Application of evolutionary optimization techniques for finding the optimal set of concentric circular antenna array

In this paper the maximum sidelobe level (SLL) reductions, optimal beam patterns and optimal beam widths of various designs of three-ring planar concentric circular antenna arrays (PCCAA) are examined using three different classes of evolutionary optimization techniques to finally determine the global optimal three-ring PCCAA design and then establish some sort of ranking among the techniques. Apart from physical construction of a PCCAA, one may broadly classify its design into two major categories: uniformly excited arrays and non-uniformly excited arrays. The present paper assumes non-uniform excitations and uniform spacing of excitation elements in each three-ring PCCAA design and a design goal of maximizing SLL reduction associated with optimal beam patterns and beam widths. The design problem is modeled as an optimization problem for each PCCAA design and solved using different evolutionary optimization techniques to determine an optimum set of normalized excitation weights for PCCAA elements, which, when incorporated, results in a radiation pattern with optimal (maximum) SLL reduction. Among the various PCCAA designs, one which yields the global minimum SLL with global minimum first null beamwidth is the global optimal design. In this work the three-ring PCCAA containing (N₁ = 4, N₂ = 6, N₃ = 8) elements proves to be such global optimal design. The optimization techniques employed are real coded GA (RGA), canonical PSO (CPSO), craziness based PSO (CRPSO), evolutionary programming (BEP), hybrid evolutionary programming (HEP). While ranking the techniques after 30 total runs for each design, HEP, CRPSO, RGA, CPSO, BGA hold the first five ranks in order of optimization capability. HEP yields global minimum SLL (?32.86 dB) and global minimum BWFN (77.0°) for the optimal design. BEP often changes the rank from second to fifth depending on the design set. Further, when compared to a uniformly excited PCCAA having equal number of elements and same radii a reduction of major lobe beamwidth is also observed in the optimal non-uniformly excited case.     

Detecting anomalous packets in network transfers: investigations using PCA,autoencoder and isolation forest in TCP

Machine Learning - Large-scale scientific workflows rely heavily on high-performance file transfers. These transfers require strict quality parameters such as guaranteed bandwidth, no packet loss...     

A Three-Phase AC-to-DC Power Transistor Converter-Controlled DC Motor Drive

A three-phase ac-to-dc power transistor converter that is suitable for speed control of a dc motor and for a variable dc power supply is investigated. The availability of power transistors of fairly high power rating has made the task of ac-to-dc conversion relatively easy due to simplification of the converter circuit configuration. The external performance characteristics of the power transistor converter-controlled separately excited dc motor are obtained by employing the equal pulsewidth modulation (EPWM) control technique with 18 pulses per half-cycle of the supply voltage and are verified experimentally. There is good agreement between experimental and theoretical results. The external performance is also evaluated under constant torque operation. The use of inductance-capacitance (LC) filters for reducing the effect of predominant supply harmonics is briefly discussed. Experimental oscillograms of some typical waveforms are illustrated to verify the basic principles of operation.     

Comments on “A fuzzy neural network and its application topattern recognition”

This note analyzes the unsupervised fuzzy neural network (FNNU) of the original paper by Kwan and Cai (ibid., vol.2, p.185-93, 1994) and finds the following: the FNNU is a clustering net, not a classifier net, and the number of clusters the network settles to may be less or more than the actual number of pattern classes (sometimes it could even be equal to the number of training data points); the huge number of connections in the FNNU can be drastically reduced without degrading its performance; and the algorithm does not have any learning capability for its parameters. Computational experience shows that usually the performance of a multilayer perceptron (MLP) is comparable to that of even a supervised version of FNN (trained by gradient descent algorithm) in terms of recognition scores, but an MLP has a much faster convergence than the supervised version of FNN     

Effects of mass transfer and free convection on the unsteady mhd flow past a vertical porous plate with constant suction

Effects of free convection currents and mass transfer on the unsteady flow of an electrically conducting and viscous incompressible fluid past an infinite vertical porous plate subjected to uniform suction, in the presence of transverse magnetic field, have been studied taking into account that the external flow velocity varies periodically with time in magnitude but not in direction. The effect of the induced magnetic field has been neglected. Approximate solutions to the transient flow, the amplitude and the phase of the skin-friction and the rate of heat transfer have been derived. During the course of the discussion, the effects of the Grashoff number Gr, the modified Grashoff number Gc (depending on the concentration difference), the Schmidt number Sc, the Eckert number Ec, the magnetic field parameter M, and the frequency ω have been discussed.