• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.13-15
• /
• 2006

When we use multilayer perceptron model for pattern classification probmems, we allocate one output node for each class. In this paper, we increase the number of output nodes for each class and investigate the performance of multilayer perceptrons through the simulation of isolated-word recognition problems.

• Journal of the Korean Society of Industry Convergence
• /
• v.20 no.4
• /
• pp.275-284
• /
• 2017

As the high-level of the industrial and information age, the electricity become the indispensable element in the daily life including OA, FA, and computer, electric home appliances, and etc. In particular, The continuous use of the high capacity power supply system by applying a Switching Mode Power Supply(SMPS) according to the increase of the secondary side output terminal of the power load of the refrigerator of the home appliance or automation of the plant is pressed. The purpose using the way with this kind of high-capacity altogether is to supply the output voltage and output current regardless of the input voltage or to the external environmental conditions of the secondary-side load fluctuation. In this paper, a combination of a Buck Converter with Boost Converter by making a constant current source to control the inductor current and maintain stable power supply side operating characteristics, when load variations. While maintaining the same characteristics as conventional Buck Converter, and offer a DC-DC Converter system with the new switch pattern having a wide output range capable of operating in Buck-Boost Converter. In addition, after theoretical analysis, we carry out simulations and experiments to verify the validity and performance comparing with a conventional DC-to-DC converter.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.4
• /
• pp.534-540
• /
• 2014

In this paper, we propose a fuzzy combined Polynomial Neural Network(PNN) for pattern classification. The fuzzy combined PNN comes from the generic TSK fuzzy model with several linear polynomial as the consequent part and is the expanded version of the fuzzy model. The proposed pattern classifier has the polynomial neural networks as the consequent part, instead of the general linear polynomial. PNNs are implemented by stacking the simple polynomials dynamically. To implement one layer of PNNs, the various types of simple polynomials are used so that PNNs have flexibility and versatility. Although the structural complexity of the implemented PNNs is high, the PNNs become a high order-multi input polynomial finally. To estimate the coefficients of a polynomial neuron, The weighted linear discriminant analysis. The output of fuzzy rule system with PNNs as the consequent part is the linear combination of the output of several PNNs. To evaluate the classification ability of the proposed pattern classifier, we make some experiments with several machine learning data sets.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.1
• /
• pp.86-93
• /
• 2006

The TAM(Topographic Attentive Mapping) network based on a biologically-motivated neural network model is an especially effective one for pattern analysis. It is composed of of input layer, category layer, and output layer. Fuzzy rule, for input and output data are acquired from it. The TAM network with three pruning rules for reducing links and nodes at the layer is called fuzzy TAM network. In this paper, we apply fuzzy TAM network to pattern analysis of core competency model for subcontractors of construction companies and show its usefulness.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.58 no.4
• /
• pp.602-608
• /
• 2009

In this thesis, output voltage, current and power of solar module were classified by irradiation from data of overall operating characteristics collected for one year in order to manage efficient photovoltaic generation system and deliver maximum power. In addition, from these data, correlations between irradiation of photovoltaic cell and amount of power given by photovoltaic cell was quantitatively examined to deduce optimization of the design and construction of photovoltaic generation system. As I-V characteristics according to a temperature range of 10~50[$^{\circ}C$], the area of I-V characteristics were increased with an increase in temperature. Since this area corresponds to the power, output power is thought to have increased with temperature. As output power characteristics according to a temperature range of 10~50[$^{\circ}C$], output power was increased with an increase in temperature. Since output power increases with temperature increase, the result corresponds well to the related equation on temperature and output power. As I-V characteristics according to a irradiation range of 100~900 [$W/m^2$], voltage and current were increased with an increase in irradiation. The result is thought of as an increase in output power with increasing irradiation. As output power characteristics according to a irradiation range of 100~900 [$W/m^2$], output power was increased with increasing irradiation. This result corresponds well to the related equation on irradiation and output power.

