AMIRKABIR
Year:2001 | Volume:12 | Issue:48|Papers Count:25

The paper presents a simple yet powerful fuzzy logic based static VAR compensator (FLSVC) applied to an industrial power network consisting of three phase synchronous, asynchronous and DC motor loads. In the proposed fuzzy logic controller (FLC), the speed and acceleration variations of a specific machine are taken as the inputs. To demonstrate the effectiveness and capabilities of the employed FLSVC, several nonlinear time domain are performed. The results how that over a wide range of operating conditions and disturbances, the FLSVC improves remarkably the voltage profile and the overall dynamic performance.

Hybrid laminates, materials that consist of alternate aluminium alloy and composite laminae, are functionally graded material systems of particular interest to the aerospace industry. In this study, the use of genetic algorithm in stacking sequence optimization of thin rotating hybrid laminated cylindrical shells is studied. The centrifugal and Coriolis forces are included in the theory and the Navier-type solutions are presented for simply supported boundary conditions. To validate the formulation, results are compared to those from other shell theories. For the optimization problem, two different objective functions that include the weight and Natural frequency of the rotating cylinder are considered. the objective functions are constrained for the maximum number of contiguous plies. Optimal stacking sequences at different rotational speeds are presented.

A novel kind of microwave mixer is introduced which its operation is based on traveling wave mode. It is shown that increasing the drain-source capacitance of a FET will simultaneously improve the conversion gain and the noise figure of the mixer. Besides, It is shown that exciting the input signal from one end of the gate electrode and extracting the out put signal from the opposite end of the drain electrode. Can improve the device performance,  provided that two remaining ends of the gate and the dram are properly loaded. The analysis is performed using sliced modeling of active circuits. .This structure can be used in high power application because it is specially used for FETs with large gate width. Also due to the traveling wave structure, it is useful for wideband circuits. This idea can also be applied to analyses and improve the signal and noise performances of other microwave and mm-wave integrated circuits.

With use of the Trotter's technique simulation for one - dimensional XYZ Heisenberg systems of spin -1/2 have been performed. It is shown that the one- dimensional quantum systems can be transformed to the two- dimensional classical systyems by Trotter's technique. The extra dimension found to be like an imaginary time which has been related to the temperature of the system. Several chanis of one - dimensional spin systems are simulated by Monte carlo method and the thermodynamic quantities of these are calculated at low temperature by means of this technique. Comparison is made with Bonner &Fisher's exact data which is found tobe in good agreement with their results.

A novel design procedure is presented that determines the weight matrices of a neural network conroller for a class of nonlinear multi-input systems. The plant is a linear system supposed to bounded inputs. In contrast to concentional recursive training methods, the novel design method is based upon an one-parameter optimization which directly specifies the weight matrices. The stability of the closed loop system is guaranteed analytically.

With the growing use of Electrocardiogram signal for detection and diagnosis of heart disorders, the efficient transmission of this signal over telephone lines or mobile radio is becoming more and more important. A specialist in a major hospital can use the system and receive information on a patiend in a rural area and send back a plan of treatment specific for that patient.In this article, we present different stage for design and fabricate an intelligent telephonic transmission System with 12-Lead ECG. This paper includes two sections on hardware and software. The digital hardware circuitry takes benefit of an microprocessor. Microprocessor software takes hospital telephone number and transmit/receive serial data to modem. In software section a real time algorithm is used to draw ECG signal and save this digital data to remote computer hard disk and compression ECG signal via four algorithms.

The behavior of discharge lamps (Na and Hg), which are widely used in the lighting of passages, streets, stadiums and factories, and their role in the generation of harmonic distortions in a power system have been studied both theoretically and experimentally. Due to the poor power factor (0. 4-0. 55), these lamps are usually equipped with capacitors, which can also affect the severity of harmonic generation. To reduce system unbalance status, three single-phase components are commonly installed in equal numbers within the three phases. As a result, triplen harmonics are basically additive in the neural, generating a dominant third harmonic current. It is shown that the third harmonic current is larger when the capacitors star-point is disconnected. The good agreement between the theoretical results and their experimental counterparts demonstrates the validity of the theoretical model used in this paper.

In this paper Kalman filtering problem for 2-D systems was considered. For this purpose the wave Advanced Model (WAM) was used. Since this model is variable structure it is not possible to extend the theories available for I-D case to 2-D case. Thus, first it is necessary to derive the new results directly for the WAM which is I-D equivalent of 2-D systems. Second by minimization of estimation error, the Kalman filtering in 2-D systems will be obtained.

An Intelligent Tutoring System (ITS) can be effectively employed for a power plant simulator so that the need for instructor is minimized. In this paper using the above concept as well as object oriented programming and SIMULINK Toolbox of MATLAB, an intelligent tutoring power plant simulator is proposed. Its successful application on a typical 11 MW power plant is demonstrated.

Hip replacement is used for inoperable femur head injuries and femur fractures where internal fixation can not be used. This operation is one of the most common orthopedic operations that many research have been done about it. Among these we can mention implant and cement materials and composites optimization and also implant shape optimization. This study was designed to optimize artificial hip joint position (placement) to decrease maximal von misses stress. First, a model of femur and implant were made and then a computer program was written with the ability to change the position of implant range in the femur. In each of these positions, the program simulated femur and implant according to finite element method and made, applied forces were weight and muscle traction. Our findings show that a small deviation of the implant  from femur bone center causes a considerable decrease in von misses stress that consequently results in longer maintenance of, the implant.

