Anda di halaman 1dari 32

1

NAFENS World Congress


vancouver, Nay 2007
MULTI-PHYSICS MODELLING AND
SIMULATION:
PROGRESS AND CHALLENGES
M. Cross, T.N. Croft, D. McBride, A.K. Slone,
and A.J. Williams
Centre for Civil and Computational
Engineering
School of Engineering
University of Wales, Swansea
m.cross@swan.ac.uk
1
NAFENS World Congress
vancouver Nay 2007
Why Nulti-physics Nodelling ?
Large number of real world problems require multi-
physics simulation tools.
Examples
Solidification problems - Solder Joints
Fluid-Structure interaction - Flutter in aircraft wings
Need to solve for integrated physics
Ensure two-way coupling
Fluid Flow
Stress
Analysis
Electromagnetics
Heat Transfer
NAFENS World Congress
vancouver Nay 2007
Challenge of multi-physics
Multi-physics - closely coupled interactions amongst
separate continuum physics phenomena
CAE analysis software essentially phenomena specific :
- CFD - Fv techniques with segregated iterative solvers
- CSN - FE techniques with direct solver structure
- CAA 8 CEN uses any of FEfFvfBEfFDTD techniques
plus heritage software approaches that go with each.
Nust ensure accurate filtering and mapping of data for
volume source and boundary data
Nust ensure that where needed, the mesh and geometries
deform compatibly
2
NAFENS World Congress
vancouver Nay 2007
Directly Directly Coupled Coupled Problems Problems
K K
f f
K K
fs fs
K K
ft ft
K K = = K K
sf sf
K K
s s
K K
st st
K K
tf tf
K K
ts ts
K K
t t
a a
f f
a a = = a a
s s
a a
t t
f f
f f
f f = = f f
s s
f f
t t
Ka Ka = = f f
A single code for all A single code for all
phenomena phenomena
coupling direct coupling direct
NAFENS World Congress
vancouver Nay 2007
Explicit Explicit or or implicit implicit
K K
f f
n n
a a
f f
n n
= = f f
f f
n n
- - g g
1 1
( (a a
s s
n n- -1 1
, ,a a
t t
n n- -1 1
) )
K K
s s
n n
a a
s s
n n
= = f f
s s
n n
- - g g
2 2
( (a a
f f
n n
, ,a a
t t
n n- -1 1
) )
K K
t t
n n
a a
t t
n n
= = f f
t t
n n
- - g g
3 3
( (a a
s s
n n
, ,a a
t t
n n
) )
Impact of using Impact of using
distinct solvers distinct solvers
for each phenomenon for each phenomenon
Staggered Staggered Solution Solution: :
1
NAFENS World Congress
vancouver Nay 2007
Classifying multi-physics
Nost vendors advertise multi-physics
Nost vendors offer multi-disciplinary
Nulti-disciplinary - using data generated by one code as input
into another - loose or one way coupling
(e.g. electric field loading a thermal calculation)
Nulti-physics - two way exchange of information, which could
involve implicit convergence within a time-step
(e.g. thermo-mechanical)
Closely coupled multi-physics - time and space accurate
exchange of data
(e.g. dynamic fluid-structure interaction)
NAFENS World Congress
vancouver Nay 2007
Key issues in closely coupled
multi-physics simulation
Phys-A
Phys-B
Interpolation from one set
of variables to another =>
compatibility of mesh
Virtual single database of mesh
data & simulation variables
Solver strategy
- direct vs iterative
- Eulerian vs Lagrangian
Is coupling strategy compatible
with scalable parallelism, EVEN if
software components are parallel
2
NAFENS World Congress
vancouver Nay 2007
Multi-physics: coupling issues
Existing `component' phenomena analysis software -
coupling
Need a filter structure for code interoperability:-
exchange information directly from each others database
without opening closing files
Options here derive from tools for developed for
parallelisation - especially PvN, NP! and specifically
NpCC!
!nteroperability not equivalent to simulating physics of
coupling
Remaining Challenge - parallel scalability of interoperable
codes
NAFENS World Congress
vancouver Nay 2007
MpCCI - a tool for code interoperability
Emerged from an EU project - public domain OPEN
SOURCE tools
www.scai.fraunhofer.de/mpcci.0.html
A number of NpCC! facilitated code couplings:
- ABAQUS + FLUENT for DFS!
- STAR-CD + NASTRAN for DFS!
NpCC! here to stay and facilitates genuine coupling -
