Volume 63 (2012): Issue 1 (January 2012)

The paper re-designs the conventional 6/8 switched reluctance motor (SRM) by a particular way to minimize the losses. The flux loops are shortened by making each two stator teeth incorporated in one magnetic circuit only. Subsequently the flux does not cross in some iron parts of the stator core backs. These unused iron sections are taken off consequently the iron losses decreased. Now the stator consists of three separated sections and they are fixed inside non-magnetic cylinder. The copper losses also are decreased because less copper windings are used to produce the same amount of the output torque as the flux paths become short then the flux density increases. The analysis and comparison of the 6/8 SRM before and after the re-design process are presented. A m files Matlab software is used to simulate the dynamic performance. Then the paper proceeds to examine different control techniques to the new design. The hysteresis and PI controllers are used as classical method to control the SRM. Then the artificial neural network (ANN) is used to test the new control techniques.

Image segmentation is first and very important step in image analysis. The main idea of image segmentation is to simplify and change image into easier and meaningful form to analyze. Image segmentation is process, which locate objects in image. Many segmentation algorithms have been created for different applications. The algorithms are used in traffic applications, army applications, web applications, medical applications, studying and many others. In present time, do not exist restful objective methods to evaluate segmentation algorithms. This paper presents evaluation criterion based on measurement of precision of boundary segmentation. Moreover, the automatic segmentation algorithms in comparison with human segmentation results were tested. Four most used image segmentation algorithms, namely, Efficient graph based, K-means, Mean shift and Belief propagation are compared by designed criterion. The criterion computes three evaluation parameters like precision, recall and F₁ and the results are presented in the tables and graphs at the end of the paper.

This paper investigates the optimal motion planning for an hyper dynamic manipulator. As case study, we consider a golf swing robot which is consisting with two actuated joint and a mechanical stoppers. Genetic Algorithm (GA) technique is proposed to solve the optimal golf swing motion which is generated by Fourier series approximation. The objective function for GA approach is to minimizing the intermediate and final state, minimizing the robot's energy consummation and maximizing the robot's speed. Obtained simulation results show the effectiveness of the proposed scheme.

The paper addresses the problem of the robust output feedback PI controller design for complex large-scale stable systems with a state decentralized control structure. A decentralized control design procedure is proposed for static output feedback control which is based on solving robust control design problems of subsystems' size. The presented approach is based on the Generalized Gershgorin Theorem and uses the so-called equivalent subsystems approach to consider the interactions in the local robust controller design. The resulting decentralized control scheme has been successfully tested on two examples: a linearized model of three interconnected boiler-turbine subsystems and a linear model of four cooperating DC motors where the problem is to design four local PI controllers for a large scale system which will guarantee robust stability and performance of the closed-loop uncertain system.

In this paper we discuss high throughput fixed point-to-point radio links in frequency bands above 70 GHz. These links are interesting as a commercial alternative to fiber connections in IP computer networks. Characteristics of equipment used for radio transmission above 70 GHz are presented. As rain is being the predominant propagation effect that causes link outages, standard rain model is described. Performance analysis in the case of adaptive radio equipment is presented. A new approach to point-to-point radio link performance analysis, that uses two dimensional rain cell models and performance criterion based on end-user experience, is proposed. This approach is compared with the standard point-to-point radio link performance analysis procedure, based on calculation of percentage of time when link is unavailable and percentage of time when link has reduced capacity. It is shown on characteristic examples that proposed performance analysis gives better insight into how rain event would really affect end-user.

In this paper we consider phase process second order statistics of generalized Rice (Beckmann) multipath fading channel. Closed-form expression for JPDF of phase and random FM noise is derived. Furthermore expressions for the PDF and CDF of random FM noise are obtained. The level-crossing rate of the phase process is then obtained for any phase crossing level. Obtained expressions reduces to known ones for Hoyt, Rice and Rayleigh fading channels, since these are the special cases of generalized Rice fading channel. Moreover, derived analytical expressions are compared with results obtained by computer simulation where excellent agreement is achieved. Presented results can be applied for analyzing the statistics of FM spikes in the case of data transmission over generalized Rice fading channels.

In distribution networks switching devices play critical role in energy restoration and improving reliability indices. This paper presents a novel objective function to optimally allocate switches in electric power distribution systems. Identifying the optimized location of the switches is a nonlinear programming problem (NLP). In the proposed objective function a new auxiliary function is used to simplify the calculation of the objective function. The output of the auxiliary function is binary. The genetic algorithm (GA) optimization method is used to solve this optimization problem. The proposed method is applied to a real distribution network and the results reveal that the method is successful.

Brownian oscillator, ie a micron-sized or smaller particle trapped in a thermally fluctuating environment is studied. The confining harmonic potential can move with a constant velocity. As distinct from the standard Langevin theory, the chaotic force driving the particle is correlated in time. The dynamics of the particle is described by the generalized Langevin equation with the inertial term, a coloured noise force, and a memory integral. We consider two kinds of the memory in the system. The first one corresponds to the exponentially correlated noise and in the second case the memory naturally arises within the Navier-Stokes hydrodynamics. Exact analytical solutions are obtained in both the cases using a simple and effective method not applied so far in this kind of problems.

A novel low complexity soft decision technique which allows the decoding of block codes with tree structure is proposed. These codes are shown to have a convenient tree structure that allows Fano decoding techniques to be used to decode them. The Fano algorithm searches through the tree structure of the block code for a path which has the optimal value of the Fano metric function. When a new candidate codeword is found, an optimality check is performed on it by using the threshold. If checked successfully, the candidate codeword is the most likely codeword and the search stops. The basic idea of this approach is to achieve a good error performance progressively in a minimum number of steps. For each decoding step, the error performance is tightly bounded and the decoding is terminated at the stage where either optimum or near optimum error performance is achieved. As a result, more flexibility in the trade off between performance and decoding complexity is provided. Some examples of the tree construction and the soft decision Fano decoding procedure are discussed.