In this paper the large amplitude and non - linear vibration of a string is considered. The initial tension, lateral vibration amplitude, diameter and the modulus of elasticity of the string have main effects on its natural frequencies. Increasing the lateral vibration amplitude makes the assumption of constant initial tension invalid. In this case, therefore, it is impossible to use the classical equation of string with small amplitude transverse motion assumption. On the other hand, by increasing the string diameter, the bending moment effect will increase dramatically, and acts as an impressive restoring moment.Considering the effects of the bending moments, the nonlinear equation governing the large amplitude transverse vibration of a string is derived. The time dependent portion of the governing equation has the from of Doffing equation. This equation is solved using the perturbation theory. The results of the analysis are shown in appropriate graphs, and the natural frequencies of the string due to the non-linear factors are compared with the natural frequencies of the linear vibration of a string without bending moment effects.

In a new de-regulated power industry, the Independent System Operator (ISO) provides open access; to transmission System; for all existing transactions. For a reliable operation, ISO is also responsible for providing ancillary services one of which is reactive power requirement. In this paper a new reactive power pricing strategy for an open access environment is proposed in which by using Fair resource allocation concept, initially, each transaction consumption of the generation of each reactive power resource is determined. The costs, namely investment and operational are then calculated.

In this paper, a numerical simulation for analysis of the deep drawing process of ax symmetric products (Cup Drawing) is given. It is based on a combination of both Finite Element and Finite Difference Methods. The analysis of process is thus simplified and the computational time is reduced. Bending and unbending at the die profile, friction, shearing force and change in thickness are taken into account. The equilibrium equations of forces in the meridian and thickness directions including the effect of the contact force of the tool and the equilibrium of peripheral moment are satisfied in all parts of the material. A contact model has also been given to judge the contact stating of each element with tool.By using the simulation, the deformation features of the circular blank can be obtained. These include the geometrical shapes of blank at each deformation step, stress and strain distributions and change of thickness. Moreover, the results of analysis regarding peripheral strain are compared with the experimental results. Good agreements between the analysis and experimental results show the validity of the simulation.

In design and optimization of rail vehicles two important parts are dynamic behavior and stability analysis when passing a curve. In passing a curve one, has to consider new problems which are not present analysis of straight line motion, such as centrifugal force, different loading on right and left wheels, and track inclination, ... All these are important in dnamics of vehicle. Since the stability is effected when the vehicle passes through a curve, it is necessary to do a full, stability study.In a previouse work [9], for increasing the critical speed a certain type of flexible wheel was proposed, and it was showed that the lateral stability was increased due to wheel fexibity. To further conupt of stability furthure the concept the stability of that vehicle when passingthrovgh through a curve in studied here.In this work it is shown that flexible wheel will increase the lateral stability of the wheel axle set when passing on a curve.

Development of discrete - time Gabor transform for vibration signature analysis of machinery and fault diagnosis is the main aim of this paper. At first, we use time - frequency Gabor. transformation for fault diagnosis of machinery and then, we pay attention characteristics evaluation of this method and compare with other present time or frequency domain methods for fault diagnosis of a healthy and faulty rotating shaft with the same working condition and compare the results for fault diagnosis. This procedure is repeated and time domain vibration data for a faulty gearbox is used for fault diagnosis

This Paper describes a new method for multiobjective coordination of demand side management procedures for corrective actions. This method relies on a two stage very fast simulated reannealing and is capable to recognize the effectiveness of corrective action of demand side. A number of cases are presented to illustrate the consequences of introducing of demand side action. In particular, it is shown that using this method provides a basis for negotiations between the system operators and potential providers of demand side corrective actions.

A neuro-based approach is proposed for locating cloud-to-ground lightning strokes. Due to insufficient experimental data, we have used the results of an electromagnetic simulator for training the developed artificial neural network. The simulator utilizes the well-known transmission line and is capable of predicting the electromagnetic field due to a return stroke channel (RSC) for various parameters associated with the shape of the channel base-current. The training process has been successfully done using the Levenberg-Marquard technique. The simulation results demonstrate that the RSC locations can be predicted with an absolute error not greater than 1 km for RSCs located within 80 km of a lightning detection station.

In this paper, Langevin rate equations governing the behavior of a single-mode semiconductor laser are solved numerically and various noise characteristics are calculated from the results of numerical solution. Using the stochastic dynamic equations of field and population inversion, including nonlinear gain, we obtain intensity autocorrelation function, intensity noise versus injection current characteristic, intensity and frequency noise spectra, lineshape, and linewidth of laser.

Lexicons play a central role in natural language processing (NLP) systems, so design and implementing computational lexicons are of foundamental activities in development of such systems. In this paper first, we define lexicon, introduce its parts, study the difference factors in lexicons and describe their various types. Then after discussing the theoretical aspects design and implementation of lexicons, We will study a practical instance in a Persian natural language processing system. In this system, we will consider the mentioned theretical parameters and factors and initiate them practically.

Modeling is considered as the first step in scientific investigations. Several alternative models may be candidated to express a phenomenon. Scientists use various criteria to select one model between the competing models. Based on the solution of a Fuzzy Decision-Making (FDM) problem, this paper proposes a new method in model selection. The method enables the scientist to apply all desired validity criteria, Including both statistical and visual, even linguistic criteria, systematically by defining a proper Possibility Dititribution Function (PDF) due to each criterion. Finally, minimization of a utility function composed of the PDFs will determine the best selection. The method is illustrated through a modeling example for the "Average Daily Time Duration of Electrical Energy Consumption in Iran