Training

• Oasys Software
  • Oasys PRIMER - An Introduction
  • Oasys PRIMER - Spotwelds and Connections
  • Oasys PRIMER - Seat and Dummy Positioning
  • Oasys PRIMER - Automatic Assembly of Multiple Crash Cases
• LS-DYNA Basic
  • LS-DYNA Introductory Course
  • LS-DYNA Advanced Impact Analysis (non-automotive)
  • Oasys LS-DYNA Environment Automotive Crash Modelling
• LS-DYNA Specialist
  • Polymeric Material Modelling in LS-DYNA
  • Crashworthiness Simulation using LS-DYNA
  • Blast & Penetration Modelling in LS-DYNA
  • Implicit Capabilities of LS-DYNA
  • Fluid / Structure Interaction in LS-DYNA
  • Heat Transfer & Thermal-Stress Analysis in LS-DYNA
  • Smoothed Particle Hydrodynamics (SPH) in LS-DYNA
  • Airbag Modelling with LS-DYNA
• Other Training
  • Optimization with LS-OPT
  • Stochastic Analysis with LS-OPT
• Worldwide Courses

• Distributors
• Contact
• Links
• Careers

Search

Stochastic Analysis with LS-OPT

Given by Dr. Heiner Müllerschön of DYNAmore GmbH

Duration: 1 Day

Preceding participation in the seminar "Optimization with LS-OPT" is recommended

Robustness and reliability analysis are of particular importance to investigate and improve the quality of your FE-models. For this, several methodologies are available in LS-OPT.

Single FE-simulations always leads to just one specific response without considering variations due to uncertainties of input parameters. Typically, scatter in the input parameters is present for material properties, contact conditions, sheet thicknesses or geometrical values.

By considering such input variations for the simulation engineer usually the following questions arise:

1. What is the range/distribution of the variation of the simulation responses? Is the model still robust?
2. How is the significance/contribution of parameters (input) with respect to specific responses (output)?
3. Is the model reliable with respect to constraints (failure, damage, requirements, ...)?
4. What is the probability that a particular failure limit exceeds?

LS-OPT is a useful software tool in finding solutions to these questions. The methodologies in LS-OPT, such as Monte Carlo simulation using Latin Hyper Cube sampling, Meta-model based Monte Carlo simulations or reliability analysis, are discussed in detail in this seminar. Of course, as prerequisite basic principles of statistic and probability theory are addressed as well. In a workshop the participants have the possibility to practice the application of stochastic analysis with LS-OPT. 

Content:

• Introduction, Terminology
• Basic principles of statistic and probabilistic theory
  • Confidence Intervals
  • Correlation Coefficients
  • Monte Carlo vs. Latin Hypercube Sampling
• Stochastic methodologies
• Monte Carlo analysis
• Monte Carlo analysis using Meta-Models
• Statistical distributions
• Normal (Gauß)
• Weibull
• Uniform
• Lognormal
• User defined
• Reliability analysis
• Interpretation of Ant-Hill Plots
• Separation of deterministic and chaotic responses
• Variance and correlation plots
• Post-Processing in LS-OPT and result interpretation
• Visualization of statistical quantities on the FE-model
• Examples

[back to course list]