Summary of Activities

1
System Analysis towards System Optimization
Multi-disciplinary Design Optimization K
Sudhakar Amitay Isaacs SK
Sane J Jayaraman
PM Mujumdar ARDB Centre for Systems Design
Engineering Department of Aerospace
Engineering Indian Institute of Technology
Bombay October 13, 2003
2
Multi Disciplinary Analysis (MDA) towards System
Optimization Multi-disciplinary Design
Optimization K Sudhakar Amitay Isaacs
SK Sane J Jayaraman
PM Mujumdar ARDB Centre for
Systems Design Engineering Department of
Aerospace Engineering Indian Institute of
Technology Bombay October 13, 2003
3
CASDE - Activities

• MTech in Systems Design Engineering
• Courses of study
• System Modeling Simulation
• Optimization for Engineering Design
• Multi-disciplinary Design Optimization (MDO)
• Applied Mechatronics (hands on course)
• Statistical Methods for Analysis Design
• Systems Engineering Principles  
• Also available as short courses
• Coordinates a Special Interest Group on MDO
  (SIG-MDO)

4
Systems Engineering Process

• System represents system under design
  development
• Super-system represents operational scenario
  for the systems
• under design development.

5
Systems Engineering Process
Context
Super System
Requirements to lower level
System
Solution from lower level
Sub System

• System represents system under design
  development
• Super-system represents operational scenario
  for the systems
• under design development.

6
Systems Engineering Process
Context
Super System
Requirements to lower level
System
Solution from lower level
Sub System

• Level 1 Good understanding of system
  knowledge base, heuristic
• Computationally less
  expensive
• Usually not available for
  new systems.
• Level 3 Physics based modeling
  computationally intensive,
• applicable to new systems
  (VV?)

7
Systems Engineering Process
Context
Super System
Requirements to lower level
System
Solution from lower level
Sub System
8
CASDE Activities

• Research activity
• High fidelity models in design loop ( CFD, . .)
• Multi-Disciplinary Analysis (MDA) leading to
• Multi-disciplinary Design Optimization (MDO)

• MDO Studies
• Hypersonic Vehicle (DRDL)
• Combat Aircraft (ADA)
• Launch Vehicle (VSSC)

9
System Optimization / Design

• Design Problem Statement
• Not always readily available
• Not easy to reconstruct
• Need for capturing problem as we go along
• Optimization
• System parameterization, x
• Objectives Optimize f1(x), f2(x), . .
  . .
• Constraints Equality h1(x), h2(x), . .
  . . 0
• Inequality g1(x),
  g2(x), . . . . ? 0
• Multi Disciplinary Analysis ? f1(x), . , h1(x),
  . , g1(x), .

10
System Optimization / Design

• Multi Disciplinary Analysis Capability
• Strengths exist in disciplinary analysis
• Focus on Analysis for Design
• Capture knowledge with traceability
• Focus on verification / validation

11
System Optimisation / Design

• Need for a group to
• Capture design problems
• Define needs for Analysis for Design
• Extract / Establish traceability
• Perform Verification / Validation
• Explore design methodologies

SIG-MDO is doing this informally MSO-DMES
gives hopes of formalizing this
12
System Optimisation / Design (Contd.)

• IT Enabled Design Environment
• Stress on digital models
• Need for distributed environment
• Support to system designer for abstracting
• Enabling disciplinary authority / control
• MDO Frameworks!
• Optimization Technology
• Goes beyond theory
• Goes beyond S/W packages
• Guided search to improve than hunt for optima

13
System Level Analysis
14
System Level Analysis
15
System Level Analysis
16
System Level Analysis
17
Design Problem

• Design an Air-Breathing Hypersonic Vehicle
• Cruise M6.5 at H 30-35 km
• Constraints
• Dimensional constraints on overall length, height
  and width
• Take-off gross weight
• Intake entry conditions
• Control deflection within allowable values
• Vehicle drag to be less than thrust deliverable

18
Parameterization of HSTDV Body
Design variables XD ?1, ?2, ?3 , ?n_plan , ?wc
, wfac_pl, tfac_pl,, Hcruise
19
Hypersonic Vehicle Discipline
Interactions
Analysis Model
Output
Input variable
Ext. Compression Model AM1
?1
Y1 l1, l2, l3, h1, h2, h3 Y2 ma , MI, ?, pst
Y1
?2
Ext. Configuration Model AM2
Y3 (X,Y,Z) Y4 TOGW , C.G., Vol, Fuel mass
?3
Aero Model AM3
Y5 CN, Cm, CA
?n_pl
Trim Model AM4
Y6 TOGW_up, ?T , ?T , D
?w_c
SW
Thrust Model AM5
Y7 Th_deliv, Lp, Mp
ST
Hcr
Performance Model AM6
Y8 Cruise Range
?n_pl, SW ?1, Hcr
Variables not shared
Y1 Response from AM1 required as
input in AM2
Shared variables
20
System Level Analysis
21
MDO Framework

• MDO Framework is an important infrastructure for
  design
• Do MDO Framework exist?
• iSIGHT,
• Phoenix integration,
• etc
• Most design offices are migrating to MDO Framework

22

• System Engineering Processes
• Mr J Jayaraman
• MDO Framework for System Optimization
• Mr. Amitay Isaacs
• Contd. . . .