Reduced-order modeling in uncertainty quantification and computational fluid dynamics - Part III of III
For Part I, see MS-Mo-D-32
For Part II, see MS-Mo-E-32

Date: August 12
Time: 16:00--18:00
Room: 306A

(Note: Click title to show the abstract.)

Chen, Peng (ETH Zurich (Swiss Federal Inst. of Tech. in Zurich))
Quarteroni, Alfio (EPFL)
Rozza, Gianluigi (SISSA, International School for Advanced Studies)

Abstract: This proposed minisymposium is about the development and application of reduced-order modeling techniques in the fields of uncertainty quantification and computational fluid dynamics for control, optimization and design. Large-scale computing is commonly faced in these fields due to the high computational complexity of solving parametric and/or stochastic systems described by, e.g. partial different equations, which may lead to unaffordable computational burden for real-world application. In order to tackle this challenge, reduced-order modeling (e.g. RB, POD, EIM, PGD) techniques with the aim of capturing and utilizing the most important features of these systems are particularly in need for real-time and/or many-query computing.

This minisymposium focuses on the development and application of reduced-order modeling techniques in following themes: 1. efficient and reliable a posteriori error estimates for reduced solution and output; 2. forward uncertainty quantification problems, e.g. sensitivity analysis, risk prediction or reliability analysis with scientific and engineering applications; 3. stochastic inverse problems (model calibration, parameter identification) by variational or Bayesian approach; 4. control, optimization and design in computational fluid dynamics possibility under uncertainties.

Reduced-order modeling techniques have undergone fast development during the last decade and become a new frontier in scientific computing. Their increasing popularity is witnessed by many minisymposia at congress and conferences around the world, such as ICIAM, ICOSAHOM, WCCM, SIAM CSE, SIAM UQ, ECCOMAS, ENUMATH. The aim of this minisymposium is to discuss the most recent development of these techniques with emphasis in the field of UQ and CFD and identify new directions and perspectives. For this purpose we have invited 12 speakers with great expertise from several universities around the world, e.g. (MIT, Stanford, Paris VI, EPFL, TU Munich, CAS, Sandia National Laboratories, etc.)

Different approaches for the approximation by reduced basis approximations
Mula, Olga (RWTH Aachen)
Maday, Yvon (Laboratoire J.-L. Lions, Univ. Pierre et Marie Curie)

Stochastic collocation methods on unstructured meshes and their applications to UQ
ZHOU, TAO (AMSS, the Chinese Acad. of Sci.)
Narayan, Akil (Univ. of Massachusetts Dartmouth)

Hierarchical Bayesian Learning for Two-Dimensional Turbulent Bottom Gravity Currents
Lin, Jing (Massachusetts Inst. of Tech.)
Lolla, Tapovan (Massachusetts Inst. of Tech.)
Haley, Patrick (Massachusetts Inst. of Tech.)
Lermusiaux, Pierre (MIT)


Code: Type-Date-Time-Room No.
Type : IL=Invited Lecture, SL=Special Lectures, MS=Minisymposia, IM=Industrial Minisymposia, CP=Contributed Papers, PP=Posters
Date: Mo=Monday, Tu=Tuesday, We=Wednesday, Th=Thursday, Fr=Friday
Time : A=8:30-9:30, B=10:00-11:00, C=11:10-12:10, BC=10:00-12:10, D=13:30-15:30, E=16:00-18:00, F=19:00-20:00, G=12:10-13:30, H=15:30-16:00
Room No.: TBA