• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.3
• /
• pp.18-31
• /
• 2002

In this study, we introduce a concept of advanced neurofuzzy polynomial networks(ANFPN), a hybrid modeling architecture combining neurofuzzy networks(NFN) and polynomial neural networks(PNN). These networks are highly nonlinear rule-based models. The development of the ANFPN dwells on the technologies of Computational Intelligence(Cl), namely fuzzy sets, neural networks and genetic algorithms. NFN contributes to the formation of the premise part of the rule-based structure of the ANFPN. The consequence part of the ANFPN is designed using PNN. At the premise part of the ANFPN, NFN uses both the simplified fuzzy inference and error back-propagation learning rule. The parameters of the membership functions, learning rates and momentum coefficients are adjusted with the use of genetic optimization. As the consequence structure of ANFPN, PNN is a flexible network architecture whose structure(topology) is developed through learning. In particular, the number of layers and nodes of the PNN are not fixed in advance but is generated in a dynamic way. In this study, we introduce two kinds of ANFPN architectures, namely the basic and the modified one. Here the basic and the modified architecture depend on the number of input variables and the order of polynomial in each layer of PNN structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the ANFPN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed ANFPN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Applied Microscopy
• /
• v.32 no.1
• /
• pp.17-30
• /
• 2002

The retinae of Todarodes pacificus and Octopus minor are divided into four layers that are an outer segment, a rod base region, an inner segment, and a plexiform layer, respectively. The retina of Octopus minor is about $20{\mu}m$ thicker ($400{\sim}420{\mu}m$) than that of Todarodes pacificus ($385{\sim}400{\mu}m$). A retina is composed of visual cells and supporting cells. The microvilli of length $0.6{\sim}0.7{\mu}m$ are packed densely on top of the supporting cells of Octopus minor while they are not found in Todarodes pacificus. The visual cells and supporting cells have pigment granules that exclude light. In case of Todarodes pacificus, the pigment granules of the visual cell are larger ($2.0{\times}0.5{\mu}m$) than those of the supporting cell ($1.0{\times}0.3{\mu}m$). But, the sizes of both cells are similar in Octopus minor. In the upper portion of a visual cell, microvilli shaped like a comb are forming a rhabdome (diameter, 60 nm) of a hexagonal structure. The rhabdome consists of 4 rhabdomere and the total area of a rhabdom of Octopus minor is larger than that of Todarodes pacificus. The synaptosome constructing a plexiform layer in Todarodes pacificus are divided into two types, each of which possess electron dense-core vesicles and electron lucent vesicles, respectively. Octopus minor also has two types of synaptosomes but each type comprises a mixture of electron dense vesicles and electron lucent vesicles, and electron lucent vesicles only, respectively, which is different from the case of Todarodes pacificus.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.38 no.4
• /
• pp.106-127
• /
• 2010

Although multi-layer planting methods are more widely used as a method for clustered planting and environmental programs such as plant remediation, difficulties have been faced in applying those to planting design. This study develops a basic planting model that can be applied to multi-layer planting in basis on an analysis of forest structures in the Seoul area. An optimal number of clusters was determined through the ISA (Indicator Species Analysis), and 7 basic clusters were found through a cluster analysis by using PC ORD 4.0 software specifically developed for ecological analysis. The 7 basic clusters include the following communities: the Quercus acutissima Community, Sorbus alnifolia-Quercus mongolica Community, Pinus rigida-Pinus densifiora Community, Rododendron mucronulatum var. mucronulatum-Quercus mongolica Community, Juniperus rigida-Quercus mongolica Community, Rododendron mucronulatum var. mucronulatum-Pinus densiflora Community, and Rododendron sclippenbachii-Quercus mongolica Community. The study also selected 57 species with at least a 10% frequency among the plant species existing in the Seoul area and suggested both a companion species and available similar alternative species by conducting an additional interspecific association analysis. This study may help to enhance usefulness of the model in architectural planting design. In addition, the two results named above were synthesized to develop a multi-layer planting model that can be utilized in landscape planting design by selecting similar alternative species through the interspecific association analysis, which includes 7 clusters of natural plants. The multi-layer planting model can be widely applied to design planting because the model has an average target cover range based on the average value of a transformed likelihood.

