Slide contents
- Agenda
- Motivation for parametric variation
- Motivation
- Understanding alternative designs
- Understand a design
- Engineering a design
- Benefits of a parametric design variation
- Design improvement
- Design improvement- goals
- Improve conflicting properties -1
- Improve conflicting properties -2
- Dealing with tolerances -1
- Dealing with tolerances -2
- Matching simulation and test - 1
- Matching simulation and test - 2
- System simulation
- Multiphysics simulation based on system coupling
- Model reduction for Nonlinear Components
- Behind ptiSLang - dynardo
- Understand your design - parametric workflow in ANSYS
- Multiphysics analysis of an eletric-thermal actuator
- The ANSYS workbench philosophy
- Thermal-electric actuator
- Where to get the parameters
- CAD-MODEL variation
- Which CAD system provides parametric interfaces?
- CAD parameters
- Use the SpaceClaim direct modeler
- Parametric material modeling
- MS Excel
- Fully automated simulation workflows in APDL-1
- Fully automated simulation workflows in APDL-2
- Understand your design
- Example: Notch -1
- Example: Notch -2
- Example: Notch -3
- Manual variations
- Manual variations - normally 3designs
- Manual vs automatic sampling
- The automatic sampling
- Understand your design - typical questions
- Content
- How to evaluate 1000 designs? -1
- How to evaluate 1000 designs? -2
- How to evaluate 1000 designs? -3
- How to evaluate 1000 designs? -4
- How to evaluate 1000 designs? -5
- How to evaluate 1000 designs? -6
- The optiSLang meta-model of optimal prognosis (MOP)
- Coefficient of prognosis (CoP)
- The optiSLang meta-model of optimal prognosis (MOP)
- Coefficient of prognosis (CoP)
- Accuracy and numerical noise
- Reviewing the results
- Robust parameter settings
- BUT - Do we need always converging and regeneratable models?
- Restart option
