Uncertainty Quantification for Smooth Functional Data with Application to Material Properties

Williams, Brian J., Holland, Troy Michael, Phillips, Matthew Ryan · 2025

Read Uncertainty Quantification for Smooth Functional Data with Application to Material Properties by Williams, Brian J., Holland, Troy Michael, Phillips, Matthew Ryan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/doe-osti-3003145-uncertainty-quantification-for-smooth-fu.txt.

Description

Brian J. Williams, Troy Michael Holland, and Matthew Ryan Phillips of Los Alamos give a methods roadmap in report LA-UR-25-31138, issued November 12, 2025. The problem is uncertainty quantification for smooth functional data: experiments whose output is a whole curve, such as strain versus stress, rather than a single number. Material property studies produce curves like these by the hundreds.

The recipe runs in steps. Curves are first registered onto a common set of index values. Principal component analysis then compresses them, keeping the few modes that explain at least a user-set share of variance, typically 99 percent. Regression on each mode's scores ties curves to experiment settings. Sampling from the fitted model yields predicted mean and bounding curves at new settings. An example uses stress and strain measured on a foam core punch.

Catalog record

CREATORWilliams, Brian J., Holland, Troy Michael, Phillips, Matthew Ryan
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE29 KB
LICENSEPublic Domain
RIGHTSPUBLIC DOMAIN
SOURCEDOE OSTI

Source: https://www.osti.gov/biblio/3003145

Browse the full library · About the Tower of Babel