Application of submodeling technique to reduce time spent modeling remote magnetic field sensors

Abstract

Currently, various magnetic field sensors located at a distance from the signal source are widely used in technology. The development and optimization of the designs of such sensors is often carried out using computer modeling, which requires not only large computing resources, but also significant time costs. Thus, the development and application of special methods to reduce the time spent on modeling these converters are relevant. The solution of differential equations in the magnetic field model of the converter is performed using the finite element method implemented in the COMSOL Multiphysics program. The solution of the chain model is obtained based on the general provisions of the theory of circuits using the built-in electric circuit editor in the COMSOL Multiphysics program. The use of a submodeling technique is proposed to solve transient processes in magnetic field sensors with a magnetic circuit and located at a great distance from the signal source. It is proposed to use the solution of a two-dimensional model as initial conditions. A technique to determine the size of a submodel is presented by comparing the results of two two-dimensional models, one of which does not consider the magnetic circuit. The factors influencing the effectiveness of the proposed technique are determined. The developed submodel tuning technique reduces the simulation time of magnetic field sensors with an error of replacing the complete model with a submodel of no more than 1 %. Application of the proposed submodeling technique will improve the efficiency of the design of magnetic field sensors.