Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

ACMD Seminar: Uncertainty aggregation through model development and assessment towards prediction

Andrew White
Thermal Fluid Systems, Rolls-Royce Corporation

Tuesday, July 11, 2023, 3:00-4:00 PM ET (1:00-2:00 PM MT)

A video of this talk will be made available to NIST staff in the Math channel on NISTube, which is accessible from the NIST internal home page. It will be taken down from NISTube after 12 months at which point it can be requested by emailing the ACMD Seminar Chair.

Abstract: The goal of developing and exploiting physics-based models of engineered products and systems is to predict and understand the behaviors of the product/system to an adequate level of accuracy. The models incorporate many simplifying assumptions and approximations, and thus introduce many possible sources of error and uncertainty. Therefore, physical testing remains an important source of truth to ensure the validity of the model-based design. However, the test and measurement of complex systems also provides imperfect information due to constraints in the representation of the full system, the operational environment, and measurement error and uncertainty. The methods of verification, validation, and uncertainty quantification (VVUQ) are focused on the estimation of both model and measurement sources of uncertainty. Much of the published literature focuses on certain individual aspects of the VVUQ process. A few studies have tried to combine the contributions of these sources. However, gaps exist when attempting to apply these frameworks in industrial applications, e.g., multivariate model outputs. This research therefore pursues a Bayesian paradigm for end-to-end aggregation of multiple sources of error and uncertainty, considering all the steps of verification, validation, and uncertainty quantification. The process is demonstrated using typical gas turbine engine heat transfer finite element model of a turbine disc. The uncertainty aggregation process results in a demonstrated, pragmatic solution to the estimation of 'error bounds' for the final model predictions, while also identifying potential areas for model improvements.

Bio: Andrew White is a technical specialist in aerospace thermofluid and thermomechanical analysis at Rolls-Royce. He has a B.S. in Physics and M.S. Mechanical Engineering from Purdue University (2006). After spending over a decade gaining expertise in various kinds of physics-based modeling and simulation, he began focusing on the application of verification, validation, and uncertainty quantification (VVUQ). He earned a Robust Design Green Belt (2015) by demonstrating probabilistic heat transfer, turbine disc stress and life analysis. Andrew completed a PhD from at Vanderbilt University (2022) focused on the application of VVUQ methods to gas turbine modeling and simulation. Key focus areas include surrogate modeling, Bayesian inference, and validation metrics. He continues to advance the use of VVUQ within Rolls-Royce.

Host: Zach Grey

Note: This talk will be recorded to provide access to NIST staff and associates who could not be present to the time of the seminar. The recording will be made available in the Math channel on NISTube, which is accessible only on the NIST internal network. This recording could be released to the public through a Freedom of Information Act (FOIA) request. Do not discuss or visually present any sensitive (CUI/PII/BII) material. Ensure that no inappropriate material or any minors are contained within the background of any recording. (To facilitate this, we request that cameras of attendees are muted except when asking questions.)

Note: Visitors from outside NIST must contact Lochi Orr at least 24 hours in advance.


Created June 6, 2023, Updated July 12, 2023