Levitation induction heating represents the first step of levitation melting, which is an advanced technology used for processing of superclean metals or metal alloys. A workpiece is lifted by the Lorentz forces in time variable magnetic field produced by a system of suitably placed field coils and then heated by the Joule losses. The paper deals with optimization of the heating process from the viewpoint of the velocity of heating the workpiece. Investigated is the influence of its shape (sphere, cylinder, truncated cone) and arrangement of the field coil. The theoretical analysis is supplemented by a series of particular examples and discussion of the results obtained. 

The paper presents contribution to parameter identification of an induction motor by genetic algorithms (GA) programme. Its correctness was verified by identification of the model assembled by five rated parameters. Then phase current and speed of the 1.5 kW motor were measured during start by direct connection to mains. Unknown parameters of the mathematical model were identified by GA. Because of varying motor parameters some differences appear among model time responses and real curves. However, certain constant parameters distinct from rated values but ensuring satisfactory matching with measurement may be found. Deviations of time responses were evaluated at measurement of speed, current or both magnitudes. Particular solutions are illustrated by corresponding graphs. 

The aim of this paper is to propose the use of temporal logic in dynamic weaving of advices in aspect oriented programming (AOP), based on process functional paradigm. Namely, temporal logic appears to be well suited tool to design pointcuts whose task is selection of join points throughout source code (static weaving) or program execution (dynamic weaving). Because it is well known that join point model and techniques used for selecting them (pointcuts) are crucial elements of aspect oriented programming, we tried to propose the way of better (at least we think it is better for describing dynamic join points) pointcut formulation in terms of temporal logic. By now we have developed process functional language (PFL), which is simple but yet powerful abstraction used to illustrate mechanisms of programming languages. PFL includes structures and principles supporting functional, imperative and even object oriented and aspect oriented paradigms (which is still under development), so it is multi paradigmatic language. This paper is presenting reflection as a basic property of PFL used as a tool for weaving semantics of aspect code into original code. Code examples in this paper use PFL, but some examples are written also in AspectJ to clarify informal semantics of its operations (particularly cflow operation, since it is used for selecting dynamic join points).

This paper introduces a method for restoration of missing arbitrarily shaped image segments using combined discrete orthogonal transforms (DOT). The missing segment texture is successively interpolated/extrapolated using the texture of the correctly decoded segments belonging to the same object, thus having the same texture pattern.

Two approaches to calculating distribution energy losses are developed in this paper. The first one is based on the fuzzy load flow, and the other one uses fuzzy clustering technique. The attention at the first approach is devoted to forming the fuzzy numbers that represent loads. Data accessible from the measurements in corresponding substations are considered in this process. Using formed fuzzy loads one fuzzy load flow calculation is made. Results of calculation are fuzzy power losses. Defuzzification gives the deterministic value of average power losses that multiplying with the number of hours for analyzed period gives energy losses. For the second approach, the range of the coefficient that defines fuzziness of clustering is determined, as well as an optimal number of clusters. Analyses shows that the best results are obtained for fuzziness coefficient on the range of 1.1-2, and the number of clusters up to 20. 

The paper deals with algorithms of entropy coding in the standard videocodecs H.261,263,264 and MPEG-1,2,4 considerably increasing their total data compression. These algorithms are variable length coding by using Huffman codes, arithmetic coding, universal variable length coding, contex based adaptive binary arithmetic coding employing context modeling and binarization of nonbinary symbols. In generally they employ the statistical nonstationary properities of videosignals, which depend on the content of visual scenes. 
The tables of Huffman code used in all standard videocodecs are fixed and optimized for the bounded range of bit rates. To encrease the range in the videocodec H.26L there is used a new algorithm of entropy coding based on universal variable length code. Disadvantage of both Huffman and universal variable length codes is that each symbol is encoded by integer number of bits even if with different lenth. Therefore the videocodec H.263 alows using arithmetic coding that brings saving of bit rates about 3-4% in interframe and up to 10% in intraframe mode of the videocodec. The bigest encreasing of the performance of entropy coding can be achieved by contex based adaptive binary arithmetic coding that removes of an intersymbol redundancy. By its applying in the videocodec H.264 the saving of bit rates may be near 32% compared to the universal variable length code at the same quality of coded videosignals.
All algorithms of entropy coding are applied on entropy coding not just of the main information about transform coefficients in intraframe and interframe modes of the standard videocodecs but even of additional information.

The place of generation of technologic heat is important applied criteria for classification of electric heating. According to that, the direct heating is if the heat is generated directly in the heated material, or the indirect heating if material is heated indirectly by the heat generated in conversion element. Possibilities of practical application of both methods are typical for resistant heating. Considering indirect heating, the charge is out of influence of electromagnetic field, this directly does not influence its temperature field. Temperature distribution in the charge volume depends only on its heat-physical properties and conditions of heat exchange on surface, i.e. thermo kinetic processes. Fourier's equation is applied in field solution. Considering direct heating, temperature distribution in the charge volume is more complicated. It is conditioned by interaction of temperature and electromagnetic field, which is generated by the current flowing in the charge. Solution of temperature field in directly heated charge must for that reason respect current density distribution on its cross-section, this leads to applying of heat conduction equation. We introduce its application for different types of temperature fields in the solid charge in this article. 