initial experiences not trouble free, but . .
3
NAFENS World Congress
vancouver Nay 2007
Sector Specific 'multi-physics' Software
Castings
PROCAST http:JJwww.ues-software.com
MAGMASOFT http:JJwww.magmasoft.com
Forming
DEFORM http:JJwww.deform.com
SUPERFORGE http:JJwww.mscsoftware.com
FORGE3 http:JJwww.transvalor.com
Polymers
C-Mold http:JJwww.moldflow.com
Joining Processes
SYSWELD http:JJwww.esi-group.com
Electronic cooling
Flotherm http:JJwww.flomerics.com
NAFENS World Congress
vancouver Nay 2007
Commercial CAE analysis - a web survey
Tools claiming multi-physics capabilities:
ANSYS/Multi-physics http://www.ansys.com
ABAQUS http://www.abaqus.com
ADINA http://www.adina.com
ALGOR http://www.algor.com
AUTODYN http://centurydynamics.com
CFD-ACE http://www.esi-group.com
DYNA http://www.lsc.com
COMSOL http://www.comsol.com/
LMS software http://www.lms.com
MSC- NASTRAN http://www.mscsoftware.com
PHYSICA+ http://www.physica.co.uk
STAR-CCM+ http://www.adapco.com
1
NAFENS World Congress
vancouver Nay 2007
Sector Specific 'multi-physics' Software
Castings
PROCAST http:JJwww.ues-software.com
MAGMASOFT http:JJwww.magmasoft.com
Forming
DEFORM http:JJwww.deform.com
SUPERFORGE http:JJwww.mscsoftware.com
FORGE3 http:JJwww.transvalor.com
Polymers
C-Mold http:JJwww.moldflow.com
Joining Processes
SYSWELD http:JJwww.esi-group.com
Electronic cooling
Flotherm http:JJwww.flomerics.com
NAFENS World Congress
vancouver Nay 2007
Commercial CAE analysis - a web survey
Tools claiming multi-physics capabilities:
ANSYS/Multi-physics http://www.ansys.com
ABAQUS http://www.abaqus.com
ADINA http://www.adina.com
ALGOR http://www.algor.com
AUTODYN http://centurydynamics.com
CFD-ACE http://www.esi-group.com
DYNA http://www.lsc.com
COMSOL http://www.comsol.com/
LMS software http://www.lms.com
MSC- NASTRAN http://www.mscsoftware.com
PHYSICA+ http://www.physica.co.uk
STAR-CCM+ http://www.adapco.com
2
NAFENS World Congress
vancouver Nay 2007
Key route to closely coupled multi-disciplinary {multi-
physics) simulation
Basic requirements of a SSF:
- consistency of mesh for all phenomena
- compatibility in the solution approaches to
each of the phenomena
- single database & memory map so that no
data transfer 8 efficient memory use
between programs
- facility to enable accurate exchange of
boundary or volume sources (e.g. body force)
- enables scalable parallel operation for all physics
interactions
Alternative approach:
Single Software Framework
NAFENS World Congress
vancouver Nay 2007
Single software frameworks for
multi-physics
CONSOL
- FE solver technology with some
phenomena specific modules
- Nice environment; easy to use
STAR-CD (8 STAR-CCN+ emerging)
- Fv solver technology, polyhedral element
- CFD based with strong meshing
PHYS!CA
- Nixed Fv (for flow, etc) 8 Fv or FE (for
stress)
3
NAFENS World Congress
vancouver Nay 2007
`Straightforward' multi-physics
CFD
Potential
CSN
Anything involving
Potential Equation
coupled to some other
physics
Potential equation:
-Heat
-Electric field
-etc
NAFENS World Congress
vancouver Nay 2007
Multi-physics Modelling
Physics Requirements
Fluid Flow
Heat transfer
Solidificationfphase change
Stress
Electro-magnetics
Acoustics
Geometry
Complex
Large simulations
NULT!-PHYS!CS
UNSTRUCTURED
PARALLEL
Key issue: CFD capability
1
NAFENS World Congress
vancouver Nay 2007
Continuous casting process:
example of CFD based multi-physics
Mixture of liquid steel
and argon injected into
rectangular mould
Liquid metal flux sits on
top of mould
Water cooled mould
extracts energy forming a
solid steel shell
Continuous withdrawal
B.G. Thomas
NAFENS World Congress
vancouver Nay 2007
Solution domain
2
NAFENS World Congress
vancouver Nay 2007
Free surface (SEA)
Solves:
where f is the fraction
of metal in a cell
van Leer scheme used to
reduce smearing of interface
continuity equation solved for volume not mass
properties a linear combination of phases present
0 . = +

