From Uncertainty Propagation to Uncertainty Quantification: Using Analytics to Take Uncertainties out of Engineering Simulations

On Demand

Presented by Dr. Mark Andrews, SmartUQ UQ Technology Steward

He is responsible for advising SmartUQ on the industry’s uncertainty quantification needs and challenges. Dr. Andrews is SmartUQ’s principal investigator for the Probabilistic Secondary Flow and Heat Transfer Model project as part of the Probabilistic Analysis Consortium for Engines (PACE). Prior to working at SmartUQ, Dr. Andrews spent 15 years at Caterpillar where he worked as a Senior Research Engineer, Engineering Specialist in Corporate Reliability, and Senior Engineering Specialist in Virtual Product Development. He has a Ph.D. and M.S. in Mechanical Engineering from the New Mexico State University, as well as a BS in Mechanical Engineering from the University of New Mexico.

A common approach to assessing the variability in an engineering simulation model is to conduct Uncertainty Propagation. This method represents known input variabilities using probability distributions and then randomly samples from these distributions using Monte Carlo routines. Another method of performing Uncertainty Propagation utilizes predictive models (a.k.a., emulators or surrogate models) to propagate the uncertainties from the inputs.

In addition to propagating input uncertainties, Uncertainty Quantification (UQ) provides a more comprehensive framework for assessing various sources of uncertainties in engineering simulations. This framework includes several key analytics techniques:

Building predictive models: trained to mimic complex engineering simulations.

Sensitivity Analysis: ranks parameters by their ability to influence the results.

Statistical Calibration: handles the disagreement and uncertainty between the simulation model and physical tests.

Inverse Analysis: determines an underlying distribution for ill-conditioned and sparse model input.

This webinar will discuss and compare sampling-based and emulator-based methods for Uncertainty Propagation. Using example problems and software demonstrations for illustration, the webinar will also show how using additional UQ methods improves the decision-making process. The audience for the webinar includes engineers, data scientists, and managers who want to learn more about the methods and benefits of quantifying uncertainties in their engineering simulations.

Webinars

From Uncertainty Propagation to Uncertainty Quantification: Using Analytics to Take Uncertainties out of Engineering Simulations

Questions?