Enhancing Engineering Design and Analysis Interoperability Part 3: Steps toward Multi-Functional Optimization

1
Enhancing Engineering Design and Analysis
Interoperability Part 3 Steps toward
Multi-Functional Optimization
First MIT Conference on Computational Fluid and
Structural Mechanics Cambridge, Massachusetts
USA June 12-15, 2001

• Rod Dreisbach
• The Boeing Company
• Computational Structures Technology
• www.boeing.com

Russell Peak Georgia Tech Engineering Information
Systems Lab eislab.gatech.edu
2
Maturation of product life cycle knowledge
3
Typical Current Approach Optimize idealized
parameters (vs. detailed design)
Detailed Design Model
Analysis Model (with Idealized Features)
Channel Fitting Analysis
4
Multi-Functional Optimization (MFO)

• Term as coined at Boeing
• Multitude of operational functional requirements
• Concurrent consideration during product design
  process
• Idealized design variables used in optimization
  associated directly with product (detailed design)

5
Progress onNecessary Components

• Design-Analysis Integration
• CAD-CAE Associativity
• Connect diverse CAE models to same CAD model
• Varying discipline, behavior, fidelity, method,
  tool
• Multi-directional
• Object-Oriented View of Optimization
• Enhanced FEA Modeling for Built-Up Structure

6
X-Analysis Integration Techniques
a. Multi-Representation Architecture (MRA)
b. Explicit Design-Analysis Associativity
c. Analysis Module Creation Methodology
7
COB-based Constraint Schematic for
Multi-Fidelity CAD-CAE InteroperabilityFlap Link
Benchmark Example
8
Test Case Flap Linkage Analysis Template Reuse
of APM
Linkage Extensional Model (CBAM)
Flap link (APM)
reusable idealizations
9
Flap Link APMImplementation in CATIA v5
Design-Idealization Relation
Design Model
Idealized Model
10
CATDAK OverviewXaiTools CATIA Design-Analysis
Knowledge Manager
Design Idealizations (APM)
CAD-Analysis Template Coordination
Analysis Template Usage (CBAMs)
CATDAK
XaiTools
Analysis Inputs
CATIA Model
API
VBScripts
Analysis Outputs (Design Updates)
VBScripts
Analysis Templates
Traditional Solvers
API application programming interface
11
Updating CAD Model from Analysis Template Results
12
Progress onNecessary Components

• Design-Analysis Integration
• CAD-CAE Associativity
• Connect diverse CAE models to same CAD model
• Varying discipline, behavior, fidelity, method,
  tool
• Multi-directional
• Object-Oriented View of Optimization
• Enhanced FEA Modeling for Built-Up Structure

13
Thesis Abstract
14
Partition of Engineering Entities
Enhanced Optimization Model (EOM)
Math Opt Model
Engineering
Opt. Model
Solution Method Model
Find
Design variable
Notation
solder joint

height (h)
PWB material type
(a )
Maximize

Solder Fatigue life
THESIS FOCUS
Context-Based Analysis Model
Analysis Building Block
Printed Wiring Assembly (PWA)
Solution Method Model
Solder
body
T
Component
Component
1
0
Joint
body
body
Solder Joint
3
4
body
PWB
2
Printed Wiring Board (PWB)

