Plenary Speakers
Darkhan Akhmed-Zaki
Head of the Computer Sciences Laboratory
Scientific Research Institute of Mechanics and Mathematics, Al-Farabi Kazakh National University
Almaty, Kazakhstan
Information system for oil recovery analysis
Now are known a lot of studies and computer programs which investigated oil recovery processes (UTCHEM, ECLIPSE, GPAS, STARS). But it are realized computer modeling based on knowledge sets (features, behavior, state and etc.) of well-known oil-recovery technology in oil industry. Our article addresses about modernization of the information system for oil recovery analysis (ISAR-2) and its deployment carried out in between 2005-2015. ISAR-2 includes following software modules: distributed databases that store geo-physical and physic-hydro dynamical properties of the fluids/stratum; databases of mathematical and engineering models for two and three phases fluids flow in anisotropic and nonuniform porous media; interactive web visualization of computing results in 2D-3D via Internet connection; high-performance computing models for supercomputer and mobile platform execution. ISAR-2 functionality can easily be extended by adding new models of technological oil recovery, and has quick access to the hydrodynamics simulator capabilities using built-in authorization.
Alexander V. Avdeev
Intel Russia/CIS Business Development Manager, PhD
Software and Services Group
Novosibirsk, Russia
Intel software for solving research and industrial problems: Modern trends of high performance computing
Hardware technologies in High Performance Computing are continuously undergoing major changes and rapidly increasing performance capabilities, but the software and the underlying code legacy is often left unchanged or even neglected. This leads to performance gaps and underutilized hardware assets. The talk is devoted to Intel programming techniques and software tools required to achieve the highest performance on systems built on the current and future generations of Intel® processors, including coprocessors Intel® Xeon Phi ™. We will consider the latest Intel software tools - Intel® Parallel Studio XE 2016 (which include optimized compilers, math libraries and tools for optimization and parallelization of programs) and examples (success stories) of Intel software usage for solving industrial and research problems. We will also discuss Intel University/Academic programs.
Igor V. Bychkov
Director of ISDCT, Academician of RAS
Matrosov Institute for System Dynamics and Control Theory (ISDCT) of SB RAS
Irkutsk, Russia
Technologies of heterogeneous programming systems integration in the informational computing environment of mathematical modeling and data analysis
The solution of complicated task usually requires usage of distributed programming systems realized on various platforms. The informational computing environment for mathematical modeling and data analysis was developed in the ISDCT of SB RAS. It uses WPS standard for service publication and provides the set of technologies that ease the solution of complex tasks by making the development and deployment of distributed services easier. Also, the JavaScript extension was developed in order to build services interaction.
Peyman Givi
Director, Laboratory for Computational Transport Phenomena
Professor, Mechanical & Petroleum Engineering, University of Pittsburgh
Pittsburgh, Pennsylvania, USA
Quantum Computing and Its Potential for Turbulence Simulations
A tutorial will be provided of quantum computing (QC) and the way it has made significant speed-up in various simulations. A review will also be provided of the large eddy simulation of turbulent flows via the stochastic filtered density function (FDF) methodology. The potentials of the quantum speed-up in FDF simulation via QC appear to be significant. This can results to a revolutionary means by which turbulence simulations can be conducted in future.
Andreas Griewank
Professor of Mathematics at the Humboldt University, Germany,
and Yachay Tech University, Ecuador
Berlin, Germany
Nonsmooth Numercis via Piecewise Linearization
Classical numerical methods exploit smoothness of nonlinear problem functions to approximate them locally by linear or higher order models. Many application areas of current interest involve functions that are nonsmooth, i.e. have kinks or jumps, for example due to phase transitions.
We demonstrate that certain classes of nonsmooth functions can be well approximated by piecewise linear functions, and how these models can be utilized for the fundamental numerical tasks of equation solving, optimization, and the integration of differential equations.
Simon Jayaraj
Professor
National Institute of Technology
Calicut, Inidia
Modeling and Simulation of Fluid Flow and Mixing in Micro Channels Using Immersed Boundary Method
Important recent literature available in the area of micro channel flow modeling and simulation are reviewed. The topics covered include the physics of flows in micro channels and integrated modeling of micro channel flows. An immersed boundary method (IBM) based on discrete forcing approach is elaborated for numerical simulations of micro channel flow and mixing. Standard backward facing step is considered to ensure the accuracy and feasibility of the method. Further the micro-channel is constricted with the help of triangular blocks or rectangular blocks with the intention of enhancement in mixing efficiency. Both electro-osmotic effects and channel constriction simultaneously incorporate the active and passive mixing techniques in th
Christophe Josserand
Research director at CNRS, PhD and Habilitation
Institut D'Alembert, CNRS & UPMC
Paris, France
Numerical simulation of multiphase flows
In this talk, I will present powerful numerical methods allowing the precise resolution of multiphase flow problems. I will in particular present the iconical problem of drop impact on liquid and solid surfaces, which is a key-problem of surface flows as well as a problem involving multiple applications. I will in particular focus on the interplay between the surface deformation and the gas dynamics.
Sergey Kabanikhin
Corresponding Member of RAS, Director-elect
Institute of Computational Mathematics and Mathematical Geophysics SB RAS
Novosibirsk, Russia
The size of the domain of measurements is the regularization parameter in continuation problem
Gayaz S. Khakimzyanov
Professor, Research Scientist
Institute of Computational Technologies SB RAS
Novosibirsk, Russia
On Some Problems of Numerical Modelling of Surface Waves in the Framework of the Shallow Water Model
Anvarbek Meirmanov
Professor
Kazakh-British Technical University
Almaty, Kazakhstan
Macroscopic Mathematical Models of Physical Processes in Porous Media via Microstructure
Mathematical models of physical processes play important rule in oil, gas and mining industries. There are different types of such models, but all of these macroscopic models do not take into account the peculiarities of processes at the microscopic (pore) level. The particular qualities of the process just postulates at the macroscopic level. That is why all these models involve a lot of phenomenological functions and constants, which obviously cannot be determined uniquely.
Our approach is based upon on the commonly accepted classical continuum mechanics at the pore level and rigorous homogenization technique.
Oleg I. Potaturkin
Professor, Deputy Director on Science
Institute of Automation and Electrometry of SB RAS
Novosibirsk, Russia
Spectral-Spatial Classification of the Earth Surface Types Using Hyperspectral Remote Sensing
Potential and restrictions of hyperspectral satellite images application at the Earth surface monitoring are discussed. Spectral and spectral-spatial methods of vegetative cover types classification based on reflection spectra comparison of researched and reference objects belonging to various classes were investigated. The estimation of methods efficiency is calculated on the basis of remote sensing data processing and ground-truth observations comparison. It is shown that the examination of spatial dependence between pixels is useful for classification.
Boris Ya. Ryabko
Professor, Head of Laboratory
Institute of Computational Technologies SB RAS
Novosibirsk, Russia
Optimization of the Internet Search Based on Laws of Information Theory and Psychology
Nowadays Internet plays important role in all spheres of human life. The problem of search of the necessary information in Internet and time spent for search becomes especially actual. Electronic systems of search include booking (theatrical, aviation, railway, etc.), literature search, search of goods on the Internet and e -learning . That is why the problem of minimization time of search becomes especially actual.
Results of our research allow to reduce search time several times, using methods of Information Theory and laws of psychology of perception.
Nickolay Y. Shaparev
Professor, Chief Scientist
Institute of Computational Modelling SB RAS
Krasnoyarsk, Russia
Modeling of absorption and transfer of radiation in an expanding sphere
The absorption and transfer radiation in a self-similarly expanding gaseous sphere illuminates by laser, resonance and continuum radiation has been numerically studied. The absorption coefficient, spectral and spatial characteristics of the absorbed and scattered radiations, and the influence of optical thickness and the ra-dial expansion velocity on them has been investigated. It is shown that expansion results, in a reduced optical thickness of the medium.
Yury N. Zakharov
Professor, Head of Departament
Kemerovo State University
Kemerovo, Russia
Computational and Experimental Research of Soil Erosion at the Bottom of Gravity Platforms
The presentation shows computational simulation of flow and surface waves affecting soil erosion at the bottom of “Prirazlomnaya” sea platform model and comparison between solutions obtained and results of laboratory experiments carried out in hydrodynamic test basin of XXIII State Marine Research and Design Institute (Saint-Petersburg, Russia).
Fluid model described by three-dimensional system of time-dependant Navier-Stokes equations is used for numerical calculation of flows near the platform.