No. 1, 2005


Jarmila Dedkova, Libor Dedek : Methods for computation of general signals induced in high-resistive place chamber, pp.1-21

Different approaches of computation of signals induced in the Resistive Plate Chamber (RPC) are presented. A simple one-dimensional model of the RPC is analyzed, in order to enlighten on the processes which occur in the numerical modeling of the three-dimensional field in the chamber. Further a generalized solution for numerical evaluation of output signals from the RPC is proposed. Simplified solution of the output signals due to the moving charge between two parallel electrodes is presented. Finally the field equations for scalar transient electric potential due to the motion of a free charge in the presence of high resistive dielectrics are derived. Also the equations for the currents and voltages of the pick-up electrodes indicating the charge velocity and position are given. Some results of the calculated signals are obtained using a new program based on the Finite Element Method.

You Xiaojie, Geng Wang, Viktor Valouch : Voltage balance of separate DC capacitors for CSAPF, pp.23-33

The application of cascade shunt active filter (CSAPF) based on H-bridge multilevel converter topology is hindered by the problem of voltage unbalance of separate DC capacitors that leads to system instability. This paper presents a new control algorithm of CSAPF for balancing voltages of separate DC capacitors. The experimental results demonstrate the effectiveness of the proposed approaches.

Ivo Dolezel, Pavel Solin, Martin Zitka, Bohus Ulrych : On electromagnetic stirring of molten metals, pp.35-51

The stirring of molten metals by means of electromagnetic fields is a successful novel technology that has the potential to increase the quality of metal and alloy products by degasification of the melts, improving their homogeneity and preventing the formation of dendritic crystals during solidification. This paper presents a numerical model of this process based on the incompressible Navier-Stokes equations endowed by appropriate free-surface boundary conditions. The equations are discretized by the second-order semi-implicit Oseen-Crank-Nicolson scheme on fully-staggered finite difference grids. The electromagnetic (Lorentz) forcing term is determined from the Maxwell equations. Various properties of the model and numerical scheme are discussed, and numerical examples are presented.

Petr Filip, Jiri David, Radek Pivokonsky : On empirical modelling of viscosity regarding ageing of emulsions, pp.53-66

The aim of this contribution is to present an empirical model taking into account ageing of highly concentrated emulsions. This model enables smooth passage from modelling a viscoelastic fluid (shear-thinning behaviour) corresponding to a `fresh' material to describing a viscoplastic fluid (viscoplastic behaviour) corresponding to an `aged' material including yield stress modelling.

Jaroslav Feda : Double mechanism of deformation of granular materials, pp.67-81

Creep behaviour in the oedometer of the granulated clay was studied. Two deformation mechanisms have been found, diffusion of deformation and garlandlike compression. They develop after a critical time period of about 1 day (transition from the inborn to the compression structure). Secondary consolidation occupies shorter time intervals (standard behaviour). The author proposes the diffusion deformation to be the fundamental one, however, only in a homogeneous sample. Otherwise a hybrid deformation takes place.

Pavol Fedor, Daniela Perdukova : Fuzzy model of washing machine drive, pp.83-91

The article presents a simplified washing machine model. First, a model is created by analysis of force relations in a washing machine; then a fuzzy model based on selected input and output samples is generated. Finally, a comparison of these two models is presented.

Jozef Vojtko, Irena Kovacova, Dobroslav Kovac, Ladislav Madarasz : Neural networks for error reduction of transformer type elastomagnetic sensors, pp.93-106

This paper deals with possibilities of error correction of elastomagnetic sensors, which are used for measuring massive pressure force within the range 0-200 kN. For this class of sensors it is very hard to obtain the correct result of measured force from sensor output. Therefore, designed solution is based on utilizing neural networks as a universal approximator which provides a conversion from sensor output into measured force and suppresses errors.