Search results

The purpose of this paper is to develop a viscous-type frequency dependent scalar Preisach hysteresis model and to identify the model using measured data and nonlinear numerical field analysis. The hysteresis model must be fast and well applicable in electromagnetic field simulations.

Iron parts of electrical machines are made of non-oriented isotropic ferromagnetic materials. The finite element method (FEM) is usually applied in the numerical field analysis and design of this equipment. The scalar Preisach hysteresis model has been implemented for the simulation of static and dynamic magnetic effects inside the ferromagnetic parts of different electrical equipment.

The comparison between measured and simulated data using a toroidal core shows a good agreement. A modified nonlinear version of TEAM Problem No. 30.a is also shown to test the hysteresis model in the FEM procedure.

The dynamic model is an extension of the static one; an extra magnetic field intensity term is added to the output of the static inverse model. This is a viscosity-type dynamic model. The fixed-point method with stable scheme has been realized to take frequency dependent anomalous losses into account in FEM. This scheme can be used efficiently in the frame of any potential formulations of Maxwell's equations.

The classical Preisach model and its modifications are one of the most generally applied simulations to model the behavior of magnetic materials, to describe hysteresis phenomena and different properties, as noncongruent minor loops, frequency dependence, temperature dependence, accommodation, and so on. Artificial neural networks (NNs) are widely used in fields of research where the solution of problems with conventional methods on traditional computers is very difficult to work out, for example system identification, modeling and function approximation. NNs can be considered as universal approximation for functions based on the theorem of Kolmogorov‐Arnold. In this paper a new NN model of scalar hysteresis characteristics is introduced. The examined method is built on the function approximation and continuous interpolation capability of NNs. The anhysteretic magnetization curve and a set of the ascending and a set of the descending first order reversal branches can be stored in a system of three neural networks. Different properties of magnetic materials can be simulated by a simple knowledge‐based algorithm. Value of differential susceptibility can be expressed in analytical form. Finally hysteresis characteristics predicted by the introduced model are compared with the results of the Preisach simulation.

The purpose of this paper is to present a Preisach model to simulate the vector hysteresis properties of ferromagnetic materials.

The vector behavior has been studied at low frequency applying a single‐sheet tester with a round‐shaped specimen, and the locus of the magnetic flux density vector has been controlled by a digital measurement system. An inverse vector Preisach hysteresis model has been developed and identified by using the measured data.

Finally, the inverse model has been inserted into a finite element procedure through the combination of the fixed point technique and the reduced magnetic scalar potential formulation. The developed single‐sheet tester measurement system has been simulated. The applicability of the realized measurement system as well as the developed model has been proven by comparing measured and simulated results.

The identification technique is original, based on a previous work of the author.

The purpose of this paper is to present the importance of model accuracy in closed loop control by the help of parallel finite element model of a voltage-fed solenoid with iron core.

The axisymmetric formulation of the domain decomposition-based circuit-coupled finite element method (FEM) is embedded in a closed loop control system. The control parameters for the proportional-integral (PI) controller were estimated using the step response of the analytical, static and dynamic model of the solenoid. The controller measures the error of the output of the model after each time step and controls the applied voltage to reach the steady state as fast as possible.

The results of the closed loop system simulation show why the model accuracy is important in the stage of the controller design. The FEM offers higher accuracy that the analytic model attained with magnetic circuit theory, because the inductance and resistance variation already take into account in the numerical calculation. Furthermore, parallel FEM incorporating domain decomposition to reduce the increased computation time.

A closed loop control with PI controllers is applied for a voltage driven finite element model. The high computation time of the numerical model in the control loop is decreased by the finite element tearing and interconnecting method with direct and iterative solver.

On the basis of the Kolmogorov‐Arnold theory, the feedforward type artificial neural networks (NNs) are able to approximate any kind of nonlinear, continuous functions represented by its discrete set of measurements. A NN‐based scalar hysteresis model has been constructed preliminarily on the function approximation ability of NNs. An if‐then type knowledge‐base represents the properties of the hysteresis characteristics. Vectorial generalization to describe isotropic and anisotropic magnetic materials in two and three dimensions with an original identification method has been introduced in this paper.