INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE (ENGLISH)
Year:2003 | Volume:14 | Issue:1|Papers Count:14

Square root is one of the operations that are being performed by the hardware in the recent generations of the processors. The hardware implementation of the square root operation is achieved by different means. One of the popular methods is the non-restoring algorithm. In this paper the non-restoring algorithm is improved. Then, in order to create an environment for modular design, a number of blocks are suggested. Using these blocks it is possible to construct any square root circuit with an arbitrary number of inputs. The suggested methodology results in an expandable design with minimum area usage. Therefore, to increase the accuracy of a circuit, special wiring is not required.

Static and dynamic performances of DC/DC converters with PWM and sliding mode controllers are investigated. To improve the performance of the PWM controller, a linear compensator is proposed and used. For sliding controller, important parameters such as sliding coefficients and filter time constant are carefully computed and used. Finally, a DC/DC buck converter with PWM and sliding controller is designed, modeled and constructed. Theoretical and experimental results are compared and the distinguished features and limitations of each control technique are presented.

Load is one of the most important part of the system which has to be modeled accurately in determining the power system status. Measurement and component based are two different approaches for determination of models and dynamic load modeling. The component based load modeling approach is used in this paper. To study the load modeling using component based approach three sets of data such as: load class mix data, load composition data and load characteristics data are required. These three sets of data are usually not available with the power system network utilities. These data are to be prepared by the utilities from other relevant data available. This paper deals with preparation of such data for load modeling purposes. In this paper load class mix data are obtained using the consumer bills, load composition data are obtained using statistic analysis and load characteristics data are obtained using laboratory measurements for different load components in the Boshehr Power System Network (BPSN) simulation studies. It has been observed that the load characteristics data obtained can be used for the entire networks in the world and also load composition data are useful to study all buses for the BPSN. These data can also be useful as a data base for load modeling simulation in all other parts of the country having similar weather and load pattern as that of BPSN. Finally, static and synamic load modeling for BPSN presented in this paper.

Since the core iron loss is directly related to the distribution of localized flux density, an understanding of flux behavior within magnetic cores would therefore lead to a better core design and consequently lower iron losses. This paper describes the purpose built microprocessor controlled measuring system and the developed software for localized flux density in electrical machines cores, as well as the comparison of the localized in-plane flux density in the stator cores of two single-phase induction motors.

This paper introduces a cross-relation (CR) based matched field processor (MFP) for source localization in a shallow water environment where the source generate a random signal. The estimation formulas are given for non-stationary (NS) and wide sense stationary(WSS) random sources. For each case two formulations are proposed, a self-CR and a cross- CR according to which channel output signal used to construct the estimator. We demonstrate the performance of the proposed estimators in source localization, and compare their results with two common estimators, the Bartlett and Minimum Variance (or Capons) matched field processors. The comparison is carried out first by simulation of a wide band, WSS source in a shallow water environment that represents the continental shelf offshore of Vancouver Island. Subsequently, comparison real ship is also carrier out using signal recorded at the same site. The simulation results show that in the low signal-to-noise ratio (SNR) the cross-CR MFP gives superior result in comparison with the self-CR, Bartlett and MV processor with respect to resolution and side lobe level. For real ship signal both CR based MFPs show similar results and in comparison with Bartlett processor, they give lower side lobe level throughout the ambiguity surface and narrower main lobe around source location.

Three-dimensional (3-D) images have recently received wide attention in the medical and industrial applications. The research involves the development of a non-invasive contour capturing method for facial surfaces using the structured light technique. The research focuses on how the images are obtained using the structured light optic system and the subsequent image processing of data to give a realistic 3-D image. The image processing and feature extraction steps in the structured light system produce a binary image consisting of single pixel width pattern of the structured light, which show where the projected lines are. But the contours of the objects mean that certain parts of the projected grid can not be seen on the captured image (dead shadow areas) and breaks occur in the projected grid itself around the holes or sharp edges. In order to produce a 3-D data set, all of the projected lines must be perfectly formed, of single-pixel thickness and be continuous. The grid lines are interpreted and decisions must be made as to whether any given pixel belongs to which line. The algorithms attempt to provide solutions for both noise and breaks and to fill the disconnected areas.