u
t
NAFENS World Congress
vancouver Nay 2007
z
Solid regions appear in blue
End
View
Top view
Solidification Strand
3
NAFENS World Congress
vancouver Nay 2007
Argon bubble injection:
closely coupled L-E approach
L particles
Incorp in cont
CFD code
Embed with
mass flux
effects
Eval
integrated
path
NAFENS World Congress
vancouver Nay 2007
Clustering of argon bubbles
1
NAFENS World Congress
vancouver Nay 2007
Computations were also performed to estimate the effects of
EMB on the free surface . For this the Maxwell equations were
solved, which with the usual MHD assumptions, lead to:
Continuity of magnetic flux:
Ohm's Law for conducting metals
Magnetic Transport, or Induction equation
Lorentz force:
Note: Terms containing the velocity U, are only important when Rm
(=LU/h)> 1
0 = B .
- = E where ), B U + E ( = J


m
2
1
= where, B + ) B U ( =
t

1
NAFENS World Congress
vancouver Nay 2007
B=0T
B=0.4T
Flow suppressed
here
Fluid behaviour under EMB conditions
NAFENS World Congress
vancouver Nay 2007
Coupled EM-flow calculations
For most practical calculations in metals
processing:
The EM field influences the flow and
thermal fields
BUT the thermo-fluid phenomena has little
influence of the EM fields
Hence, essentially one way coupling
So calculate the EM field and calculate the
thermal and flow loads in the CFD
calculation
Can implement above model in any good
CFD code!
2
NAFENS World Congress
vancouver Nay 2007
Welding processes simulation - natural
multi-physics
Processes involve:
free surface flow
electromagnetic forces
heat transfer with solidification/melting
development of non-linear stress
Ideal candidate for multi-physics modelling
NAFENS World Congress
vancouver Nay 2007
T-Junction arc weld simulation
1
NAFENS World Congress
vancouver Nay 2007
Experiment and simulation
T-junction section,
highlighting HAZ region
X
Y
Z
Model: T_J
CASE1: PHYSICA Results
Step: 1 TIME: 0
Nodal LFN
Max = 1 Min = 0
.909E-1
.182
.273
.364
.455
.545
.636
.727
.818
.909
FEMGV 5.1-01 28 FEB 2000 Greenwich University

G E R
E
N
of
Y
T
I
W
I
U
I
N
V
E
t h e
H
C S
R
the
UNIVERSITY
of
GREENWICH
NAFENS World Congress
vancouver Nay 2007
Distortion of T-junction due to heat source
Heat source
2
NAFENS World Congress
vancouver Nay 2007
Weld pool dynamics
Velocity vectors in crossection
Lorentz force distribution in the
weld-pool
NAFENS World Congress
vancouver Nay 2007
Distortion of T-junction due to heat source
Distortion
3
NAFENS World Congress
vancouver Nay 2007
Welding multi-physics BUT . .
Welding involves:
free surface fluid flow
heat transfer and solidification/melting
electro-magnetic fields
non-linear stress
BUT . . no coupling back:
from thermo-fluids to EM field
from stress calculation to thermo-fluids
SO . . reasonably loosely coupled
NAFENS World Congress
vancouver Nay 2007
Generic Dynamic Fluid Structure Interaction
Closely coupled multi-disciplinary problem
Time & space accurate
Very challenging in every respect.
Issue of GCL
Implementation of boundary conditions.
Features of single software framework:
Consistency of mesh.
Single database & memory map.
Compatibility in the solution approaches FV-UM.
Traction
boundary
condition
CFD
CSM
Deformation
Mesh
adaptation
1
NAFENS World Congress
vancouver Nay 2007
Three Phase Approach
CMD
1 +