• ETRI Journal
• /
• v.25 no.2
• /
• pp.101-108
• /
• 2003

This paper proposes an improved vascular pattern extraction algorithm for person verification applications. The proposed direction-based vascular pattern extraction (DBVPE) algorithm is based on the directional information of vascular patterns. It applies two different filters to the input images: row vascular pattern extraction filter (RVPEF) for effective extraction of the abscissa vascular patterns and column vascular pattern extraction filter (CVPEF) for effective extraction of the ordinate vascular patterns. We use the combined output of both filters to obtain the final hand vascular patterns. Unlike the conventional hand vascular pattern extraction algorithm, the directional extraction approach prevents loss of the vascular pattern connectivity. To validate the DBVPE algorithm, we used a prototype system with a DSP processor. The prototype system shows approximately a three-times better false acceptance rate (FAR) than the conventional single filter algorithm.

• Journal of Energy Engineering
• /
• v.17 no.3
• /
• pp.145-152
• /
• 2008

Approaching to the era of high energy price and energy sources scarcity, the demand for governmental intervention to mitigate the short-term shocks is highly increasing. When any energy policy is implemented, double-side effects would be derived. To begin with positive aspect, by decreasing energy import, unnecessary currency outflow can be prevented and the resultant saved money will be appropriately allocated. Furthermore, industrial competitiveness will be assured by reducing use of expensive energy. On the contrary, inappropriate energy saving policy may lead to unexpected negative effects that would hinder improvement in productivity due to indiscreet replacing energy by equipments. In order to enhance effectiveness of energy policy, efforts should be made in advance to understand the energy use pattern of each industry sector which composes the economy. Therefore, in this study, an energy input-output method, one of the macroscopic approaches, is applied to analyze energy use patterns of each industry sector in Korea. Using this method, a quantitative assessment is performed to obtain the energy use intensity and the amount of energy uses with respect to energy types.

• East Asian Journal of Business Economics (EAJBE)
• /
• v.7 no.4
• /
• pp.51-58
• /
• 2019

Purpose - This paper aims to analyze the difference of technological innovative pattern by industry. Research design and methodology - we try to identify the major factors which can exert an effective influence on actual innovation output, utilizing the result of Korean Innovation Survey. By doing so, this work can make a comparison with Pavitt (1984) and succeeding discussion on sectoral pattern of innovation Results - Analysis on major industry in Korea shows that there are substantial differences in terms of the source of innovation, organization-related factor, and appropriation mechanism among each industry, and differential strategy to be proper for the nature of each industry is needed. There is some variation within industries which deemed as same type of sector defined by Pavitt. Conclusions - This analysis call for elaborate analysis on sectoral pattern of innovation, considering the change and difference of innovative environment as well as differential business strategy and way to do innovate, which is proper considering the nature of innovative pattern in each industry for successful technological innovation in Korea. At the same time, proper policy measure considering the differential pattern of technological innovation is needed.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.1
• /
• pp.1-14
• /
• 2004

This paper surveys some existing direct adaptive feedback control schemes for linear time-invariant systems with actuator failures characterized by the failure pattern that some inputs are stuck at some unknown fixed or varying values at unknown time instants, and applications of those schemes to aircraft flight control system models. Controller structures, plant-model matching conditions, and adaptive laws to update controller parameters are investigated for the following cases for continuous-time systems: state tracking using state feed-back, output tracking using state feedback, and output tracking using output feedback. In addition, a discrete-time output tracking design using output feedback is presented. Robustness of this design with respect to unmodeled dynamics and disturbances is addressed using a modified robust adaptive law.

• The Journal of the Korea Contents Association
• /
• v.9 no.1
• /
• pp.123-130
• /
• 2009

When we apply MLPs(multilayer perceptrons) to pattern classification problems, we generally allocate one output node for each class and the index of output node denotes a class. On the contrary, in this paper, we propose to increase the number of output nodes per each class for performance improvement of MLPs. For theoretical backgrounds, we derive the misclassification probability in two class problems with additional outputs under the assumption that the two classes have equal probability and outputs are uniformly distributed in each class. Also, simulations of 50 isolated-word recognition show the effectiveness of our method.