
=

The
modeling &
simulation (M&S)
research theme focuses
mainly on fundamental
research of algorithms
and mathematical models
of computer graphics,
visualization, and
visual analytics, about
static and dynamic
descriptions in the
data. Projects in
M&S are either
standalone or derived
from/related to projects
from the other four
research themes.

To
date, M&S projects from our
group are organized into four
categories:
 Mapping
techniques including
(1.1)
Decalmapping
for multivariate
visualization
(Rocha
et al.
17a,
from RT3: Vis&VA);
and (1.2)
Texture
synthesis
for illustrative patterns
(dos
Passos et al.
'10,
from RT1: NPR);
 Volume
modeling for interactive
volume data manipulation and
visualization, in
collaboration with medical
researchers and clinicians
(Chen
et al.
'08,
also RT2: SBIM,Patent:
US20090322748),
(Roberts et al.
['09,
'10],
Patent: US20110074780
A1);
 Geometric
and topological modeling
including four subcategories:
(3.1)
Pointset
sampling
for reconstruction and
rendering of implicit surfaces
 i.e., Proença
et al.
'07,
Vital
Brazil et al.
'10a
(from RT1: NPR);
(3.2)
Modeling
and rendering
byexample
for curve synthesis
(Brunn
et al.
'07,
from RT1: NPR
),
stippling marks
(Kim
et al.
'09
from RT1: NPR),
and terrain synthesis
(Brosz et al.
['06,
'07b]);
(3.3)
Procedural
modeling
for plant growth
(Streit
et al.
'05),
in collaboration with
botanical scientists; and
(3.4)
Multiresolution
filters
for image and mesh
reconstruction and
focus+context visualization
(Hasan
et al.
'15,
also RT3: Vis&VA);
 Computational
methods for bridging
geoscience and petroleum
engineering. One of the
major challenges in modeling
petroleum reservoirs is to be
able to integrate,
visualize,
and analyze
various data of different
types, which is susceptible to
high uncertainty. This process
is not a straightforward task
and usually requires
nontrivial and timeconsuming
workflows. Examples of such
data integrations can be
demonstrated in terms of
SAGD
modeling, well
test
calibration in conventional
reservoirs or numerical rate
transient modeling in
tight
and shale
reservoirs.
The primary objective of this
research is to be able to
create and adapt modern
workflows to efficiently
integrate dynamic and static
data. Modern geostatistical
simulations
in combination with flow
simulations are usually
required to ensure consistent
integrations. In addition, to
enhance the integration
workflow and reduce the
computational costs, efficient
optimization techniques are
also needed to bring the
computational efficiency to a
next level. Our research
include five
subcategories: (4.1)
Geostatistical
simulations
with conditioning hard and
soft data (Hamdi
& Costa Sousa
'16,
Khani et al.
['17,
'18 (a,
b)]);
(4.2)
Geological
welltest
simulation
(Hamdi et al.
['14,
'15a])
including gas condensate
system (Hamdi et al.
['13,
'15b]);
(4.3)
Efficient
optimization and uncertainty
quantification
for reservoir modeling
(Hajizadeh
et al.
'13,
Hamdi et al.
['15c,
'17(a,
b,
c),
'18]);
(4.4)
Global
sensitivity
analysis
of complex reservoir models
(Karami Moghadam et al.
['16,
'18],
with '18 from RT3:
Vis&VA);
and (4.5)
Flow
diagnostics
in unstructured grids
integrated with geomodeling
(Zhang et al. ['17
(a,
b),
'18]
, Rapid
Reservoir
Modeling
project).

M&S
Publications
G&PE
= Geoscience & Petroleum Engineering
venue
top

