Bayesian calibration and probability bounds analysis solution to the Nasa 2020 UQ challenge on optimization under uncertainty

A. Gray, A. Wimbush, M. DeAngelis, P. O. Hristov, E. Miralles-Dolz, D. Calleja, R. Rocchetta

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)

Abstract

Uncertainty quantification is a vital part of all engineering and scientific pursuits. Some of the current most challenging tasks in UQ involve accurately calibrating, propagating and performing optimisation under aleatory and epistemic uncertainty in high dimensional models with very few data; like the challenge proposed by Nasa Langley this year. In this paper we propose a solution which clearly separates aleatory from epistemic uncertainty. A multidimensional 2nd-order distribution was calibrated with Bayesian updating and used as an inner approximation to a p-box. A sliced normal distribution was fit to the posterior, and used to produce cheap samples while keeping the posterior dependence structure. The remaining tasks, such as sensitivity and reliability optimisation, are completed with probability bounds analysis. These tasks were repeated a number of times as designs were improved and more data gathered.

Original languageEnglish
Title of host publication30th European Safety and Reliability Conference, ESREL 2020 and 15th Probabilistic Safety Assessment and Management Conference, PSAM 2020
EditorsPiero Baraldi, Francesco Di Maio, Enrico Zio
PublisherResearch Publishing Services
Pages1111-1118
Number of pages8
ISBN (Electronic)9789811485930
Publication statusPublished - 2020
Event30th European Safety and Reliability Conference, ESREL 2020 and 15th Probabilistic Safety Assessment and Management Conference, PSAM 2020 - Venice, Virtual, Italy
Duration: 1 Nov 20205 Nov 2020

Conference

Conference30th European Safety and Reliability Conference, ESREL 2020 and 15th Probabilistic Safety Assessment and Management Conference, PSAM 2020
Country/TerritoryItaly
CityVenice, Virtual
Period1/11/205/11/20

Bibliographical note

Funding Information:
This research is also funded by the Engineering & Physical Sciences Research Council with grant no. EP/R006768/1, “Digital twins for improved dynamic design”. The authors would also like to acknowledge the gracious support of this work through the local authorities under grant agree- ment “ITEA-2018-17030-Daytime”. This work has been carried out within the EUROfusion Consortium and Euratom research and training programme under grant agreement No 633053. The views and opinions expressed herein do not necessarily reflect those of the European Commission.

Publisher Copyright:
Copyright © ESREL2020-PSAM15 Organizers.Published by Research Publishing, Singapore.

Keywords

  • 2nd-order distribution
  • Bayesian calibration
  • Epistemic uncertainty
  • Optimization under uncertainty
  • Probability bounds analysis
  • Uncertainty propagation
  • Uncertainty reduction

Fingerprint

Dive into the research topics of 'Bayesian calibration and probability bounds analysis solution to the Nasa 2020 UQ challenge on optimization under uncertainty'. Together they form a unique fingerprint.

Cite this