摘要:

Analysis and design of a Ka‐band (20‐GHz) satellite offset‐fed shaped reflector antenna with a single feed for producing a contoured beam to cover the Taiwan region is presented. The design of this shaped reflector antenna is for future geostationary (GEO) satellite communication or DBS (direct broadcasting satellite) applications to serve the Taiwan region. The geometrical optics (GO)/aperture field (AF) method and aperture phase optimization procedures are employed for reflector surface shaping to produce the desired contoured beam. The physical optics (PO) method is used to accurately calculate the far‐field radiation pattern. Numerical simulation results show that the designed shaped reflector produces a Taiwan region converge beam and effectively reduces the antenna gain level in the nearby area of mainland China. Further reduction of the antenna gain level can be achieved by shifting the boresight direction of the satellite antenna for a certain small angle.

ISSN:0253-3839    

DOI:10.1080/02533839.1995.9677686

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

The effects of processing temperature and mechanical mixing speed on glass transition temperature, and pot life, activation energies for chemical reaction and molecular relaxation of diaminodiphenyl sulphone (DDS) cured tetraglycidyl diaminodiphenyl methane (TGDDM) epoxy have been evaluated. Viscometry study indicates that the pot life of TGDDM/DDS resin mixtures is significantly affected by the cure temperature and mixing speed. The anomalous behavior at the mixing speed of 900 rpm is likely to be due to a result of poor mixing. Differential scanning calorimetry (DSC) studies show that the glass transition temperature of uncured resin exhibits aTgoof 5 ‐ 19 °C, which decreases with increasing mixing speed. In general, a higher mixing speed would create a higher mixing power of the system, is a result of increasing fluid molecular friction and shear stress. This may be a consequence of molecular chain scissions of oligomer that probably reduce the average molecular weight, and therefore reduce the glass transition temperature. The activation energy for the curing reaction of this epoxy system, around 91 ‐ 98 kJ/mole, was found to be only slightly affected by the mixing speed. Dynamic mechanical analysis (DMA) studies show that two main relaxations can be ob‐ served for this epoxy system. The low temperature relaxation at ‐20 ‐ ‐50 °C is theβtransition, with a low activation energy of about 56 ‐ 59 kJ/mole under increasing mixing speed, whereas the high temperature relaxa‐tion at 230 ‐ 260 °C is the a transition, which is associated with the glass transition temperature, with an activation energy of 240 ‐ 400 kJ/mole, and was strongly affected by the mixing speed. It is believed that these activation energies provide a unique method of characterizing the molecular segmental motion of epoxy networks which were affected by the mixing process.

ISSN:0253-3839    

DOI:10.1080/02533839.1995.9677687

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

The structure of the Reynolds stress in a turbulent flow over the multiple windbreaks of half‐circular sections across‐wind was investigated experimentally in the boundary layer wind tunnel. The quadrant analysis theory was applied to analyze the turbulent stress signals by partitioning the Reynolds stress into inward interaction, ejection, outward interaction, and sweep events in accordance with whichever quadrant the (u’, w‘)belongs to.u’ andw’ are the stream wise and transversewise velocity fluctuations, respectively The results show that whenZ/H <1,X/H <20, the sweep events are the largest contributors to the Reynolds stress followed by the ejection events while the inward and outward interactions are negative contributors.Zis the distance from the wall,Xis the downstream distance, andHis the height of windbreak. The stress contributions by sweeps and ejections are almost the same whenZ/H< 1,X/H >30.

ISSN:0253-3839    

DOI:10.1080/02533839.1995.9677688

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

Based upon the selective attention in human visual perception, a conceptual computation model performing the recognition of the meaningful dotted pattern from a dotted image will be presented. This consists of two major processes. A passive process performs images partition for the segmentation, and an active process performs distance computation for the pattern recognition. The principle of this model is illustrated by a set of Arabic numerals, and its further application is also pointed out.

ISSN:0253-3839    

DOI:10.1080/02533839.1995.9677689

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

In contrast to the conventional LQG method using calculus of variations, this paper presents an alternative approach based on operator theory and spectrum factorization. This approach, which works toward transfer function domain, has merits in its simple and straightforward development. The novelty of this method is not the results achieved, but rather in their derivation. A loop shaping design procedure which incorporates the classical loop shaping into LQG/LTR synthesis is developed such that the robustness properties of the optimal full state feedback are guaranteed atboththe plant input and the plant output. The design procedure is applied to a position controller design in a servo system, and the performance is evaluated by a digital signal processor, the TMS320C30. Experimental results have shown effectiveness of this design method.

ISSN:0253-3839    

DOI:10.1080/02533839.1995.9677690

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

The convergence property of thelp‐norm algorithm for polynomial‐perceptron having different error signal distributions will be analyzed in this paper. To see the effect of error signal on the convergence rate, two types of activation functions are considered in the analysis: one is of a linear type and the other is of a sigmoidal type. Different activation functions yield different ranges of output signal and, in turn, yield different error signal distributions. Linear activation function causes the error signal to be distributed in an uncertain way, while sigmoidal activation function causes it to be distributed in a tightly bounded region. Based on this difference the convergence property of thelp‐norm algorithm, 1 ≤p ≤2, is investigated in this paper. Expressions of average learning gains are obtained in terms of the power metricp, the error probability, and the upper bound of the error signal distribution. Analytic results indicate that it is of particular value in using thelp‐norm algorithm for the perceptron using sigmoidal activation functions. Computer simulation of an adaptive equalizer using this algorithm confirms the theoretical analysis.

ISSN:0253-3839    

DOI:10.1080/02533839.1995.9677691

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

ISSN:0253-3839    

DOI:10.1080/02533839.1995.9677675

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor