Physical Fields of Ship, Ocean and Atmosphere
DOI: https://doi.org/10.24866/2227-6858/2021-2-7

Shevkun S., Samoylov N.

SERGEY SHEVKUN, Candidate of Physics and Mathematical Sciences, Associate Professor, 
ResearcherID: AAF-6258-2021, ORCID: https://orcid.org/0000-0002-5312-1776, ScopusID: 6508158641, shevkun.sa@dvfu.ru
NIKITA SAMOYLOV, MS-Student (Corresponding Author), samoilov.ns@students.dvfu.ru
Polytechnic Institute, Far Eastern Federal University
Vladivostok, Russia

Application of the neural network approach for solution of direct and inverse problems of scattering

Abstract: In this paper, we consider one of the variants of the neural network approach for solving direct and inverse scattering problems – using a radial-basis neural network. The results obtained are compared with the results of other numerical methods. Solution of the Helmholtz equation is considered as well. Numerical experiments were carried out to restore the dependence of the square value of the wave number on the depth using a radial-basis neural network. The results of the study can be used to solve various applied problems related to hydroacoustic and underground communication systems, technical means for ocean exploration, as well as for navigation.

Keywords: direct scattering, inverse scattering, neural network, Helmholtz equation

Contribution of the authors: the authors contributed equally to this article.
The authors declare no conflicts of interests.

See the reference in English at the end of the article