JOURNAL OF FACULTY OF ENGINEERING (UNIVERSITY OF TABRIZ)
Year:2003 | Volume:29 | Issue:1 (31 ELECTRICAL ENGINEERING)|Papers Count:11

This paper presents a three-phase PWM inverter in which power transistors are used as switches. In this type .of inverter current-sharing reactors are used, and in each phase four transistors are used compared to two transistors used in conventional inverters. By using this inverter more amount of current can be attained as well as output voltage harmonics are greatly reduced and output voltage waveform will be a curve of five levels of voltage which resembles more closely to a sine wave. The correction role of increase in number of transistors in each phase, which causes increase in power level and a reduction in output voltage harmonics has been shown using time-scale waveforms, frequency spectrum and the effect of variations of modulation index on amplitude of harmonic component. Finally, experimental results of such, an inverter design in a laboratory have been has presented which completely agrees with theory.

In general, most of satellites have nonlinear dynamic together with uncertainty and restricted constraints on their actuators. In this paper, stability of satellite by using sliding control as a robust control is proved in such a way that influences of above-mentioned limitations has been reduced. In this regard, after deriving dynamic and kinematic equation based on quaternion, sliding surface will be designed such that it will be stable. It is then illustrated that discrete sliding control law will not guarantee stability, but continuous one will guarantee satellite stability by proving. Finally, fine performance of designed control law is shown by simulation on a spinning satellite considering practical aspects.

In this experimental work, the polycrystalline samples of ternary compound InGaTe2 and binary compound PbT2 were initially synthesized and then by grinding and mixing given amounts of PbTe and InGaTe2 (as impurity), two types of compounds given by the formula (PbTe)1-x (InGaTe2)x having two different mole percentages of x=0.5 and x=1 were synthesized, respectively. By studying the variations of thermoelectric power (a) and electrical conductivity (er) of samples versus the temperature, the values of energy gap for the above two compounds were evaluated. Experimental results show that, the pure PbTe is a typical p type semiconductor, but introducing the ternary compound InGaTe2 into the crystal, its conductivity changes from p to n. The energy gap of PbTe is increased by increasing the mole percentage of the InGaTe2 impurity. Increasing the impurity content, up to x=0.5 mole percent causes the conductivity to increase, but further increase of the impurity content, causes the conductivity to decrease. Thermoelectric power increases with the temperature and its maximum value is larger for the compound with lesser impurity content.

Low-Density Parity-Check (LDPC) codes were introduced by Gallager in 1962. After more than three decades the near Shannon limits performance of these codes rediscovered by MacKay and Neal in 1996.Currently, LDPC code is one of the hot topics for research. In spite of simple iterative decoding for these codes by Sum-Product Algorithms (SPA), encoding and constructing of these codes have major difficulty. There exist a variety of methods for constructing LDPC codes. In this paper, we present and compare the most important of these methods. In addition we introduce a new constructing method to enhance the error performance.

Discontinuities are used to produce impedance and transfer functions in microwave devices. Various types of filters, phase shifters, mixers and high frequency oscillators may employ some kind of discontinuity. The longitudinal E-plane metallic strip is the main part of pure meta: insert (PM\) and fin-line filters. In this paper, the least squares boundary residual method (LSBRM) is employed as an effective procedure for the analysis of longitudinal E-plane thick and thin metallic strips inside waveguides. The fields inside every section of the waveguide are expanded in terms of modal functions. Then, the boundary conditions on the tangential electric and magnetic fields over the waveguide interfaces (together with a weighting factor) are used to construct an error function. The minimization of the error function determines the amplitude of the excited modes and finally the equivalent circuit of the metallic strip. The numerical results obtained by the computer simulation agree very well with those derived by other methods, in cases available in the literature. The present method is effectively employed for the design of microwave band pass filters.

Electromagnetic launchers have very important advantages over powder guns, and then increasing the efficiency of these launchers is one of the important goal of researchers around the world. One of the most important factor to increase efficiency is enlargement "inductance gradient" or L'. This paper presents two methods for calculation L' in cylindrical rails construction. The results of two methods is compared and showed they are almost identical. The second method used for square rails construction and the results is compare with reported result by famous scientist. Finally inductance gradient L' for railgun with auxiliary rails calculate by second method. The methods proposed in this paper have advantages over to numerical method because compact formulas obtained for L' . By these formulas we can study the effects of different dimensions on L' values. A table of L' for different dimensions of railgun is presented in this paper.

Propagation analysis of the acoustic waves generated by the vessels and submarines nearby the coast, where the seabed shape is not uniform and is a function of distance, needs a method which consider not only acoustic energy transferred from the seabed in shallow waters but can be applied in low frequencies as well. In this article wave equation in cylindrical coordinate is formulated and then by separating functions, parabolic equations are developed. Considering the variable seabed shape, this equation is solved using Finite Difference Method. Solving this equation provides development of software which can calculate acoustic signal amplitude reached to any point, based on the input data such as frequency, signal amplitude and, amplitude and radiation pattern of the source, title angle, sound velocity in the water and under the seabed and also environment parameters.

This paper presents equivalent circuit models for both relative intensity noise (RIN) and phase/frequency noise spectrum (FNS) in a single semiconductor laser diode. The model for the electrical phase noise of a single mode laser is proposed for the first time. These equivalent circuit models are derived from the rate equations including the Langevin noise sources. Then, RIN and FNS are calculated in terms of electrical parameters. Finally, we explain an indirect experimental method used to measure RIN and FNS of a typical optical communication laser diode. Behavior of the experimental results is in agreement with those calculated by circuit models.

The classic traveling - salesman problem is to determine a tour that will minimize the total distance or cost involved in visiting several cities and returning to the starting point. This paper describes a new algorithm for accelerating the computational performances of branch exchange heuristics for symmetric traveling salesman problem. The improvement in performance is obtained by considering only exchanges that have a good chance of producing a better solution. This algorithm is faster than other heuristics' algorithms and computation times increase much less rapidly with problem size. The new algorithm makes it possible to solve large-scale traveling salesman problems.

Cumulative Probability Distribution (CPD) is a statistical approach in comparison of multi-path fading models. One certain problem was encountered in this comparison method, which is its inability in distinguishing between certain sets of data. This problem has been covered thoroughly in this paper and a modified two dimensional CPD (MCPD) is introduced. It outputs three-dimensional and uses cross sectioning techniques. It also has the advantage of distinguishing between signals that are very similar to each other and have very close CPD.