uu.seUppsala University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Treating model defects with a Gaussian Process prior for the parameters
Uppsala University, Disciplinary Domain of Science and Technology, Physics, Department of Physics and Astronomy, Applied Nuclear Physics.
Uppsala University, Disciplinary Domain of Science and Technology, Physics, Department of Physics and Astronomy, Applied Nuclear Physics.ORCID iD: 0000-0002-7595-8024
2017 (English)Conference paper, Oral presentation with published abstract (Other academic)
Abstract [en]

The covariance information in TENDL is obtained by propagating uncertainties of, e.g., TALYSparameters to the observables, and by attempting to match the parameter uncertainties to the experimental data. This results in model-driven covariances with strong energy‐energy correlations, which can lead to erroneously estimated uncertainties for both differential and integral observables.Further, the model driven approach is sensitive to model defects, which can introduce biases and underestimated uncertainties.To resolve the issue of model defects in nuclear data (ND) evaluation, models the defect with a Gaussian process. This can reduce biases and give more correct covariances, including weakerenergy‐energy correlations. In the presented work, we continue the development of using Gaussian processes to treat model defects in ND evaluation, within a TENDL framework. The Gaussian processes are combined with the Levenberg‐Marquardt algorithm for non‐linear fitting, which reduces the need for a prior distribution. Further, it facilitates the transfer of knowledge to other nuclides by working in the parameter domain. First, synthetic data is used to validate the quality of both mean values and covariances provided by the method. After this, we fit TALYS parameters and a model defect correction to the 56Fe data in EXFOR.

Place, publisher, year, edition, pages
2017.
National Category
Subatomic Physics
Identifiers
URN: urn:nbn:se:uu:diva-339285OAI: oai:DiVA.org:uu-339285DiVA, id: diva2:1175320
Conference
Workshop on TALYS/TENDL developments, 13‐15 November 2017, Prague
Available from: 2018-01-17 Created: 2018-01-17 Last updated: 2018-01-24Bibliographically approved

Open Access in DiVA

Presentation GP-model(6386 kB)8 downloads
File information
File name IMAGE01.pdfFile size 6386 kBChecksum SHA-512
5e1d44a1f1635a71d0acf6abf2203488d7a70d4bf699035e1aec1bc8ea68ad4f85b88163a467750554eb5e31a232df75a816c42813d4d4d80ca809e1831b078b
Type imageMimetype application/pdf
book of abstracts(217 kB)11 downloads
File information
File name SUMMARY01.pdfFile size 217 kBChecksum SHA-512
f9e26e39eeed259050614ff757a3a6a29587443a398461f0d4474858d54a3a2a184ab4e91aeb61cd0d24faac20261f92c5f85fe1d727910836fda7329fb76735
Type summaryMimetype application/pdf

Authority records BETA

Helgesson, PetterSjöstrand, Henrik

Search in DiVA

By author/editor
Helgesson, PetterSjöstrand, Henrik
By organisation
Applied Nuclear Physics
Subatomic Physics

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 32 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf