Product and System Optimisation

Incorporation of Product and System Optimization (PSO) Methods into Compact, Reliable Design Cycles

Workshop Description

This workshop is concerned with educational & awareness issues, including identifying barriers to successful uptake of emerging and relatively mature technologies. The workshop will be structured into two parts, a discussion session where issues are identified and technical presentations where specific methods are presented along with summaries of state of the art. The technical focus for this workshop is Design of Experiments and Response Surface Analysis. The workshop will also address the following questions:

• What are the real barriers to uptake?

• What can we conclude from the results of FENET survey on PSO? survey report

• What type of documentation for best practices is needed?

Presentations

Optimized and Reliable Structures - A Contradiction in Terms?
R. Helfrich (INTES GmbH - Germany)
The ESPRIT project ASRA-HPC
Integration of structural mechanics, optimization and reliability analysis

• solution of a combined optimization and reliability problem in an one-step approach

State-of the art technology for both domains

• High Performance Computing
• Adequate solution algorithms

GUI support for problem setup
Solution of industrially relevant problems

Discrete Variable Optimization using MSC.Nastran SOL200
Hans Sippel, MSC.Nastran Europe

• Discrete Variable Optimization
• Conventional optimization uses Mathematical Programming that yields continuous design variables
• These design variables are not immediately usable ( i.e. you cannot make a plate of thickness 0.37582 cm ).
• With discrete optimization, the user can specify thicknesses available: according to standard gauges
• Three methods: Conservative Discrete Design (CDD) / Design of Experiments (DOE) / Rounding up/off
• Implemented as a postprocessing step to a continuous solution, this means 1 additional FE – analysis