Modeling of heat transfer processes on hybrid architecture computers

The paper is devoted to the development of methods and algorithms for computer simulation of heat transfer processes, which play a significant role in the field of continuum mechanics. Finding the temperature fields during welding to determine the kinetics of temperature stresses and deformations helps to increase the technological and operational indicators of welded structures. This paper proposes a method of determining temperature fields during welding of thin-sheet elements of structures. This paper proposes a method of determining temperature fields during welding of thin-sheet elements of structures. A non-stationary thermal problem was obtained as a result of the simulation of the heat transfer process, taking into account the nonuniformity of the heating of the welded elements during the welding process. A discrete problem was obtained for the numerical solution of this problem by the finite element method – a first-order system of ordinary differential equations, that is, the Cauchy problem. This paper proposes a high-performance fourthorder Runge-Kutta computer algorithm for solving the Cauchy problem. The algorithm is developed for modern supercomputers built on new multi-core processors. The proposed algorithm is based on a multilevel model of parallel computing. At the upper level of parallelism, calculations performs in form of parallel processes, memory is distributed between them, also ensures synchronization of calculations and information exchanges. Parallelization at this level is expedient to be carried out by means of MPI. At the second level of parallelism, calculations performs using OpenMP directives or Intel MKL library software modules using shared memory. The third level of parallelism involves data processing with vector arithmetic and logic devices, which are included in the program automatically with the help of the compiler. The developed algorithm was tested on solving the problem of determining the temperature field, which arises during the study of the life cycle of welded structures. Such calculations are used, in particular, in the problem of determining the kinetics of temperature deformations and stresses during welding of thin plates. Some results of testing the proposed method and the developed algorithm on the SKIT supercomputer of the V.M. Glushkov Institute of Cybernetics of the National Academy of Sciences of Ukraine are presented.

Authors

Institutions

Publication Details

Journal
PROBLEMS IN PROGRAMMING
Published
2024-09-01
DOI
https://doi.org/10.15407/pp2024.02-03.078
Primary Topic
Heat Transfer and Optimization
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Modeling of heat transfer processes on hybrid architecture computers

А. В. Попов, Oleksandr Duchenko, A.N. Nesterenko
PROBLEMS IN PROGRAMMING
Heat Transfer and Optimization
article

Modeling of heat transfer processes on hybrid architecture computers

А. В. Попов, Oleksandr Duchenko, A.N. Nesterenko
article en

Abstract

The paper is devoted to the development of methods and algorithms for computer simulation of heat transfer processes, which play a significant role in the field of continuum mechanics. Finding the temperature fields during welding to determine the kinetics of temperature stresses and deformations helps to increase the technological and operational indicators of welded structures. This paper proposes a method of determining temperature fields during welding of thin-sheet elements of structures. This paper proposes a method of determining temperature fields during welding of thin-sheet elements of structures. A non-stationary thermal problem was obtained as a result of the simulation of the heat transfer process, taking into account the nonuniformity of the heating of the welded elements during the welding process. A discrete problem was obtained for the numerical solution of this problem by the finite element method – a first-order system of ordinary differential equations, that is, the Cauchy problem. This paper proposes a high-performance fourthorder Runge-Kutta computer algorithm for solving the Cauchy problem. The algorithm is developed for modern supercomputers built on new multi-core processors. The proposed algorithm is based on a multilevel model of parallel computing. At the upper level of parallelism, calculations performs in form of parallel processes, memory is distributed between them, also ensures synchronization of calculations and information exchanges. Parallelization at this level is expedient to be carried out by means of MPI. At the second level of parallelism, calculations performs using OpenMP directives or Intel MKL library software modules using shared memory. The third level of parallelism involves data processing with vector arithmetic and logic devices, which are included in the program automatically with the help of the compiler. The developed algorithm was tested on solving the problem of determining the temperature field, which arises during the study of the life cycle of welded structures. Such calculations are used, in particular, in the problem of determining the kinetics of temperature deformations and stresses during welding of thin plates. Some results of testing the proposed method and the developed algorithm on the SKIT supercomputer of the V.M. Glushkov Institute of Cybernetics of the National Academy of Sciences of Ukraine are presented.

PROBLEMS IN PROGRAMMING(2-3)
V.M. Glushkov Institute of Cybernetics (UA)
Openalex Percentile: Top 73%
Heat Transfer and Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.