Post-graduate theses
Current Record: 29 of 299
|
Identifier |
000456628 |
Title |
Searching for a signature of turnaround in velocity profiles of galaxy clusters with machine learning |
Alternative Title |
Αναζητώντας την "υπογραφή" της ακτίνας αναστροφής στις ταχύτητες των γαλαξιών σε σμήνη με χρήση μηχανικής μάθησης |
Author
|
Τριανταφύλλου, Νικόλαος Π.
|
Thesis advisor
|
Παυλίδου, Βασιλική
|
Reviewer
|
Bonfini, Paolo
Κορκίδης, Γεώργιος
|
Abstract |
Context: ΛCDM is currently recognized as the standard model in cosmology, but recent
tensions between different types of high-accuracy data indicate potential inconsistencies.
This generates an urgent need for a new examination of this model using novel observables.
Pavlidou et al. have recently illustrated that the turnaround density, the average density
within the turnaround radius of galaxy clusters, can be utilized as a probe for this purpose.
Aims: We search for an efficient way of measuring the turnaround radius of galaxy clusters
on the plane of the sky.
Methods: We use the MultiDark Planck 2 and Virgo simulations to acquire projections of
galaxy clusters and calculate their turnaround radius based on their velocity profiles. We
use these data to train (and test) a Convolutional Neural Network (CNN), involving several
layers of filters, normalization and regularization techniques.
Results: We find that: (a) the turnaround radius is correlated to the central mass of a
galaxy cluster and that the mass distribution around the cluster is almost an irrelevant
feature to the model’s prediction. (b) The velocity dispersion of the galaxies also contains
information related to the turnaround radius. The model’s accuracy is not significantly
affected by the absence of information inside the R200 of the central overdensity in each
cluster.
|
Language |
English |
Issue date |
2023-07-28 |
Collection
|
School/Department--School of Sciences and Engineering--Department of Physics--Post-graduate theses
|
|
Type of Work--Post-graduate theses
|
Permanent Link |
https://elocus.lib.uoc.gr//dlib/f/5/2/metadata-dlib-1687763058-714550-7476.tkl
|
Views |
542 |