Previous work
Analyzable
Peak et al. 2000,
Analysis Tools
Product Model
1999,
Tamburini
Design Tools
Wilson, 2000
15
Optimization Model Diversity
OPTIMIZATION MODEL CLASS
Optimization Object 1
Optimization Object 2
Min Weight
Min Weight
subject to
subject to
Stress Buckling Design variables Area, Material
Stress Design variables Area
1D EXTENSIONAL STRESS MODEL
Analysis Model(s) Enhancement and/or Addition
Objective, design variable, and/or constraint
function enhancement
OPTIMIZATION MODEL CLASS
Optimization Object 1
Optimization Object 2
Optimization Object 3
Min Weight
Min Weight, Cost
Min Weight
subject to
subject to
subject to
g (x)lt0
g (x)lt0
g (x)lt0
h(x) 0
h(x) 0
h(x) 0
X(H)
X(H,LL,LR)
X(H,LL,LR,Mat)
2D PLANE STRAIN MODEL
16
Optimization Model Enhancement
OPTIMIZATION MODEL I
Minimize
Weight
Subject to
Normal Stress Margin of Safety
Design variables
X
A
OPTIMIZATION MODEL II
Minimize
Weight
Subject to
Normal Stress Margin of Safety
Design variables
X
A, material
17
Minimization of Weight of a LinkageX(area)
subject to (extensional stress)
deformation model
Extensional Rod
(isothermal)
product structure
linkage
analysis context
L
al1
effective length,
eff
D
L
L
o
L
x
1
x
2
mode
shaft tension
cross section
area,
A
al2
A
material
linear elastic model
youngs
modulus,
E
al3
condition
linkage
s
E
density, r
flaps down
reaction
e
F
goal
optimization
minimize weight
weight,W
constraint
allowable stress
s
L
eff
MSstress
Design Variable
A
18
Minimization of Weight of a LinkageX(area,
material) subject to (extensional stress)
deformation model
Extensional Rod
(isothermal)
product structure
linkage
analysis context
L
al1
effective length,
eff
D
L
L
o
L
x
1
x
2
mode
shaft tension
cross section
area,
A
al2
A
material
linear elastic model
youngs
modulus,
E
al3
condition
linkage
s
E
density, r
flaps down
reaction
e
F
goal
optimization
minimize weight
weight,W
constraint
allowable stress
s
L
eff
MSstress
Design Variable
area,A
material
19
Optimization Model Enhancement
OPTIMIZATION MODEL III
OPTIMIZATION MODEL IV
Minimize
Weight
Subject to
Normal Stress Margin of Safety
Buckling Margin of Safety
Design variables
X
A, material
20
Minimization of Weight of a LinkageX(area)
subject to (extensional stress, buckling load)
deformation model
Extensional Rod
(isothermal, buckling)
product structure
linkage
analysis context
L
effective length,
eff
D
L
L
o
L
x
moment of inertia,
I
1
x
cross section
2
mode
shaft tension
area,
A
A
material
linear elastic model
youngs
modulus,
E
condition
linkage
s
E
density, r
flaps down
reaction
load,P
e
F
goal
optimization
minimize weight
Extensional Rod
weight,W
(Buckling)
constraints
L
o
E
P
s
cr
I
allowable stress
L
eff
MSstress
MSbuckling
Design Variables
A
21
Minimization of Weight of a LinkageX(area,
material) subject to (extensional stress,
buckling load)
deformation model
Extensional Rod
(isothermal, buckling)
product structure
linkage
analysis context
L
effective length,
eff
D
L
L
o
L
x
moment of inertia,
I
1
x
cross section
2
mode
shaft tension
area,
A
A
material
linear elastic model
youngs
modulus,
E
condition
linkage
s
E
density, r
flaps down
reaction
load,P
e
F
goal
optimization
minimize weight
Extensional Rod
weight,W
(Buckling)
constraints
L
o
E
P
s
cr
I
allowable stress
L
eff
MSstress
MSbuckling
A
Design Variables
material
22
Progress onNecessary Components

• Design-Analysis Integration
• CAD-CAE Associativity
• Connect diverse CAE models to same CAD model
• Varying discipline, behavior, fidelity, method,
  tool
• Multi-directional
• Object-Oriented View of Optimization
• Enhanced FEA Modeling for Built-Up Structure

23
Chip Package Products Shinko
Quad Flat Packs (QFPs)
Plastic Ball Grid Array (PBGA) Packages
24
Traditional VTMB FEA Model CreationManually
Intensive 6-12 hours
VTMB variable topology multi-body
FEA Model Planning Sketches - EBGA 600 Chip
Package
25
Advanced Product Information-Driven FEA
Modeling Challenges

• Main challenges
• Differences between design analysis geometries
• Variable topology multi-body geometries
• FEA requirements node matching, aspect ratio
• Relative body sizes
• Degree of indirect inter-body coupling
• Mixed analytical bodies
• Idealized inter-body interfaces
• Loads interfaces on non-explicit boundaries
• Idealization-induced anomalies
• Ex. - Shell mid-/outer-face matching
• Arbitrary shapes (complex 3D surfaces )

26
Multi-Representation Architecture Context

• Composed of four representations (information
  models)
• Provides flexible, modular mapping between design
  analysis models
• Creates automated, product-specific analysis
  modules (CBAMs)
• Represents design-analysis associativity
  explicitly

27
Approach Outline Test Cases

• Benchmark test cases
• diving board
• eWidget
• simplified PBGA
• Production test cases(representative
  production-like problems for industry)
• Chip package (Shinko)
• Thermal analysis - Phase 2
• Thermomechanical (stress) analysis - after Phase
  2
• Air frame structural analysis (Boeing)
• PWA/B (JPL/NASA,)
• Thermomechanical, ...

28
Chip Package Test Cases (for Shinko)
29
Airframe Structural Analysis Radar Support
Structure (for Boeing)
Automatic FEA Pre/Post-processing Solution (in
vendor-specific Solution Method Model)
Design Model
30
PWA Thermomechanical Analysis(for JPL/NASA, ...)
Goal Generalization of previous work Zhou, 1997
31
Summary

• Progress
• Design-Analysis Integration (maturing)
• CAD-CAE Associativity
• Connect diverse CAE models to same CAD model
• Varying discipline, behavior, fidelity, method,
  tool
• Multi-directional
• Object-Oriented View of Optimization (initial
  progress)
• Enhanced FEA Modeling for Built-Up Structure
  (in-progress)

• Further work needed
• High-level operational criteria, such as Product
  Design Requirements and Objectives
• Need to leverage recent optimization tools
• Ex. iSIGHT, ProductCenter, etc.
• Provide enhanced modularity knowledge capture