MDOB Overview

1
MDOB Overview

• Dave Rudy
• Assistant Head, MDOB

Integrated Design Center Meeting Feb. 27, 2003
2
MDO Definition

• Multidisciplinary Design Optimization (MDO) is
• a collection of methodologies
• for the design of complex engineering systems and
  subsystems
• that coherently exploits the mutually interacting
  subsystems synergism

3
MDR Conceptual Elements
Analysis and Approximations
Organization and Culture
Design Problem Formulations and Solutions
Information Processing and Management
Cost-fidelity trade-off
Design problem formulation
Software engineering
MD Training
Approximations
Problem de- /re- composition
Data and software standards
MD process insertion
High-fidelity results inclusion
Optimization algorithms
Data visualization, storage, management
MD in integrated product teams
Parametric product data models
Optimization procedures
MD environment
MD in existing organization
SD analysis and sensitivity analysis
SD optimization
Human interface

• based on Special Session on Industry Needs at
  1998 AIAA/MAO (AIAA Paper 98-4737)

MD analysis and sensitivity analysis
MD optimization
MD computing
4
(No Transcript)
5
(No Transcript)
6
Approximation Management

• Accomplishment
• A rigorous mathematical framework was developed
  for exploiting low-fidelity models to reduce the
  cost of high-fidelity optimization while still
  obtaining the correct optimum
• Demonstrations for aerodynamic optimization have
  been made for various types of low-fidelity
  models, including response surfaces, coarse grids
  and simpler physics
• Significance
• Navier-Stokes optimization using unstructured
  grids was sped up by a factor of 5 for
  multi-element airfoils and by a factor of 3 for
  3-D wings
• This approximation and model management framework
  is used in LaRCs FAAST project for reducing the
  time of CFD analysis and design

7
3D Aerodynamic Design with AMMO

• Accomplishments
• Implemented efficient FUN3D-based aerodynamic
  optimization by judicious combination of
  higher-fidelity NS analyses and lower fidelity
  Euler analyses.
• Significance
• Greatest expected benefit of AMMO is in reducing
  the cost of extremely expensive, 3D design
  problems this implementation demonstrated
  approximately a 4-fold saving in high-fidelity
  analyses and sensitivity analyses
• Confirms results obtained earlier with FUN2D and
  CFL3D

8
CAD-Based Cost Estimation for a BWB

• Accomplishments
• Process assesses manufacturing cost for novel
  vehicle concepts
• Manufacturing cost models based on 10 years of
  NASA investment in physics-based manufacturing
  cost model.
• Significance
• Physics-based cost models provide realistic
  estimates for novel vehicle concepts
• Enabling technology brings manufacturing cost
  considerations directly into conceptual design
  processes.

9
Rapid Generation of Feasible Conceptual
Spacecraft Designs

• Accomplishments
• Automation of process and addition of a genetic
  algorithm to SmallSAT enable very rapid
  examination of many diverse combinations of
  spacecraft design variables
• Significance
• No conceptual spacecraft design tools known to
  have this automated capability.
• GA can be integrated with a conceptual spacecraft
  design tool in a Windows NT spreadsheet
  environment.

10
Legacy Code IntegrationApplication Trimmed
Aeroelastic Analysis

• Accomplishment Description
• Used Rational Unified Process to capture software
  requirements
• Developed integrated process from 25 individual
  lower level processes
• Implemented with UNIX Script, Python Script, and
  Model Center
• Significance
• Demonstrated disciplined yet fast turnaround
  integration approaches to successfully integrated
  off-the-shelf codes

11
Probabilistic RLV Design

• Accomplishment Description
• Tracked uncertainties due to variability in shape
  variables and uncertainty in aerodynamic
  parameters on minimum weight and pitching
  constraint.
• Demonstrated the sensitivities of the pitching
  moment constraint at Mach 10 to the uncertain
  aerodynamic data.
• Significance
• Application to conceptual multidisciplinary
  design
• Probabilistic formulation of design for
  controllability

12
In Conclusion

• Visit us at http//mdob.larc.nasa.gov