• Journal of Environmental Impact Assessment
• /
• v.12 no.4
• /
• pp.271-279
• /
• 2003

Temperature and humidity forecasting have been performed using artificial neural networks model(ANN). We composed ANN with multi-layer perceptron which is 2 input layers, 2 hidden layers and 1 output layer. Back propagation algorithm was used to train the ANN. 6 nodes and 12 nodes in the middle layers were appropriate to the temperature model for training. And 9 nodes and 6 nodes were also appropriate to the humidity model respectively. 90% of the all data was used learning set, and the extra 10% was used to model verification. In the case of temperature, average temperature before 15 minute and humidity at present constituted input layer, and temperature at present constituted out-layer and humidity model was vice versa. The sensitivity analysis revealed that previous value data contributed to forecasting target value than the other variable. Temperature was pseudo-linearly related to the previous 15 minute average value. We confirmed that ANN with multi-layer perceptron could support pollutant dispersion model by computing meterological data at real time.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.603-611
• /
• 2001

This paper proposes a novel fast layer-by-layer algorithm that has better generalization capability. In the proposed algorithm, the weights of the hidden layer are updated by the target vector of the hidden layer obtained by least squares method. The proposed algorithm improves the learning speed that can occur due to the small magnitude of the gradient vector in the hidden layer. This algorithm was tested in a handwritten digits recognition problem. The learning speed of the proposed algorithm was faster than those of error back propagation algorithm and modified error function algorithm, and similar to those of Ooyen's method and layer-by-layer algorithm. Moreover, the simulation results showed that the proposed algorithm had the best generalization capability among them regardless of the number of hidden nodes. The proposed algorithm has the advantages of the learning speed of layer-by-layer algorithm and the generalization capability of error back propagation algorithm and modified error function algorithm.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2006.08a
• /
• pp.695-698
• /
• 2006

The photosensitive poly-siloxane material used as the passivation layers for the conventional back channel etched (BCE) thin film transistors (TFTs) has been investigated. Through the organic material, the TFT array fabrication process can be reduced and higher aperture ratio can be achieved for higher LCD panel performance. The interface between the organic passivation layer and the back channel of the amorphous active region has been improved by the back channel oxygen treatment and the devices exhibits lower leakage current than the conventional silicon nitride passivation layer of BCE TFTs. The leakage currents between Indium-tin-oxide (ITO) pixels and the TFT devices and its mechanism have also been investigated in this paper.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2007.04a
• /
• pp.141-145
• /
• 2007

The part in the structure that is most affected by changes of external temperature is the protective concrete layer that protects a waterproofing layer. Also, the waterproofing layer that is situated under or on the back of such a protective concrete layer is affected by temperature and the behavior of the protective concrete layer under the condition of consolidation or close adhesion. In particular, in many cases, the damage is serious mainly around the projection (such as a parapet), crack, and joint (expansion joint). However, there is no proper way of examining again the non-exposed waterproofing layer once it has been constructed. Therefore, there is an assessment only on the physical property of materials and the capability of the layer in construction, and there is no actual assessment in consideration of its environmental condition or the condition of the use of buildings after construction. Therefore, in order to create more pleasant buildings and to enhance the durability of structures, this study conducts research into the change of capability of non-exposed waterproofing material after the application of a static load and moving load on the waterproofing layer situated under or on the back of protective concrete.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.2 no.1
• /
• pp.21-25
• /
• 2021

We measured the sizes of microstructures and the reflectance of blue feathers in the Common Kingfisher (Alcedo atthis). The colors were mainly produced in the barbs, which were composed of keratin sponge layers with air spaces and melanin rods. The reflectance spectra of back and tail feathers of the Common Kingfisher showed a peak with a broad plateau in the visible wavelength, whereas those of the wing feathers showed peaks in ultraviolet and visible and short-wavelengths. Moreover, the reflectance of back and tail feathers was higher than that of wing feathers. The blue color of the feathers comes from the keratin sponge layer due to coherent scattering. The back and tail feathers are composed of the keratin sponge layer only, and the wing feathers are composed of the keratin sponge layer and the keratin honeycomb structure. Due to the difference in these structures, it supposed that the reflectance is different. Determining why the reflectance spectra of the back and tail feathers were flattened will require further study.

