mathematical model; moisture transfer process; fractional derivatives; method of finite differences; physics-informed neural networks

NEURAL NETWORK MODELING OF THE PROCESS MOISTURE TRANSFER ON THE BASIS OF DERIVATIVES FRACTIONAL ORDER

In this paper, a mathematical model of moisture transfer in capillary-porous media in one- and two-dimensional space is presented and investigated, for its description the apparatus of fractional integrodifferentiation was used. This approach made it possible to take into account such properties of a system with a fractal structure as memory, self-organization, temporal and spatial correlations. The complexity of this mathematical model complicates its application and requires significant computing power.