Theory and Simulation of Electromagnetic Systems Governed by Machine Learning Principles

Abstract

This paper proposes the idea that electromagnetic systems can be formulated through probabilities once the system has been understood by the classical physics. With this, several physical observables are estimated. Also, with the diffusion equation, electrical circuits can be constructed. Thus, it is seen that classical electrodynamics dictated by action to distance forces, can be encompassed to the so-called Mitchell’s criteria. It has as consequence the validation of common problems in electromagnetism based at space-time probabilities more than the well-known methodologies based at vector algebra. Simulations of electric power contour plots are presented.