Webinars

Many analytics techniques that can be applied to simulation models, including optimization algorithms and uncertainty propagation studies, require large numbers of model evaluations to be effective. For many simulation models, the computational cost makes it infeasible to run the required number of simulations.

Uncertainty Quantification (UQ) tools can quantify these uncertainties while significantly reducing the number of simulations or tests required to get meaningful results. UQ tools help engineers analyze the difference between models and real-world performance so they avoid premature failure, costly warranty payments, and emergency redesigns.

This demonstration webinar will illustrate how the UQ toolset in the SmartUQ software determines the geometric parameters to minimize the bracket’s weight subject to fatigue life and maximum displacement under loading constraints.

This 30-minute webinar will cover:

• Creating and using a Design of Experiments (DOE) to select the best simulations to run for training data acquisition
• Training of advanced emulators (predictive models) to handle discontinuous output data caused by multiple bracket failure modes
• Using adaptive design techniques for efficient emulator accuracy improvement
• Propagating uncertainty about the optimum solution to account for uncertainties in material properties and the manufacturing of the optimal geometry