• 전자공학회논문지 IE
• /
• v.43 no.4
• /
• pp.1-7
• /
• 2006

In this paper, two types of blue organic light-emitting device were designed. We have analyzed the characteristics of Type I device without a hole blocking layer, and analyzed the characteristics of Type II device using a hole blocking layer of BCP or BAlq materials with 30 ${\AA}$ thickness. We obtained the ITO having the work function value of 5.02 eV using $N_2$ plasma treatment method with the plasma power 200 W. Type I device structure was ITO/2-TNATA/$\alpha$-NPD/DPVBi/$Alq_3$/LiF/Al:Li, and type II device structure was ITO/2-TNATA/$\alpha$-NPD/DPVBi/HBL/$Alq_3$/LiF/Al:Li. We have analyzed the characteristics of Type I and Type II device. The characteristics of the device were most efficiency on occasion of using a hole blocking layer of BAlq material with 30 ${\AA}$ thickness. Current density was 226.75 $mA/cm^2$, luminance was 10310 $cd/m^2$, Current efficiency was 4.55 cd/A, power efficiency was 1.43 lm/W at an applied voltage of 10V. The maximum EL wavelength of the fabricated blue organic light-emitting device was 456nm. The full-width at half-maximum (FWHM) for the EL spectra was 57nm. CIE color coordinates were x=0.1438 and y=0.1580, which was similar to NTSC deep-blue color with CIE color coordinates of x=0.14 and y=0.08.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.18 no.1 s.116
• /
• pp.62-75
• /
• 2007

This paper describes a shaped-beam antenna for increasing the antenna gain of a radiating element. The proposed antenna structure is composed of an exciting element and a multi-layered disk array structure(MDAS). The stack micro-strip patch elements were used as the exciter for effectively radiating the electromagnetic power to the MDAS over the broadband, and finite metallic disk array elements - which give the role of a director for shaping the antenna beam with the high gain - were finitely and periodically layered onto it. The efficient power coupling between the exciter and the MDAS should be carried out in such a way that the proposed antenna has a high gain characteristic. The design parameters of the exciter and the MDAS should be optimized together to meet the required specifications to meet the required specifications. In this study, a shaped-beam antenna with high gain was optimally designed under the operating conditions with a linear polarization and the frequency band of $9.6{\sim}10.4\;GHz$. Two methods constructed using thin dielectric film and dielectric foam materials respectively were also proposed in order to implement the MBAS of the antenna. In particular, through the computer simulation process, the electrical performance variations of the antenna with the MDAS realized by the thin dielectric film materials were shown according to the number of disk array elements in the stack layer. Two kinds of antenna breadboard with the MDAS realized with the thin dielectric film and dielectric foam materials were fabricated, but experimentation was conducted only on the antenna breadboard(Type 1) with the MDAS realized with the thin dielectric film materials according to the number of disk array elements in the stack layer in order to compare it with the electrical performance variations obtained during the simulation. The measured antenna gain performance was found to be in good agreement with the simulated one, and showed the periodicity of the antenna gain variations according to the stack layer number of the disk array elements. The electrical performance of the Type 1 antenna was measured at the center frequency of 10 GHz. As the disk away elements became the ten stacks, a maximum antenna gain of 15.65 dBi was obtained, and the measured return loss was not less than 11.4 dB within the operating band. Therefore, a 5 dB gain improvement of the Type 1 antenna can be obtained by the MDAS that is excited by the stack microstrip patch elements. As the disk array elements became the twelve stacks, the antenna gain of the Type 1 was measured to be 1.35 dB more than the antenna gain of the Type 2 by the outer dielectric ring effect, and the 3 dB beam widths measured from the two antenna breadboards were about $28^{\circ}$ and $36^{\circ}$ respectively.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.19 no.1
• /
• pp.23-28
• /
• 2006

In this paper, breakdown voltage and on-resistance characteristics of the surface doped SOI RESURF LDMOSFET were investigated as a function of surface doping depth. In order to verify the variation of characteristics, two-dimensional device simulation was carried out. Breakdown voltage of the proposed structure is varied from $73 {\~}138V$ while surface doping depth varied from $0.5{\~}2.0{\mu}m$. And on-resistance is decreased from $0.18{\~}0.143{\Omega}/cm^2$ while surface doping depth increased from $0.5 {\~}2.0{\mu}m$. Maximum breakdown voltage of the proposed structure is 138 V at $1.5{\mu}m$ depth of surface doping, yielding $22.1\%$ of improvement of breakdown voltage in comparison with that of the conventional SOI RESURF LDMOSFET with same epi-layer concentration. On-resistance characteristic is also improved about $21.7\%$.

