• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.591-593
• /
• 1995

Artificial neural network based pattern recognition method is one of the most probable candidate for on-line power system transient stability analysis. Especially, Kohonen layer is an adequate neural network for the purpose. Each node of Kehonen layer competes on the basis of which of them has its clustering center closest to an input vector. This paper discusses Kohonen's LVQ(Learning Victor Quantization) and points out a defection of the algorithm when applied to the transient stability analysis. Only the clustering centers located near the decision boundary of the stability region is needed for the stability criterion and the centers far from the decision boundary are redundant. This paper presents a new algorithm ratted boundary searching algorithm II which assigns only the points that are near the boundary in an input space to nodes or Kohonen layer as their clustering centers. This algorithm is demonstrated with satisfaction using 4-generator 6-bus sample power system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.6
• /
• pp.2806-2825
• /
• 2018

Unsupervised learning has shown good performance on image, video and audio classification tasks, and much progress has been made so far. It studies how systems can learn to represent particular input patterns in a way that reflects the statistical structure of the overall collection of input patterns. Many promising deep learning systems are commonly trained by the greedy layerwise unsupervised learning manner. The performance of these deep learning architectures benefits from the unsupervised learning ability to disentangling the abstractions and picking out the useful features. However, the existing unsupervised learning algorithms are often difficult to train partly because of the requirement of extensive hyperparameters. The tuning of these hyperparameters is a laborious task that requires expert knowledge, rules of thumb or extensive search. In this paper, we propose a simple and effective unsupervised feature learning algorithm for image classification, which exploits an explicit optimizing way for population and lifetime sparsity. Firstly, a sparse target matrix is built by the competitive rules. Then, the sparse features are optimized by means of minimizing the Euclidean norm ($L_2$) error between the sparse target and the competitive layer outputs. Finally, a classifier is trained using the obtained sparse features. Experimental results show that the proposed method achieves good performance for image classification, and provides discriminative features that generalize well.

• Journal of the Korean Geotechnical Society
• /
• v.24 no.10
• /
• pp.5-15
• /
• 2008

The purpose of this study is to examine the validity of recycled aggregates (RAs) as a substitute for Sand-Mat material for soft ground improvement in the housing site development. To evaluate the possibility of RAs as a substitute for sand mat material, first of all, the criteria and regulations related with the quality of lateral drain layer were collected and checked. Secondly, both of the properties of RAs were compared with the properties of natural sand for the lateral drain layer. The material properties related to coefficient of permeability, pressure at-rest state and so on satisfied most standards. On the basis of the test results, RAs were used to the construction site as lateral drain layer. Accordingly, if the quality of RAs can be managed well, the application of these RAs as lateral drain layer to replace natural sand was highly effective. Also, based on cost analysis of two materials, RAs are proved to be very competitive.

• Korean Journal of Environment and Ecology
• /
• v.19 no.3
• /
• pp.258-268
• /
• 2005

The objective of this study was to monitor the edge vegetation development process after timber harvesting at deciduous hardwood forest in Seoul National University Research Forest of Baekwoonsan (Mt), Korea The results from the 4th (2001) and 5th (2003) survey for edge vegetation development process after timber harvesting are summarized as follows. The competitive species at the harvested forest edge are Lindera erythrocarpa, Weigela subsessilis, Lespedeza maximowiczii, Lindera obtusiioba, Arajia eiata. There were Lindera erythrocarpa, Ivelgela subsessilis, Staphyiea bumaida as competitive species at harvested forest border between harvested forest edge and harvested interior. And at the forest interior, the competitive species were Lindera erythrocarpa, Lindera obtusiloba. There were differences in the competitive species at the forest edge due to the differences in direction, altitude, and existing dominant species. But Lindera erythrocarpa, Weigela subsessilis were appeared to be a competitive specieg at the southwestern slope (A site) and northeastern slope (B site). After timber harvesting, Sasa borealisshowed a good growth in early days, but showed a tendency of declining with the developing of canopy layer And coverage was generally higher at the Northeastern slope (Site B) than southwestern slop (Site A). The similarity index was lower as the distance from the forest edge becomes farther.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.34-36
• /
• 2022

In the existing text sentiment analysis models, the entire text is usually directly modeled as a whole, and the hierarchical relationship between text contents is less considered. However, in the practice of sentiment analysis, many texts are mixed with multiple emotions. If the semantic modeling of the whole is directly performed, it may increase the difficulty of the sentiment analysis model to judge the sentiment, making the model difficult to apply to the classification of mixed-sentiment sentences. Therefore, this paper proposes a sentiment analysis model BHGCN that considers the text hierarchy. In this model, the output of hidden states of each layer of BERT is used as a node, and a directed connection is made between the upper and lower layers to construct a graph network with a semantic hierarchy. The model not only pays attention to layer-by-layer semantics, but also pays attention to hierarchical relationships. Suitable for handling mixed sentiment classification tasks. The comparative experimental results show that the BHGCN model exhibits obvious competitive advantages.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.05a
• /
• pp.605-606
• /
• 2006

Today, the need for more flexible and adaptive production system and integrated management of their manufacturing information and facilities is ever increasing to cope with competitive and ever-changing global market environments and complexity of new control systems. This paper presents the whole system architecture and the technological characteristics of for each individual system layer which are able to flexibly integrate and manage high-speed and high-precision machining systems. It is investigated that monitoring and integrated management of the control systems can be realized with consideration of detailed information of various CNCs, and the management function may be easily constructed and extended using components of the manufacturing execution layer.