n n
t t
m m m
F d K =
CSD
( ) t
s
F Kd d C d M = + +

fs m
on u
fs fs p
= on t
fs
on d
Generalised
Newtonian
Continuity
CFD
NAFENS World Congress
vancouver Nay 2007
Vertex Based
Mesh
Element
Gauss
Point
Finite Volume
Cell Centred
Control
Volume
Control
Volume
Integration
Point
x
Finite Element
x
Node
Finite Volume
x
x
x
x
x
x
x
x x
x
x
PHYSICA
Spatial Discretisation for closely coupled
multi-physics
Unstructured mesh
CFD
Cell centred
Or mixed CC- VB
FV
CSM
Vertex based
FV/FE
2
NAFENS World Congress
vancouver Nay 2007
Dynamic fluid-structure interaction
Targeted at
problems involving
flow induced
vibrations
Use dynamic
structural
equations and
Navier-Stokes flow
equations
Wind direction Flow induced vibrations
NAFENS World Congress
vancouver Nay 2007
Dynamic response of structure without flow
1
NAFEMS World Congress
Vancouver May 2007
Fluid Velocity and Pressure Movies
At tip of wing
NAFEMS World Congress
Vancouver May 2007
Shear stress
xy
in wing
2
NAFEMS World Congress
Vancouver May 2007
Bio-medical multi-physics modelling:
the heart!
Heart a multi-physics system, featuring
interactions between:
- electro-chemical system
- fluid
- structure
Image from www.heartvalves.8m.com
NAFEMS World Congress
Vancouver May 2007
Geometry
2D model
Right ventricle
Blood
0.005m
Outlet Inlet
Wall
0.045m
0.01m 0.01m
0.08m
3
NAFEMS World Congress
Vancouver May 2007
Electro-fluid-structure interaction
Electrical
Structure
Mesh
dynamics
Fluid
flow
NAFEMS World Congress
Vancouver May 2007
Multi-potential heart electrical activity model:
comparison of results
PHYSICA
Clayton and Holden 02
Primary and secondary potentials Currents
1
NAFEMS World Congress
Vancouver May 2007
Coupling electrical field to structural mechanics
Assumption made that small strain model can
capture behaviour of heart wall
Dominant behaviour of wall is contraction/expansion
Shearing effects negligible
Elastic model
Change in potential results in a change in tension
in the heart wall
To model tension we introduce an electric strain
into structural mechanics equations
NAFEMS World Congress
Vancouver May 2007
Electric potential, deformation &
flow patterns
Results at various stages through a heart beat cycle
2
NAFENS World Congress
vancouver, Nay 2007
Parallel Multi-Physics
Modelling
NAFEMS World Congress
Vancouver May 2007
Exploiting parallel cluster technology:
the challenge
MpCCI and other filter
technologies
Upside enables
interaction at the code
dbase level
Downside all data
exchange must go via
filter and is a compute
bottleneck wrt
scalability on parallel
clusters
Code A
(CFD)
Code B
(FEA)
Filter tech:
Map onto
Parallel
cluster
Map onto
Parallel
cluster
BOTTLENECK BOTTLENECK
3
NAFEMS World Congress
Vancouver May 2007
Parallel Multi-Physics Framework
CAPlib JOSTLE
PHYS!CA
PDE Solver
Generic Parallel Calls Lib Nesh Partition
Simulations very Time Consuming need
Parallel capability
NAFEMS World Congress
Vancouver May 2007
Parallelisation approach
Partition of 3D unstructured mesh
by JOSTLE
Uses mesh partitioning SPMD
strategy with non-uniform
workload
Assumes a homogeneous load
balance across the mesh:
load balanced ( even no of cells per
node)
minimises sub-domain
interface elements
sub-domain connectivity
matches processor topology of the
parallel system
4
NAFEMS World Congress
Vancouver May 2007
Multi-physics Simulation parallel issues
Sub-domains have
specific physics so
partition must reflect this:
non-uniform load/node
Distinct physics uses
distinct discretisation
procedures:
secondary partitions
Sub-domains may change
as problem develops:
dynamic load balance
Solid mechanics
Fluid flow
Heat transfer
