[Dottorcomp] Seminari di Matematica Applicata. Martedì 30 marzo, Chiara Piazzola

[Stefano Lisini]

Seminari di Matematica Applicata, Dipartimento di Matematica "F.
Casorati" e Istituto del CNR IMATI "E. Magenes" di Pavia,

Martedì 30 marzo 2021, alle ore 15 precise, la Dott.ssa Chiara Piazzola
(IMATI-CNR Pavia) terrà un seminario dal titolo:

Multi-fidelity computational approaches for the uncertainty
quantification of ship performance

Il seminario si terrà su Zoom al link:
https://us02web.zoom.us/j/87518137128?pwd=eHg2eG9QZUdydFJEU1NESWN5a1lPQT09

Abstract: Ship performance depends on design and operational
parameters (e.g. sea state, advancement speed, payload) which are
intrinsically uncertain. Therefore, such uncertainties have to be
taken into account when estimating ship performance indicators, such
as the resistance to advancement. The assessment of the impact of the
uncertainties on the performance indicators is known as forward
Uncertainty Quantification (UQ) analysis.
In this talk we present a comparison of two methods for the forward UQ
analysis of a passegers ferry advancing in calm water and subject to
two operational uncertainties, namely the ship speed and payload. Upon
choosing a configuration, i.e., fixing these two parameters, the
resistance to advancement can be obtained by solving the free-surface
Navier-Stokes equations. To this end, we employ a multi-grid Reynolds
Averaged Navier-Stokes (RANS) solver. A UQ analysis typically requires
solving several configurations of the parameters, and therefore can
become very expensive.
The two UQ methods compared in this work are the Multi-Index
Stochastic Collocation (MISC) and the multi-fidelity Stochastic Radial
Basis Functions (SRBF). The estimation of the expected value of the
(model-scale) resistance to advancement, as well as of its higher
order moments and probability density function, are presented and
discussed.
Both MISC and SRBF are multi-fidelity methods, i.e., they explore the
variability of the resistance to advancement by considering an
ensemble of RANS solvers with different accuracy levels (in other
words, employing meshes with different resolutions). More precisely,
in a first step they query the low-fidelity, and hence inexpensive,
solvers for several different configurations of the ferry operational
parameters, and perform the UQ analysis based on these configurations.
The results thus obtained are then corrected by further solving a
handful of configurations over the high-fidelity, and hence expensive,
solvers.

References:
[1] C. Piazzola et al. Uncertainty Quantification of Ship Resistance
via Multi-Index Stochastic Collocation and Radial Basis Function
Surrogates: A Comparison. Proceedings of the AIAA Aviation Forum 2020.
[2] J. Beck et al. IGA-based Multi-Index Stochastic Collocation for
random PDEs on arbitrary domains. Computer Methods in Applied
Mechanics and Engineering, 2019.
[3]J. Wackers et al. Adaptive N-Fidelity Metamodels for Noisy CFD
Data. Proceedings of the AIAA Aviation Forum 2020.
---------------------

Pagina web dei Seminari di Matematica Applicata
https://matematica.unipv.it/ricerca/cicli-di-seminari/seminari-di-matematica-applicata/