• Journal of Welding and Joining
• /
• v.20 no.6
• /
• pp.59-59
• /
• 2002

Sn-Ag-Cu solder is known as most competitive in many kinds of Pb-free solders. In this study, effects of solderability with plated layers such as Cu, Cu/Sn, Cu/Ni and Cu/Ni/Au were investigated. Sn-3.5Ag-0.7Cu solder balls were reflowed in commercial reflow machine (peak temp. : 250℃ and conveyer speed : 0.6m/min). In wetting test, immersion speed was 5mm/sec., immersion time 5sec., immersion depth 4mm and temperature of solder bath was 250℃. Wettability of Sn-3.5Ag-0.7Cu on Cu, Cu/Sn (5㎛), Cu/Ni (5㎛), and Cu/Ni/Au (5㎛/500Å) layers was investigated. Cu/Ni/Au layer had the best wettability as zero cross time and equilibrium force, and the measured values were 0.93 sec and 7mN, respectively. Surface tension of Sn-3.5Ag-0.7Cu solder turmed out to be 0.52N/m. The thickness of IMC is reduced in the order of Cu, Cu/Sn, Cu/Mi and Cu/Ni/Au coated layer. Shear strength of Cu/Ni, Cu/Sn and Cu was around 560gf but Cu/Ni/Au was 370gf.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.36 no.2
• /
• pp.88-95
• /
• 1994

This paper is to evaluate the applicability of neural networks to the estimation of evapotranspiration. Two neural networks were developed to forecast daily evapotranspiration of the rice crop with back-propagation and counter-propagation algorithm. The neural network trained by back-propagation algorithm with delta learning rule is a three-layer network with input, hidden, and output layers. The other network with counter-propagation algorithm is a four-layer network with input, normalizing, competitive, and output layers. Training neural networks was conducted using daily actual evapotranspiration of rice crop and daily climatic data such as mean temperature, sunshine hours, solar radiation, relative humidity, and pan evaporation. During the training, neural network parameters were calibrated. The trained networks were applied to a set of field data not used in the training. The created response of the back-propagation network was in good agreement with desired values and showed better performances than the counter-propagation network did. Evaluating the neural network performance indicates that the back-propagation neural network may be applied to the estimation of evapotranspiration of the rice crop. This study does not provide with a conclusive statement as to the ability of a neural network to evapotranspiration estimating. More detailed study is required for better understanding and evaluating the behavior of neural networks.

• Journal of Welding and Joining
• /
• v.20 no.6
• /
• pp.783-788
• /
• 2002

Sn-Ag-Cu solder is known as most competitive in many kinds of Pb-free solders. In this study, effects of solderability with plated layers such as Cu, Cu/Sn, Cu/Ni and Cu/Ni/Au were investigated. Sn-3.5Ag-0.7Cu solder balls were reflowed in commercial reflow machine (peak temp.:$250^{\circ}C$and conveyer speed:0.6m/min). In wetting test, immersion speed was 5mm/sec., immersion time 5sec., immersion depth 4mm and temperature of solder bath was $250^{\circ}C$. Wettability of Sn-3.5Ag-0.7Cu on Cu, Cu/Sn ($5\mu\textrm{m}$), Cu/Ni ($5\mu\textrm{m}$), and Cu/Ni/Au ($5\mu\textrm{m}/500{\AA}$) layers was investigated. Cu/Ni/Au layer had the best wettability as zero cross time and equilibrium force, and the measured values were 0.93 sec and 7mN, respectively. Surface tension of Sn-3.5Ag-0.7Cu solder turmed out to be 0.52N/m. The thickness of IMC is reduced in the order of Cu, Cu/Sn, Cu/Mi and Cu/Ni/Au coated layer. Shear strength of Cu/Ni, Cu/Sn and Cu was around 560gf but Cu/Ni/Au was 370gf.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.06a
• /
• pp.97-98
• /
• 2008

n-ZnO/p-Si heterostructure is a good candidate for ZnO-based heterojunction light emitting diodes(LED) because of its competitive price and lower driving voltage. However, the conventional LED shows much lower extraction efficiency, because it has small top contact and large backside contact. In this structure, the injected current from the top contact enters the active region underneath the top contact. Thus, the emitted light is hindered by the opaque top contact. This problem can be solved by using a current-blocking layer(CBL) that prevents the current injection into the active region below the top contact.