RollsRoyce_Users

Integrating Frameworks and their use for Optimisation and Robust Design
A. Karl, G. May, C. Barcock, G. Webster, N. Bayley, Rolls-Royce plc R. Parchem, B. Meissner, Rolls-Royce Deutschland
?2005 Rolls-Royce plc The information in this document is the property of Rolls-Royce plc and may not be copied or communicated to a third party, or used for any purpose other than that for which it is supplied without the express written consent of Rolls-Royce plc. This information is given in good faith based upon the latest information available to Rolls-Royce plc, no warranty or representation is given concerning such information, which must not be taken as establishing any contractual or other commitment binding upon Rolls-Royce plc or any of its subsidiary or associated companies.

Contents
Rolls-Royce 5 steps in optimisation VIVACE
Introduction Forum 1 demonstrations
2
Optimisation and Robust Design Applications
optimisation of blade – disc connections robust analysis for structural casings
Conclusion
Rolls-Royce data – proprietary
A. Karl, October 2005

3
Civil Aerospace
Energy
Defence
Marine
Rolls-Royce data – proprietary A. Karl, October 2005

5 steps in Optimisation
Automate Process Process Integration
build up of integrated processes between various disciplines
4
Integrating Frameworks
execute design / analysis process without human interaction
Design Exploration
getting an understanding of the design space characteristic
Optimisation
achieve the best compromise regarding all requirements
Robust Design
make sure that the design performs for variable conditions
Optimisation and Six Sigma
Rolls-Royce data – proprietary
A. Karl, October 2005

VIVACE - Introduction
4 year project started in 2004 (https://www.wendangku.net/doc/8217566644.html,) ~70 partners ~75 M€ 3 main areas Aircraft Engine Advanced Capabilities Forum 1: first demonstration 20th/21st September 2005 initial results of the first 18 month of research work Forum 2 planned for next year Forum 3 planned for end of project
5
Rolls-Royce data – proprietary
A. Karl, October 2005

Component Optimisation
6
CAD
Rolls-Royce data – proprietary
FE
Data Evaluation
A. Karl, October 2005

Virtual Enterprise Collaboration Hub
Other disciplines...
WP2.4, PROOSIS
7
OEM
Performance
T2.3.3 T2.3.4
T2.3.1
WEM; CAD Based
Structural WEM
Suppliers
WP3.2 T2.3.5
Component Optimisation
Company A Company B Company ...
Local Design on Component
D
d uteDesign on Component Local trib is
Company A Company B Company ... ínterfaces Loads Basic Geometry
n sig De
ask t
ínterfaces Loads Basic Geometry
ínterfaces Loads Basic Geometry
Geometry
Geometry
WEM
Geometry
Geometry
WEM
WP3.6
Preliminary Design
Concept Design
Detailed Design
Rolls-Royce data – proprietary
A. Karl, October 2005
Iterative Design Process...

Distributed Design Task
Commercial tools organised into a coherent system supporting the business processes are used Shared data repository in Share-a-Space from Eurostep Solutions Process integration and workflow in FIPER from Engineous Software All communication between the companies is via the internet (no direct links) For the demonstration (Forum 1 Scenario) the data was stored in Stockholm a dimensional reduction service located in Belfast was used a mail service located in USA was used and the compute power was provided via laptops in Warwick
8
Rolls-Royce data – proprietary
A. Karl, October 2005

Distributed Design Task Process (Forum 1 Scenario)
9
e-mail for design update data download dimensional reduction
Rolls-Royce data – proprietary A. Karl, October 2005

Optimisation and Robust Design
Using the integrated processes, application of automated Optimisation and Robust Design methods and tools becomes possible
10
Reduced design time Decreased overall product costs Improved manufacturability Less failures and increased customer delight
Application examples
Optimisation of a blade – disc connection Robust analysis of a structural casing
Rolls-Royce data – proprietary A. Karl, October 2005

Optimisation: process definition
11
single sector: life & crushing stresses
parametric geometry
automated analysis process
Rolls-Royce data – proprietary
unzipping model
A. Karl, October 2005

Optimisation: Results of NSGA II
failed runs, automatically deselected by the software
12
Pareto front (best possible combinations)
Unzipping Reserve
improvement possibilities
manual optimisation
Stress
Multi-objective optimisation goals
Rolls-Royce data – proprietary A. Karl, October 2005

Robustness Analysis: Toolset
FE Text Input Files
13
SC03
Walker-Strain
UG – CAD
Scenario Part Filea Weight
Manufacturing Parameters & Tolerances
STEP Translator STEP File
Manufacturing Stress / Life
Excel – CAPRA Surrogate Capability Scrap & Rework Cost Manufacture Cost
Part Parameter & Expression File
Weight
Derived Manufacturing Parameters Design Parameters
Geometry
iSIGHT
Design Parameters & Tolerances Boundary Conditions
Cost
Decision Pro Casing Cost
Check Objectives to Exit Loop
Rolls-Royce data – proprietary
A. Karl, October 2005

Robustness Analysis: Results
14
typical input distributions for heat transfer coefficients Monte-Carlo Simulation typical response distributions
nominals are OK (red lines) but: significant amount of “designs” are not fulfilling the requirements
“Six Sigma” Analysis of data
Rolls-Royce data – proprietary A. Karl, October 2005

Typical Productivity Benefits
Structural Casing Example:
Set-up of automated models 5 design iterations DfPE Summary Overall design iterations Development time reduction Manual Process 0 days 10 days Not currently carried out 10 days 5 Automated Process 4 days 1 day 2 days 7 days ~50 3 days
15
30% savings in time plus 10x increase in design iterations
Blade – Disc connection:
analysis time down from 2 hours to 5 min
Rolls-Royce data – proprietary
A. Karl, October 2005

Conclusion
Commercial integrating frameworks are now capable for applications in real business process applications Optimisation and Robust Design methods can be employed for “real world” problems
16
Development time and cost reductions Improved design quality Increased customer delight
Rolls-Royce data – proprietary A. Karl, October 2005