Because of Sulfur Hexafluoride's excellent physical properties and unparalleled dielectric capabilities it has become the single choice for the electrical utility industry worldwide. This inert, non-toxic gas has excellent cooling, insulating, and arc quenching capabilities in potentially high temperature applications such as electrical breakers. 
Sulphur Hexafluoride is wide used in high voltage apparatus because of its excellent dielectric and heat-transfer properties. It is known that the insulation performance of SF6 is limited, not by its uniform dielectric strength, but by the effects of local field enhancement leading to the formation of corona discharge. For practical requirement, the research is aimed at an SF6-bassed gas mixture under the non-uniform field conditions associated usually with the presence of corona formation. It is clear then that the understanding of the mechanism of corona discharge formation in SF6/buffer gas mixture is fundamental for design of improved gaseous dielectrics. We presented the fundamental characteristics and properties of SF6 mixtures with N2 and CO2.
The present paper describes an experimental investigation into the nature of physical processes leading to negative corona current pulse in SF6 mixture with CO2 buffer gas. Current waveforms of initial stages of negative corona discharge formation in a short negative point-to-plane gap have been measured with nanosecond time resolution in the mixtures CO2+SF6 at the pressures 50 kPa and for various contents of SF6 and overvoltages. From the conclusion of this article is evident, that current pulses in this mixture with low concentration of SF6 are described using the streamer-based theory.

This paper described the morphological simulation of bones for visual frame in the virtual reality system. The paper is divided to three parts. In the first part are described introduction to virtual reality, modelling of bones and fractal geometry. This part also described a method choice for realisation of these problems. As initial tool for solving of this problem, was used L-system. Second, main part contained algorithms for generating of bone body and for bone surface. Algorithm for generating of bone body use fractal geometry functions based on L-systems. Algorithm for generating of bone surface use shrinkwrap method. This process is described as iterative algorithm. The final points set define triangular net of bone surface. By both algorithms is added a possibility of using random parameters values. This feature increased abilities of this process. In the last part are described advantages and disadvantages of this solving. Also are here suggestions for using these bones in virtual models and worlds and examples of outputs from program implementation.

Data security became very crucial topic and for that reason different data encryption algorithms have been developed. They should be parts of complex systems for secure data transfer designed into a single chip. Hardware implementations of a data encryption (decryption) algorithm appear to have a better resistance to attacks. Increasing complexity of electronics systems has made testing one of the most complicated and time consuming problems in system design and production. The microelectronic technologies are driving engineers towards design methodologies called system-on-chip (SoC). Pre-designed cores with testability blocks are used in the effective SoC design process. The paper presents two testable cores for data encryption that use the AES (Advanced Encryption Standard) algorithm. The first one is the " Memory" version and the second one is the "Composite fields" version . Two types of testability techniques - built-in self-test and wrapper architecture - were applied to the developed and implemented Rijndael cores in XILINX Spartan 3 FPGA. The self-testable core consists of a pseudo-random test pattern generator and control logic, which determines faulty or fault-free behaviour of the core. The core with wrapper allows to access all primary inputs and outputs of the core and a deterministic test set can be applied, and responses can be observed through the wrapper. Both testable encryption cores have been modeled by VHDL and they should be used in a complex secure system.

Temperature control is widely used in various processes of physical, electronic, chemical, mechanical and biological systems. All these systems are affected by temperature. Some biological processes must be performed in specified temperature range only. For this, controllers are used to keep temperature value on constant level or defined range. A relative new field of Soft-Computing area is bringing to us a fuzzy logic approach with specific design procedure by using heuristic information. The paper deals with the influence of fuzzy logic on design of fuzzy based non-linear controller for simplified temperature isolated plant, combined from temperature isolated room and heating element driven by amplitude of connected voltage from controller through power stage. The design procedure with respect to the quality of stabilization of the system, starting on system identification and continuing with design of first-order conventional controller and using this setup for fuzzy controller is suggested. This article describes major steps for design of fuzzy PI controller by using human language information which copying linear surface of conventional controller with helpful of defined meta-rules and uniformly arranged fuzzy sets on input-output universe of disclosure. A final design is verified by simulation in Matlab-Simulink environment where basic simulation diagram is drawn. It is shown that fuzzy logic based controller is a new approach of control. The cases in which the well tuned fuzzy logic controller in time varying process is more stable and more robust than the conventional approach in plants were observed. But in case when the fuzzy logic controller is tuned incorrectly it can exhibit limit cycle which can decrease lifetime of the actuator.