Methods and algorithms of computational mathematics and their applications

A structural two-level neural network approach to joint inversion of gravitational and magnetic fields

Yanfei Wang^a, D. V. Churbanov^b, R. L. Argun^c, A. V. Gorbachev^a, A. S. Leonov^d, D. V. Lukyanenko<sup>a

a</sup> Institute of Geology and Geophysics, Chinese Academy of Sciences
^b Moscow Institute of Physics and Technology
^c Lomonosov Moscow State University, Faculty of Physics
^d National Research Nuclear University MEPhI

Abstract: The paper proposes a new two-level neural network approach to joint inversion of gravitational and magnetic fields in the problem of mineral exploration. The lower level uses two neural networks with equivalent architecture. The first network calculates the distribution of gravitational field sources in a given domain based on measurements of this field on a certain remote surface. The second one finds the distribution of magnetic field sources in the same area based on magnetic field measurements. At the upper level, a structural residual between the two obtained source distributions is calculated, which determines the “degree of difference” of their geometries. The resulting neural network is trained to minimize this structural discrepancy. As a result, this two-level system implements a computational algorithm that finds geometrically close sources of fields of different nature. Test simulations demonstrate the high quality of joint inversion of model gravitational and magnetic fields within the proposed aproach.

Keywords: inverse problem, joint processing of gravitational and magnetic fields, neural networks, structural residual.

UDC: 519.642

Received: 30.06.2025
Accepted: 17.08.2025

DOI: 10.26089/NumMet.v26r322