Strategy needs to address all the above issues
NAFEMS World Congress
Vancouver May 2007
Primary & secondary partitions
Primary & secondary
meshes
Good primary & poor
Secondary partition
Good primary &
Secondary partitions
from JOSTLE
5
NAFEMS World Congress
Vancouver May 2007
Parallel multi-physics: two level approach
Implement a generic parallel version
of Multi-physics code/ MDA codes
without regard to in-homogeneity of the
computational work over the mesh(es) defining
the analysis domain
Dump the load balancing into the
mesh (re)partitioning task -
JOSTLE_DLB
Process as straightforward as possible
JOSTLE
PHYS!CA
PDE Solver
Mesh Partition
NAFEMS World Congress
Vancouver May 2007
Metal Forming - Extrusion
Involves large scale
deformation of metal work-
piece through interaction
with one or more dies
Multi-physics problem
Flow/deformation of
work-piece
Heat transfer generated
by internal friction
Stress/strain in die(s)
6
NAFEMS World Congress
Vancouver May 2007
Mixed Eulerian-Lagrangian Approach
Workpiece
Eulerian mesh
Free-surface algorithm to track deformation
Non-Newtonian material model
Heat transfer plus energy generated by internal friction
Die
Lagrangian mesh
Mechanical behaviour coupled with:
Thermal behaviour in workpiece
Fluid traction load from workpiece
NAFEMS World Congress
Vancouver May 2007
Governing Equations - Extrusion
Coupled Thermo - mechanical problem
Heat transfer significant factor in deformation process
CFD
Non-Newtonian viscosity model Plastic Norton Hoff law
Heat Transfer - Friction between die and workpiece.
Free Surface - Van Leer method
CSM
Static equilibrium equation linear elastic solid.
Coupling at the workpiece/die boundary:
Die subject to fluid traction boundary condition.
Workpiece subject to a die velocity boundary condition.
Dynamic meshes GCL. Fluid velocity relative to mesh
movement.
7
NAFEMS World Congress
Vancouver May 2007
Extrusion through U-shaped die
4.76 mm
41.27 mm
47.62 mm
Initial diameter = 200mm
Bearing length = 2.5mm
Punch speed = 5.85E-3m/s
63220 elements
69507 nodes
Workpiece = 470 Workpiece = 470 C C
Die = Die = 450 450 C C
Air = 30 Air = 30 C C
NAFEMS World Congress
Vancouver May 2007
Temperature contours in extruding work-piece
8
NAFEMS World Congress
Vancouver May 2007
Effective stress contours and deformation of die
NAFEMS World Congress
Vancouver May 2007
Parallel results
13.43 6.1 16
10.92 7.5 12
8.03 10.2 8
4.48 18.3 4
1 81.9 1
Speed-up
Run time
(hours)
Processors
Itanium IA 64 cluster running Linux OS
Eight nodes, two 733MHz processors per node
Each node with 2 Gb memory & 2Gb swap space
Single phase mesh partitions on 16 processors
9
NAFEMS World Congress
Vancouver May 2007
Challenges
Multi-physics OK where
discretisation methods are
coherent
What about distinct
methods FE-BE can be
OK
Continuum-particulate
interaction fracturing
What happens when it
involves combustion, heat
transfer, fluid flows, etc
Can I do this scalably in
parallel?
Fracturing
NAFEMS World Congress
Vancouver May 2007
Conclusion
Multi-physics simulation is emerging in a commercially
supported manner
Most successful multi-physics is based upon loose or one-
way coupling even then, heroic computing
Close coupling in time and space another ball game - Key
here are procedures for time & space accurate simulations;
DFSI a key exemplar
Multi-physics essentially compute intensive leads to
challenge of parallel scalability for multi-physics
simulation tools
Can do for bespoke single software solutions, but for
multi-code components, not so clear!
Challenges for the future integrating components using
essentially distinctive model paradigms & solver strategies