BEGIN:VCALENDAR
VERSION:2.0
PRODID:Events Booking
BEGIN:VEVENT
UID:664a3eb18b45a
SUMMARY:A Free Online VKI Seminar - Multi-scale chemically reacting flows:
model reduction and machine learning
DESCRIPTION:A Free Online VKI Seminar
Multi-scale chemically reacting flo
ws: model reduction and machine learning
Our Guest Speaker: Prof. Riccar
do Malpica Galassi\, assistant professor\, Sapienza University of Rome
T
his event is open to public in respect with the VKI eligibility criteria.
Before registering\, please check if you are eligible.
Abstract
High
-fidelity simulations of multi-dimensional turbulent reacting flows at lar
ge Reynolds and Karlovitz numbers with detailed chemistry and transport ar
e an essential tool to provide physical insights and to serve as a valu
able database for engineering model validation. Despite rapid advances i
n CPU/GPU high performance computing hardware\, the challenge remains due
to the large number of reactive scalar variables and the wide spectrum o
f spatial and temporal scales. Therefore\, there are continuing motivation
s to develop advanced reduced order models (ROM) to improve computational
efficiency while preserving the fidelity and predictive capability of the
simulations.
In describing the temporal evolution of chemically reacting
systems\, the large number of characteristic chemical time scales associa
ted with individual reaction pathways cause the stiffness problem\, often
demanding an unnecessarily large number of time steps to integrate the equ
ations to the desired practical time. In fact\, the fastest chemical s
cales are orders of magnitude smaller than the flow scales of interest.
Model reduction takes effect by recognizing that fast processes constrain
the slow dynamics to evolve on a lower-dimensional\, invariant manifold. T
o this end\, the computational singular perturbation (CSP) framework emplo
ys an eigenvalue decomposition of the local Jacobian matrix to identify th
e fast/slow spectral gap and projects the system onto the slow invariant m
anifold (SIM)\, allowing a significantly accelerated time integration with
an efficient explicit algorithm\, along with a corrective projection for
fast exhausted modes. In fact\, the slow dynamics\, free of the fast sca
les\, is not stiff anymore and evolves within the SIM at a pace which is
orders of magnitude larger than the system's fastest timescale. While t
he CSP-based solvers have demonstrated effective computational acceleratio
ns by orders of magnitude in time steps\, a major computational overhead
remains in the operation of the large Jacobian matrix to compute the loca
l CSP projection basis. Recently\, the renewed interest in CSP-based solve
rs has been catalyzed by the advent of machine learning techniques. Data
-driven approaches may be fruitfully employed to learn projection operato
rs and non-stiff latent spaces\, enhancing the efficacy of CSP in capturin
g the slow system dynamics and reducing computational complexities.
Biog
raphy
Riccardo Malpica Galassi is an assistant professor at Sapienza Uni
versity of Rome. He received his PhD from the Mechanical and Aerospace Dep
artment at the University of Rome in 2018\, where subsequently\, he spent
over three years as a post-doctoral fellow. Following this\, he spent two
years as a Marie Curie post-doctoral Fellow at the Aero-thermo-mechanics d
epartment at Université Libre de Bruxelles.He is currently working on the
physical understanding of the processes that characterize combustion phen
omena\, on reduced order models and digital twins\, on multi-scale adaptiv
e solvers\, on machine learning for combustion\, on uncertainty quantifica
tion\, and reactive flows CFD\, with special interest towards turbulent co
mbustion and spray combustion.
LOCATION:Sint-Genesius-Rode\, 1640 Flemish Brabant\, Belgium
DTSTART:20240613T140000Z
DTEND:20240613T150000Z
DTSTAMP:20220307T101637Z
ORGANIZER;CN=VKI Secretariat:MAILTO:secretariat@vki.ac.be
GEO:50.73507;4.396005
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Sint-Genesius-Rode\, 1640 F
lemish Brabant\, Belgium;X-APPLE-RADIUS=72;X-TITLE=von Karman Institute fo
r Fluid Dynamics:50.73507;4.396005
END:VEVENT
END:VCALENDAR