article/pii/S0016236116303696