Published in Sarov (Arzamas-16), Nizhegorodskaya oblast
NUCLEAR CENTER -
ALL-RUSSIAN RESEARCH INSTITUTE
OF EXPERIMENTAL PHYSICS
Русский | English
Issue No 1, 2022
SIMULATION OF LASER RADIATION TRANSPORT AND ABSORPTION IN INDIRECT DRIVE ICF TARGETS
|I. S. Chubareshko,
|A. A. Shestakov
VANT. Ser.: Mat. Mod. Fiz. Proc. 2022. No 1. P. 3-16.
The numerical simulation of the inertial confinement fusion (ICF) problems allows optimizing multiple experiments and reducing their number owing to the comparison of results obtained for different design configurations of laser targets. For ICF problems, 2D and 3D models of the laser radiation transport and absorption were developed in the geometrical optics approximation. The paper describes a 3D laser radiation transport and absorption model. The main objective was to verify and test the proposed model and demonstrate results obtained using this model, which are in a good agreement with the US Lawrence Livermore National Laboratory results.
Key words: laser radiation transport and absorption, indirect drive targets.
|ESTIMATION OF TIME CONSTANT IN ACCOUNTING FOR DELAED NEUTRONS
D. G. Modestov
VANT. Ser.: Mat. Mod. Fiz. Proc. 2022. No 1. P. 17-26.
The difference between the time constant calculation techniques for the neutron transport problem and similar techniques for arbitrary particles is small. The specific feature is in accounting for delayed neutrons, which is usually performed with the approximation, in which the transport of predecessor nuclei is considered in its simplified form. For this case, the paper presents the transport process simulation technique with such approximation. Asymptotic dependences of the bias error were obtained for it. It is shown, in particular, that for subcritical systems the behavior of such error is the same as in the method of generations used to solve a conditionally critical problem, while for supercritical systems it makes no contribution. To demonstrate the technique capabilities, the paper presents results of the time constant calculation for two systems: an all-metal critical assembly and a VVER-1000 nuclear reactor. Both subcritical and supercritical states were considered. Calculations were carried out with and without consideration of delayed neutrons. Results of these calculations confirm the asymptotic behavior of error and correctness of the proposed technique, if results are compared with those for an unsteady problem.
Key words: the neutron transport equation, neutron kinetics, time constant, critical assembly, VVER.
|URS-OF PROGRAM LIBRARY FOR CALCULATION OF MATERIAL PROPERTIES ADAPTED TO COMPUTING SYSTEMS CAPABLE OF PARALLEL AND VECTORIZED CALCULATIONS
D. G. Gordeev, N. N. Zhilnikova, D. N. Kidyamkina, V. G. Kudelkin, M. V. Kulikova, O. N. Shumilina
VANT. Ser.: Mat. Mod. Fiz. Proc. 2022. No 1. P. 27-39.
The paper presents the description of data structures, arrangement of codes and calculations in the vectorized version of URS-OF library that allows carrying out parallel shared- memory computations of thermophysical functions in OpenMP standard. The protocol for coordination between URS-OF library and an application code and paralleling techniques are described. Examples are given that illustrate the specifics of paralleling conditional structures and iterative loops with the use of mask arrays, as well as the paralleling of the SuperURS-type codes. Results of the performance analysis are represented as the acceleration metric due to optimization; results of the scalability study performed with the problem partition method are represented by the paralleling efficiency metric.
Key words: URS-OF program library, shared memory parallelization, OpenMP standard, autovectorization, vector register width, conditional structures, iterative loops, acceleration due to optimization, efficiency of paralleling.
|A 3D TEMPERATURE FIELD RECALCULATION METHOD FOR STRENGTH ANALYSIS PROBLEMS
V. I. Romanov, E. E. Maslov, C. Yu. Gulakov
VANT. Ser.: Mat. Mod. Fiz. Proc.. 2022. No 1. P. 40-47.
