JOURNAL OF IRANIAN ASSOCIATION OF ELECTRICAL AND ELECTRONICS ENGINEERS
Year:2004 | Volume:1 | Issue:3|Papers Count:7

To support new Internet applications such as voice over IP and video on demand, it is necessary to design effective congestion control and queue management algorithms. However, such a design is known to be difficult because there are a variety of services supported in the Internet and their various demands for Quality of Service (QoS). This led to a new trend in using alternative techniques, such as Fuzzy Logic Controllers (FLC) which have the ability to cope with the aforementioned problems and provide more flexibility in modeling congestion controllers as well.In this paper, we introduce Fuzzy BLUE queue management algorithm, a fuzzy improvement to the well known BLUE active queue management algorithm. The proposed fuzzy controller uses packet loss and normalized queue length as input linguistic variables. The output of fuzzy controller is the drop probability of BLUE mechanism. To evaluate the performance of the proposed model, simulation software was developed in C++ environment. Based on simulation results obtained from different trials, it can be seen that the proposed Fuzzy BLUE algorithm outperforms the traditional BLUE mechanism.

This paper presents a new topology of an Active filter (AF) and new approach for generating reference currents for the control of active filter and/or a static compensator. The reactive power theory for three phase circuits is used to obtain reference currents as control quantities for the AF. The developed control algorithm is first tested with a standard topology of the AF, thereafter, extended to control a new topology of four pole AF system. The new topology of the AF system is capable to correct Power Factor, compensate harmonics, reactive power and unbalance of three-phase four-wire balanced /unbalanced linear/nonlinear loads. The proposed new topology of the AF system fully compensates neutral current for nonlinear unbalanced loads connected in a three-phase four-wire system, thereby preventing heating and bursting of neutral conductor. The proposed new topology of the AF is free from shoot through fault and exhibits a desired performance. The simulation results show that proposed method and new topology of the AF has good performance.

In this paper, internal overvoltages on HVDC transmission line including switching, AC and DC type overvoltages which are produced respectively by pole to ground fault, interaction between static converters with lines and terminal equipment and uncontrolled starting of remote converters are studied. For this purpose all fault cases are analyzed using traveling waves and modal theories as well as abnormal condition and control strategies in HVDC converters. Then a simplified equivalent circuit of the system in the faults condition with respect to control system operation is introduced. Finally by employing a full bipolar HVDC test system in PSCAD/EMTDC, effects of various parameters on shape and magnitude of overvoltages is studied and their general characteristics are obtained.

Most available control theories assume that a control structure is given at the outset, so they do not use the provisions of control structure design. Control structure design tries to close the gap between theory and practice. In fact control structure design helps the designer to utilize the maximum capability of the system. Control structure design is clearly more important in large-scale systems.Power systems are large-scale systems that are subject to low frequency oscillations. These oscillations degrade the system performance. Power system stabilizers (PSSs) are used to improve the performance of power systems. Determination of the suitable signal for applying to power system stabilizer and determination of the PSS locations is an important issue in power systems.Some control structure design measures are introduced in this paper. These measures are used to determine the inputs, outputs and control configuration based on a systematic procedure that reduces the reliance on experiment and the art of the designer. Applicability of these measures in power systems is shown. Selection of a suitable signal for applying to a power system stabilizer and determination of the PSS locations are the major contributions of this paper.

The Sinuous antennas have a numerous applications in military and civil systems such as direction finding systems and reflector feeds due to their superior broadband characteristics and simultaneous polarization capability. In this paper, design, simulation, construction and measurements of planar sinuous antennas are investigated for four sinuous antennas, having different design parameters and frequency ranges. Feeding sections of these antennas are realized by using Microstrip-Tapered baluns. We have determined the microstrip tapered balun dimensions using Microwave Office simulation program. The performances of these antennas and Baluns are simulated using Ansoft HFSS package.Polarization patterns and return loss characteristics of the designed antennas have been measured in anechoic chamber. These measurements results are in good agreement with theory and simulation.

Some considerable factors for optimizing the signal and noise performances of a matrix amplifier for broadband applications have been discussed. It is shown that proper loading of both external (i.e. input and output) lines and also the middle electrodes improves the device performances. Besides, due to discrepancy between the phase velocities of different lines, the capacitive compensation of these electrodes has a considerable influence on the behavior of the device. Also it is shown that using a tier of HBTs at the output of the matrix amplifier is effective in increasing the power gain of the device. The proposed technique is applicable in improving the performance of Microwave and mm wave ICs.

This article explores the use of independent component analysis (ICA) approach to design a new EEG-based brain-computer interface (BCI) for natural control of prosthetic hand grasp. ICA is a useful technique that allows blind separation of sources, linearly mixed, assuming only the statistical independence of these sources. This suggests the possibility of using ICA to separate different independent brain activities during motor imagery into separate components. This work provides a natural basis for developing an efficient BCI based on single-source data obtained by independent component analysis of multi-channel EEG. The tasks to be discriminated are the imagination of hand grasping and opening and the resting state. Aclassifier is designed for each EEG channel and each source, separately. The features are formed from the 1-s interval of single source computed by ICA and of single-channel EEG data, during each trial of experiment. The mean absolute value, variance, power of beta band, theta band, and alpha band, 1 Hz-spectral components at different frequency band and autoregressive coefficients of order 17 constitute the features. Various feature vectors are formed and are fed into the neural network classifier. The multilayer perceptron (MLP) with back-propagation learning rule is used. The MLP network considered in this study consists of two hidden layers each containing hyperbolic tangent units and two output nodes. The networks were trained with data obtained during 50% of the experimental trials and were validated with data obtained during the subsequent trials. During the training, the feature vector is randomly selected from the training sets and then fed into the network. The learning process is stopped when it is apparent that the generalization performance has peaked. To assess the robustness of the proposed scheme in EEG classification, two different data sets are created for training and evaluating the network. For each of the two data sets obtained during each experiment day, a neural network is trained and evaluated. Then the results are averaged. We observe that single-source which is computed by ICA improved the EEG classification accuracy compared to the Single channel EEG data.