• Steel and Composite Structures
• /
• v.26 no.3
• /
• pp.273-287
• /
• 2018

Buckling and free vibration analysis of sandwich micro plate (SMP) integrated with piezoelectric layers embedded in orthotropic Pasternak are investigated in this paper. The refined Zigzag theory (RZT) is taken into consideration to model the SMP. Four different types of functionally graded (FG) distribution through the thickness of the SMP core layer which is reinforced with single-wall carbon nanotubes (SWCNTs) are considered. The modified couple stress theory (MCST) is employed to capture the effects of small scale effects. The sandwich structure is exposed to a two dimensional magnetic field and also, piezoelectric layers are subjected to external applied voltages. In order to obtain governing equation, energy method as well as Hamilton's principle is applied. Based on an analytical solution the critical buckling loads and natural frequency are obtained. The effects of volume fraction of carbon nanotubes (CNTs), different distributions of CNTs, foundation stiffness parameters, magnetic and electric fields, small scale parameter and the thickness of piezoelectric layers on the both critical buckling loads and natural frequency of the SMP are examined. The obtained results demonstrate that the effects of volume fraction of CNTs play an important role in analyzing buckling and free vibration behavior of the SMP. Furthermore, the effects of magnetic and electric fields are remarkable on the mechanical responses of the system and cannot be neglected.

• Journal of Surface Science and Engineering
• /
• v.43 no.4
• /
• pp.170-175
• /
• 2010

The GaAs epitaxial layers were grown on Si(100) substrates by molecular beam epitaxy(MBE) using the two-step method. The Si(100) substrates were cleaned with different surface cleaning method of vacuum heating, As-beam, and Ga-beam at the substrate temperature of $800^{\circ}C$. Growth temperature and thickness of the GaAs epitaxial layer were $800^{\circ}C$ and 1 ${\mu}m$, respectively. The surface structure and epitaxial growth were observed by reflection high-energy electron diffraction(RHEED) and scanning electron microscope(SEM). Just surface structure of the Si(100) substrate cleaned by Ga-beam at $800^{\circ}C$ shows double domain ($2{\times}1$). RHEED patterns of the GaAs epitaxial layers grown on Si(100) substrates with cleaning method of vacuum heating, As-beam, and Ga-beam show spot-like, ($2{\times}4$) with spot, and clear ($2{\times}4$). From SEM, it is found that the GaAs epitaxial layers grown on Si(100) substrates with Ga-beam cleaning has a high quality.

• Journal of the Korean Vacuum Society
• /
• v.9 no.3
• /
• pp.267-272
• /
• 2000

Electroplating Cu was deposited on Si(100) wafer after seed Cu was deposited by sputtering first. TaN was deposited as a diffusion barrier before depositing the seed Cu. Electroplating Cu thin films show highly (111)-oriented microstructure for both before and after annealing at $450^{\circ}C$ for 30min and no copper silicide was detected in the same samples, which indicates that TaN barrier layer blocks well the Cu diffusion into silicon substrate. After annealing the electroplating Cu film up to $450^{\circ}C$, the Cu film became columnar from non-columnar, its grain size became larger about two times, and also defects density of stacking faults, twins and dislocations decreased greatly. Thus the heat treatment will improve significantly electromigration property caused by the grain boundary in the Cu thin films.

• Journal of the Korean Geophysical Society
• /
• v.9 no.3
• /
• pp.231-240
• /
• 2006

The crustal structure of the Korean Peninsula was investigated by analyzing phase velocity dispersion data of Rayleigh waves. Earthquakes recorded by three component broad-band velocity seismographs during 1999-2004 in South Korea were used in this study. The fundamental mode Rayleigh waves were extracted from vertical components of seismograms by multiple filter technique and phase match filter method. Phase velocity dispersion curves of the fundamental mode signal pairs for 14 surface wave propagation paths on the great circle in the range 10 to 80 sec were computed by two-station method. Treating the shear velocity of each layer as an independent parameter, phase velocity data of Rayleigh wave were inverted. All the result models can be explained by a rather homogeneous crust of shear-wave velocity increasing from 2.8 to 3.25 km/sec from top to about 33 km depth without any distinctive crustal discontinuities and an uppermost mantle of shear-wave velocity between 4.55 and 4.67 km/sec. Our results turn out to agree well with recent study of Cho et al. (2006 b) based on the analysis of seismic background noises to recover short-period (0.5-20 sec) Rayleigh- and Love-wave group velocity dispersion characteristics.