During the strength analysis there is a common need in considering non-uniform temperature fields, where the deformation process runs. The goal of this paper is to describe a method for transmitting data on the space distribution of temperatures to the computational strength model of a structure. The procedure of interpolating temperatures to nodes of a finite-element grid was implemented in Python programming language to solve strength problems using known temperature values at points of an ordered set, which are, for example, nodes, or centers of grid cells for the temperature problem solution. The procedure is based on solving the problem of finding several nearest neighbors. For the effective process, data on interpolation points-nodes is represented by an octotree. An unknown temperature value at an arbitrary point of space is restored using the already known values at neighboring points with the spatial linear interpolation based on the least square method.
Results of testing the method for a model problem are given. A high accuracy of interpolation is demonstrated. The method described in the paper is successfully used by the authors to take into account the thermal state of shipping package sets for the nuclear fuel transportation and storage, when justifying their safety. Directions for the further development of the method are lined out.
Key words: a set of points, the problem of finding a nearest neighbor, octotree interpolation, the least square method.
|MPI+OpenMP IMPLEMENTATION OF CONJUGATE GRADIENT METHOD WITH THE SECOND AND FIRST ORDER INCOMPLETE BLOCK INVERSE TRIANGULAR FACTORIZATION PRECONDITIONERS
O. Yu. Milyukova
VANT. Ser.: Mat. Mod. Fiz. Proc. 2022. No 1. P. 48-61.
The paper considers a new preconditioner for solving linear algebraic equation systems with a symmetric positive definite sparse matrix, i.e. a preconditioner for incomplete block inverse triangular factorization of the first order of value. A way of using the (MPI+OpenMP)-process to construct and invert preconditioners for incomplete block inverse triangular decomposition of the second and first order of value is proposed. With the use of the (MPI+OpenMP)-process the number of blocks in such preconditioners is a multiple of the number of processors in use and the number of threads. Comparison is made between the problem runtimes with the use of the MPI process and with the use of the hybrid (MPI+OpenMP)-process for a model problem and for a number of problems from SuiteSparse collection of sparse matrices. The runtime comparison for these problems is performed using the method of conjugate gradients with the second and first order incomplete block inverse triangular factorization preconditioning.
Key words: sparse matrices, implicit block preconditioning, incomplete Cholesky triangular factorization, parallel preconditioning, method of conjugate gradients.
|THE EFFECT OF THE SUBSTRATE LAYER MATERIAL OF COOLED INFRARED FPA PHOTODETECTORS ON HEATING OF PHOTOSENSITIVE LAYER DUE TO INTERACTION WITH INTENSE LASER RADIATION
A. E. Zaponov, D. S. Konradi
VANT. Ser.: Mat. Mod. Fiz. Proc. 2022. No 1. P. 62-69.
Computational models of the intense laser radiation interaction with photodetector arrays based on InSb and CdxHg1-xTe with substrate layers made of Si, GaAs, Ge, and InAs to determine their effect on the generation and evolution of emerging thermal fields have been developed and implemented in the "Logos" engineering analysis and computer simulation software package. Temperature dependences were examined and represented in their analytical form for optical and thermophysical parameters of materials under consideration. Simulation results, which have been analyzed and generalized, are presented.
Key words: photodetector array, laser radiation, computer simulation, substrate layer material, "Logos" software package.
|A SPACE-GROUP TRANSFORMATION SYSTEM FOR INFORMATION FLOWS
I. A. Martynova, A. P. Martynov, D. B. Nikolaev, V. N. Fomchenko
VANT. Ser.: Mat. Mod. Fiz. Proc. 2022. No 1. P. 70-82.
The development of complex computing systems for classes of problems with parallel computations predetermined the need in new approaches to control data processing and this is the subject of the present paper. Operations with multidimensional dynamic structures raise requirements concerning the validity of such operations, speedup, and transparency of control over computations. The functional-algorithmic model of the space group motion proposed by the paper authors is intended for tracing the data processing and data verification using the given parameters of input and output information arrays. The model imposes no limitations on the processed data and is the basis for the multi-purpose tools used to describe and analyze interaction processes in complex dynamic systems using parallel processing of data. Results of analyzing single-thread information cryptographic systems are presented as an example of using the model and corresponding space-group transformation system for information flows.
Key words: information system, multidimensional sets, data transpormation, universal control system, cryptograpfic functions.
|[ Back ]