13e rencontre SMAI math industrie Physically correct and timely - - PowerPoint PPT Presentation

▶
13e rencontre smai math industrie
SMART_READER_LITE
LIVE PREVIEW

13e rencontre SMAI math industrie Physically correct and timely - - PowerPoint PPT Presentation

Jan 09, 2024 484 likes •910 views

13e rencontre SMAI math industrie Physically correct and timely simulations: the challenge towards Virtual Prototyping Argiris Kamoulakos, ESI Group 22/03/2011 SMAI HPC 1 Introduction Today's industry trend is towards reduction of

slide-1
SLIDE 1

13e rencontre SMAI math‐industrie

Physically correct and timely simulations: the challenge towards Virtual Prototyping Argiris Kamoulakos, ESI Group

22/03/2011 1 SMAI ‐ HPC

slide-2
SLIDE 2

Introduction

  • Today's industry trend is towards reduction of

physical prototyping through simulation driven design in order to reduce drastically the time‐to‐market of new products. This is called Virtual Prototyping. However, it relies heavily

  • n the capacity to reproduce through

simulation the correct physics and in a timely way to respect the industrial development process deadlines.

22/03/2011 SMAI ‐ HPC 2

slide-3
SLIDE 3

ESI Group Offering

3

Software & Services for End‐to‐End Virtual Prototyping

22/03/2011 SMAI ‐ HPC

slide-4
SLIDE 4

Enjeux / problématique industrielle

  • Crash Failure Analysis as an industrially

affordable solution

  • Must be performed “over-night”

22/03/2011 SMAI ‐ HPC 4

slide-5
SLIDE 5

A typical problem: automotive side‐impact

B‐Pillar model

22/03/2011 SMAI ‐ HPC

slide-6
SLIDE 6

A cumulative strain damage model

Non proportional loading Plasticity based Triaxiality dependence Lode angle dependence

The Wilkins (EWK) metal rupture model

  • Mathematical details -

22/03/2011 SMAI ‐ HPC

slide-7
SLIDE 7

Importance of "Size effect"

A severe condition imposed on model element size in order to be predictive

22/03/2011 SMAI ‐ HPC 7

Size effect:

  • critical volume to be saturated with damage in order

to start a “crack”

  • Typical values
  • Aluminum: (0.15mm)**3
  • ~6 solid elements per mm
  • High Strength Steel: (0.05mm)**3
  • 15‐20 solid elements per mm
slide-8
SLIDE 8

High fidelity modelling

  • f the B‐Pillar

~4,000 shells

~1,000,000 solids

22/03/2011 SMAI ‐ HPC

slide-9
SLIDE 9

The industrial challenge The industrial challenge

  • Crash Failure Analysis as an industrially

affordable solution

  • Example
  • Must be performed “over-night”

22/03/2011 SMAI ‐ HPC

slide-10
SLIDE 10

Car Model: 879,000 Elements Time Step: 1μs Subframe (520k-Model): 520,000 TET10-Elements Time Step: 0.05 μs Time step ratio: Rt = 20 Local model element fraction: fL = 59 % +

Problem: It is not just a matter of “size”, but algorithmic, too. (Disproportionate distribution of solution effort)

Towards extreme local refinement Towards extreme local refinement

slide-11
SLIDE 11

Casted parts pose a special issue

with porosity no porosity

22/03/2011

SMAI ‐ HPC Gas Porosity, Shrink Porosity, Micro Porosity

slide-12
SLIDE 12

Very detailed models are needed for the prediction of the grain structure

Uniform Solidification in Ni Turbocharger

CAFE (Cellular Automata FE) was developed in collaboration with: Howmet, Snecma (France), ABB, EPFL (Switzerland), AETC, Rolls Royce (UK), PCC (USA)

12

22/03/2011

SMAI ‐ HPC

slide-13
SLIDE 13

In Aeronautics the problem of Birdstrike is particularly accute

Courtesy of EC project CRAHVI

SPH

22/03/2011 SMAI ‐ HPC

slide-14
SLIDE 14

You cannot predict what you do not model for !

Incorporation of details is critical for performance prediction

14

The rivets and their effect in rupture have to be sufficiently represented by the model.

Courtesy of EC project CRAHVI

22/03/2011

SMAI ‐ HPC

slide-15
SLIDE 15

Les points de blocage

  • Mesh size

– huge meshes needed

  • Computational algorithm

– Dissimilar meshes in size and type (FE / meshless)

  • Initial state of structure

– Manufacture (filamentary composites) – Connectors (spotwelds, rivets etc.)

22/03/2011 SMAI ‐ HPC 15

slide-16
SLIDE 16
  • Manufacture of Fillamentary Composites

22/03/2011 SMAI ‐ HPC 16

slide-17
SLIDE 17

17

New Manufacturing methods for advanced composites

slide-18
SLIDE 18

18

New Manufacturing methods for advanced composites

slide-19
SLIDE 19

Virtual Properties prediction

(through virtual coupon testing)

19 Particle methods for the resin 22/03/2011

SMAI ‐ HPC

slide-20
SLIDE 20

Virtual Properties prediction

(through virtual coupon testing)

20 Particle methods for the resin 22/03/2011

SMAI ‐ HPC

slide-21
SLIDE 21
  • Modelling connectors (Spotwelds, Rivets etc.)

22/03/2011 SMAI ‐ HPC 21

slide-22
SLIDE 22

Effect of connectors in automotive crash Spotewelded beam modelling

22

Deformed spotwelds Deformed beam

22/03/2011 SMAI ‐ HPC

slide-23
SLIDE 23

23

Effect of connectors in automotive crash Spotewelded beam modelling

22/03/2011 SMAI ‐ HPC

slide-24
SLIDE 24

The challenge of Self Piercing Rivets: knowing their shape and state

22/03/2011 SMAI ‐ HPC

slide-25
SLIDE 25

The challenge of Self Piercing Rivets: knowing their shape and state

22/03/2011 SMAI ‐ HPC

slide-26
SLIDE 26

Les points de blocage

  • Mesh size

– huge meshes needed

  • Computational algorithm

– Dissimilar meshes in size and type (FE / meshless)

22/03/2011 SMAI ‐ HPC 26

slide-27
SLIDE 27
  • Modelling substructures and systems

22/03/2011 SMAI ‐ HPC 27

slide-28
SLIDE 28

28

Application to Battery Modelling

Pr Prob

  • blema

lematics: ma tics: material issues terial issues

Mic Microscale

  • scale structur

cture ( e (20 0 μm) m)

Discontin Discontinuous uous Nano-par Nano-particles les, nano-f nano-fibers

Modeling Modeling

Contin Continuous mec uous mechanics anics

Iss Issues

Contin Continuous mec uous mechanics anics pr proper

  • perti

ties es Inter Interpreta tation tion Calibr Calibration tion

Separat rators rs Anod Anodes Ca Cath thodes

22/03/2011 SMAI ‐ HPC

slide-29
SLIDE 29

29

Application to Battery Modelling

Structur ctural la al layout: ut: Re Requirements:

Str Stresses sses and str and strains in eac ins in each h la layer

Result : 200 sult : 200 elements/mm elements/mm A 20cm x 20cm ensemb A 20cm x 20cm ensemble le giv gives s 32 Million 32 Million elements elements

20 µm 200 layers plastic covers clamping: pressure shaping: bending, stretching (?), drawing (?)

22/03/2011 SMAI ‐ HPC

4mm

slide-30
SLIDE 30

30

Bullet car speed = 48 km/h Target car speed = 24 km/h HUMOS 50% male model: Driver of target car Hands attached on steering wheel Right foot on breaking pedal Left foot on the floor Restraint devices: belts Injury assessment

Ultimate goal: be predictive at the system level

HUMOS

EC project HUMOS demonstrator of real car crashes

22/03/2011

SMAI ‐ HPC

slide-31
SLIDE 31

Motion in Waves: Regular head waves of 8 meter height and a wave length of 294 meter

Marine applications are the next frontier

22/03/2011 SMAI ‐ HPC

slide-32
SLIDE 32
  • What are the answers ?
  • What needs to be assessed / improved ?

22/03/2011 SMAI ‐ HPC 32

slide-33
SLIDE 33

Multiscale ???

22/03/2011 SMAI ‐ HPC

slide-34
SLIDE 34

0.5 1 1.5 2 2.5 3 3.5 4 20 40 60 80 100 120 140

Number of CPUs Time (Hours)

Series1

128 processors : less than half hour 8 processors : a bit more than 3.5 hours

Parallelisation ??? (DMP etc.)

22/03/2011 SMAI ‐ HPC

slide-35
SLIDE 35

Infrastructure ???

1 ms

22/03/2011 SMAI ‐ HPC

slide-36
SLIDE 36

GPUs ??? For FE and mesheless !!!

22/03/2011 SMAI ‐ HPC 36

slide-37
SLIDE 37

Methodology ??? FE versus meshless for fracture

  • r

XFEM etc.

22/03/2011 SMAI ‐ HPC 37

slide-38
SLIDE 38

38

Mid‐sagittal section

HUMOS

Ultimate goal: be predictive at the system level

EC project HUMOS demonstrator of real car crashes

22/03/2011

SMAI ‐ HPC

slide-39
SLIDE 39

Towards virtual prototyping at large scale …

39

Courtesy of EADS‐CASA

slide-40
SLIDE 40

…and "full system" predictability

40

Model 1 32 processes Model 2 128 processes

22/03/2011 SMAI ‐ HPC

slide-41
SLIDE 41

22/03/2011 SMAI ‐ HPC 41