Review of methods for solving inverse problems in identifying permanent magnets of executive elements in dynamic devices

Platov South-Russian State Polytechnic University (NPI), Department of Applied Mathematics, 346428 Novocherkassk, Russia

^* Corresponding author: ulapm20@mail.ru

Abstract

Permanent magnets due to their overall-saving and energysaving properties are increasingly used in the executive elements in dynamic systems of various fields of technology (electric motors, electromagnetic drives, electromechanical devices – relays, contactors, etc.). At the same time, devices with permanent magnets have a disadvantage – under the effect of overheating, mechanical shocks and current surges in the windings, their demagnetization occurs, which can lead to disruption of operation and failure of devices. In this regard, there is a need for identification of permanent magnets, that is, assessment of their condition by investigating the distribution of the magnetization by the volume of the permanent magnets by using the solution of inverse problems. The article provides the overview of methods for solving such problems. The analysis of publications has shown that two approaches to the solution of inverse problems can be distinguished: – the linearization of some functional and the multiple solution of the direct problem of calculating the magnetic field; – solution of the ill-posed problem and determination of the pseudo-solution stable to small perturbations by regularization methods. Both approaches have disadvantages. For example, the application of mesh-based methods in field calculations leads to the need to solve systems of equations of large dimension, parallel calculations are not used, etc. Therefore, the improvement of methods for solving inverse problems in the identification permanent magnets of executive elements in dynamic devices is an actual problem.