• Economic and Environmental Geology
• /
• v.53 no.1
• /
• pp.99-110
• /
• 2020

It is a global trend to consider contaminated low-permeability zones as one of the primary management targets for the remediation of DNAPL contaminated sites. In addition, studies on the persistence caused by back diffusion of DNAPLs from low-permeability zones have been actively conducted worldwide. On the other hand, the studies for domestic groundwater contamination with the low-permeability zones are insufficient. Therefore, this study introduces the forward and back diffusions of DNAPL through low-permeability zones and suggests the importance of them by reviewing representative previous studies, especially on back diffusion and plume persistence. We proposed six diffusion scenarios and analytical solutions based on various boundary conditions of low-permeability zones. FI (forward diffusion into infinite domain) and BI (back diffusion form infinite domain) scenarios illustrate forward and back diffusion in which the depths of a low-permeability layer are assumed to be infinite. FFN (forward diffusion into finite domain with no flux boundary) and BFN (back diffusion from finite domain with no flux boundary) scenarios describe forward and back diffusion for a finite domain of a low-permeability layer with no flux boundary at the bottom. When the bottom of a low-permeability layer is considered as flux boundary, forward and back diffusion scenarios correspond to FFF (forward diffusion into finite domain with flux boundary) and BFF (back diffusion from finite domain with flux boundary). The scenarios and analytical solutions in this study may contribute to the determination of an efficient remediation method based on site characteristics such as a thickness of low-permeability zones or duration of contamination exposure.

• Journal of Navigation and Port Research
• /
• v.32 no.8
• /
• pp.589-596
• /
• 2008

In this paper, Artificial Neural Network(ANN) is applied to automatic berthing control for a ship. ANN is suitable for a maneuvering such as ship's berthing, because it can describe non-linearity of the system. Multi-layer perceptron which has more than one hidden layer between input layer and output layer is applied to ANN. Using a back-propagation algorithm with teaching data, we trained ANN to get a minimal error between output value and desired one. For the automatic berthing control of a containership, we introduced low speed maneuvering mathematical models. The berthing control with the structure of 8 input layer units in ANN is compared to 6 input layer units. From the simulation results, the berthing conditions are satisfied, even though the berthing paths are different.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.09a
• /
• pp.325-333
• /
• 1995

Back-propagation is the most widely used algorithm for supervised learning in multi-layer feed-forward networks. However, back-propagation is very slow in convergence. In this paper, a new weight initialization method, called rough map initialization, in multi-layer perceptrons is proposed. To overcome the long convergence time, possibly due to the random initialization of the weights of the existing multi-layer perceptrons, the rough map initialization method initialize weights by utilizing relationship of input-output features with singular value decomposition technique. The results of this initialization procedure are compared to random initialization procedure in encoder problems and xor problems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.11A
• /
• pp.1803-1811
• /
• 2001

In this paper, Forward Error Correction(FEC) that is satisfied with ITU-T Recommendation J.83, Annex B(North American Data Over Cable Service Interface Specifications(DOCSIS) for Multimedia Cable Network System(MCNS)) is analyzed. The FEC consist of Reed-Solomon(RS) layer, interleaving layer, randomization layer, and trellis coded modulation(TCM) layer. The effects of quantization of input symbol and of trace-back depth in the Viterbi decoder are simulated over AWGN channels.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.08a
• /
• pp.256-256
• /
• 2012

Oxide semiconductors such as zinc tin oxide (ZTO) or indium gallium zinc oxide (IGZO) have attracted a lot of research interest owing to their high potential for application as thin film transistors (TFTs) [1,2]. However, the instability of oxide TFTs remains as an obstacle to overcome for practical applications to electronic devices. Several studies have reported that the electrical characteristics of ZnO-based transistors are very sensitive to oxygen, hydrogen, and water [3,4,5]. To improve the reliability issue for the amorphous InGaZnO (a-IGZO) thin-film transistor, back channel passivation layer is essential for the long term bias stability. In this study, we investigated the instability of amorphous indium-gallium-zinc-oxide (IGZO) thin film transistors (TFTs) by the back channel contaminations. The effect of back channel contaminations (humidity or oxygen) on oxide transistor is of importance because it might affect the transistor performance. To remove this environmental condition, we performed vacuum seasoning before the deposition of hybrid passivation layer and acquired improved stability. It was found that vacuum seasoning can remove the back channel contamination if a-IGZO film. Therefore, to achieve highly stable oxide TFTs we suggest that adsorbed chemical gas molecules have to be eliminated from the back-channel prior to forming the passivation layers.