The mian goal of this paper is to design a controller for nonlinear MIMO systems. It is no convenient to control these systems by classical control methods due to their interaction and nonlinearity. This paper proposed one of the appropriate solutions by using fuzzy neural networks. Using neural networks for control of MIMO systems has led to some difficulties such as high computation cost, high training period and disruptive effects of unnecessary inputs. To cope with such nonlinear control problems, research has been underway on control using a special Local Modular Fuzzy Neural Network (LMFNN) based entirely on measured inputs and outputs. This network described by using mathematical relations, and its internal stucture is shown based on block diagram. Then the proposed controller based on this network with appropriate structure and learning algorithm is designed. This controller applied to control nonlinear MIMO systems. Experimental results are used to show the feasibility and effectiveness of the proposed controller.

High dynamic range and adjustability of gain are the main requirements in low-voltage audio-frequency transconductors. There are several methods for designing a low voltage transconductor however, most of them have a fixed transconductance over their input range. In this paper a low distortion ohmic transconductor with adjustable gain has been designed. The circuit can operate from 3 V single power supply and has a very low distortion by choosing normal dimension for W/L of feedback transistors. To consider stray capacitances of the circuit we also perform L-Edit extraction. Then the final circuit has been carried out with 0.8.um CMOS Technology. Simulation has resulted a very low distortion as small as 0.2% for a 1.8Vpp sinusoidal input.

Ultrasonic transit time flowmeters, despite their popularity suffer from the poor accuracy at low values of flow rates and small pipe diameters, where time delays in the range of nanoseconds are to be detected. To overcome this problem, a novel technique is presented in this paper. We have named it "Time Expansion Method, TEM" in which the information carried by phase angles of high frequencies are transferred to low frequencies phase angles, causing the width of the relevant time gate to be expanded. As a result, at a specified flow rate and/or pipe diameter, the system resolution and accuracy is increased. In other words, at a given resolution and accuracy, the lower flow rates inside the narrower pipes can be measured. The theory of the method is developed and its most important parameters are formulated. To practicaly study the method a prototype flowmeter was designed and successfully tested. By applying the innovative "Time Expansion Method" the flow rate information at a frequency of 2500 Khz (the working frequency of ultrasonic piezoelectric transducers) was transferred to the much lower frequency of 200Hz, expanding the related transit time 12500 times, resulting an identical improvement flowmeter sensitivity. Consequently a flowmeter with much less sophisticatied structure and expense can measure an original transit time difference of about 25 picoseconds which matches flow rate sensitivity of about 0.2mm/s or 5.5 lit/ h in a 4" (100-mm) pipe.

An exact expression for the radiated EM fields due to lossless transmission line with a wire or ground plane as return path, and arbitrary terminations has been presented. First the voltage and current along the line have been derived and then a time domain approach has been presented to derive the radiated fields. It has been shown that the radiated field is only function of the voltages and currents in the terminals,

In this paper the method of least squares is employed to design an axially symmetric contradirectional multisection coupled - line coupler together with the impedance matching of real generator and load impedances. An error function is constructed for the required coupling (C) based on the squared magnitude of the ratio of the coupler voltage to that at the incident port. It is minimized to determine the coupler geometry, namely the normalized strip conductor widths (u=w/h) and strip separation from each other (g=s/h) and the coupler length (where h is the substrate thickness). A procedure is presented to provide the initial values of u and g. The computer implementation of the proposed method shows that a proper coupler design is possible for any given coupler length. This is particularly interesting where space limitations impose contraints on the coupler length.

While the Hammerstein expression is computationally attractive for modeling of nonlinear systems, the optimal calculation of filter coefficients is practically cumbersome. A proper solution to this problem is the use of adaptive algorithms. In this paper, the Hammerstein Normalized LMS algorithm is proposed by deriving the corresponding time varying optimal step-size parameter in a closed form. The convergence behavior of this algorithm is inspected using computer simulations. The results show that the proposed algorithm achieves a faster convergence speed compared to the Hammerstein LMS algorithm, practically at the cost of an acceptable increase in computations.