• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.05a
• /
• pp.389-398
• /
• 2007

Case-based reasoning (CBR) is one of the most popular prediction techniques for medical diagnosis because it is easy to apply, has no possibility of overfitting, and provides a good explanation for the output. However, it has a critical limitation - its prediction performance is generally lower than other artificial intelligence techniques like artificial neural networks (ANNs). In order to obtain accurate results from CBR, effective retrieval and matching of useful prior cases for the problem is essential, but it is still a controversial issue to design a good matching and retrieval mechanism for CBR systems. In this study, we propose a novel approach to enhance the prediction performance of CBR. Our suggestion is the simultaneous optimization of feature weights, instance selection, and the number of neighbors that combine using genetic algorithms (GAs). Our model improves the prediction performance in three ways - (1) measuring similarity between cases more accurately by considering relative importance of each feature, (2) eliminating redundant or erroneous reference cases, and (3) combining several similar cases represent significant patterns. To validate the usefulness of our model, this study applied it to a real-world case for evaluating cytological features derived directly from a digital scan of breast fine needle aspirate (FNA) slides. Experimental results showed that the prediction accuracy of conventional CBR may be improved significantly by using our model. We also found that our proposed model outperformed all the other optimized models for CBR using GA.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.23 no.8
• /
• pp.57-66
• /
• 2009

The conventional fixed gain PI controller is very sensitive to step change of command speed, parameter variation and load disturbances. The precise speed control of interior permanent magnet synchronous motor(IPMSM) drive becomes a complex issue due to nonlinear coupling among its winding currents and the rotor speed as well as the nonlinear electromagnetic developed torque. Therefore, there exists a need to tune the PI controller parameters on-line to ensure optimum drive performance over a wide range of operating conditions. This paper proposes hybrid intelligent-PI(HIPI) controller of IPMSM drive using adaptive learning mechanism(ALM) and fuzzy neural network(FNN). The proposed controller is developed to ensure accurate speed control of IPMSM drive under system disturbances and estimation of speed using artificial neural network(ANN) controller. The PI controller parameters are optimized by ALM-FNN at all possible operating condition in a closed loop vector control scheme, The validity of the proposed controller is verified by results at different dynamic operating conditions.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.6
• /
• pp.263-268
• /
• 2017

A hot forming process is required for Mg alloys to enhance the formability and plastic workability due to the insufficient formability at room temperature. Mg alloy undergoes dynamic recrystallization (DRX) during the hot working process, which is a restoration or softening mechanism that reduces the dislocation density and releases the accumulated energy to facilitate plastic deformation. The flow stress curve shows three stages of complicated strain hardening and softening phenomena. As the strain increases, the stress also increases due to work hardening, and it abruptly decreases work softening by dynamic recrystallization. It then maintains a steady-state region due to the equilibrium between the work hardening and softening. In this paper, an efficient optimization process is proposed for the material model of the dynamic recrystallization to improve the accuracy of the flow curve. A total of 18 variables of the constitutive equation of AZ80 alloy were systematically optimized at an elevated forming temperature($300^{\circ}C$) with various strain rates(0.001, 0.1, 1, 10/sec). The proposed method was validated by applying it to the constitutive equation of AZ61 alloy.

• Journal of Powder Materials
• /
• v.22 no.5
• /
• pp.350-355
• /
• 2015

Perforated polygonal cobalt oxide ($Co_3O_4$) is synthesized using electrospinning and a hydrothermal method followed by the removal of a carbon nanofiber (CNF) template. To investigate their formation mechanism, thermogravimetric analysis, field-emission scanning electron microscopy, transmission electron microscopy, X-ray diffraction, and X-ray photoelectron spectroscopy are examined. To obtain the optimum condition of perforated polygonal $Co_3O_4$, we prepare three different weight ratios of the Co precursor and the CNF template: sample A (Co precursor:CNF template- 10:1), sample B (Co precursor:CNF template-3.2:1), and sample C (Co precursor:CNF template-2:1). Among them, sample A exhibits the perforated polygonal $Co_3O_4$ with a thin carbon layer (5.7-6.2 nm) owing to the removal of CNF template. However, sample B and sample C synthesized perforated round $Co_3O_4$ and destroyed $Co_3O_4$ powders, respectively, due to a decreased amount of Co precursor. The increased amount of the CNF template prevents the formation of polygonal $Co_3O_4$. For sample A, the optimized weight ratio of the Co precursor and CNF template may be related to the successful formation of perforated polygonal $Co_3O_4$. Thus, perforated polygonal $Co_3O_4$ can be applied to electrode materials of energy storage devices such as lithium ion batteries, supercapacitors, and fuel cells.

• Journal of Powder Materials
• /
• v.21 no.2
• /
• pp.108-113
• /
• 2014

Flake-like $LiCoO_2$ nanopowders were fabricated using electrospinning. To investigate their formation mechanism, field-emssion scanning electron microscopy, X-ray diffraction, and X-ray photoelectron spectroscopy were carried out. Among various parameters of electrospinning, we controlled the molar concentration of the precursor and the PVP polymer. When the molar concentration of lithium and cobalt was 0.45 M, the morphology of $LiCoO_2$ nanopowders was irregular and round. For 1.27 M molar concentration, the $LiCoO_2$ nanopowders formed with flake-like morphology. For the PVP polymer, the molar concentration was set to 0.011 mM, 0.026 mM, and 0.043 mM. Irregular $LiCoO_2$ nanopowders were formed at low concentration (0.011 mM), while flake-like $LiCoO_2$ were formed at high concentration (0.026 mM and 0.043 mM). Thus, optimized molar concentration of the precursor and the PVP polymer may be related to the successful formation of flake-like $LiCoO_2$ nanopowders. As a results, the synthesized $LiCoO_2$ nanopowder can be used as the electrode material of Li-ion batteries.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.3
• /
• pp.2178-2184
• /
• 2015

With the development of wireless network technologies, mobile users may use various networks easily and expect more advanced services. On the other hand, it may bring on some problems with network resource management that should lead the service provider to improve the current service quality and manage the network resource efficiently. This paper proposes the optimized radio resource management (RRM) scheme that integrates the Grey Relational Analysis (GRA) and game theory. The first applies the GRA to determine the Grey Relation Coefficient (GRC) factors that represent the network preference, and the network provider then selects the requested service that provide maximum payoff through Nash Equilibrium. Six requested services that have one application service among four different types of service classes were considered and the game was played repeatedly. In WiMAX, WLAN 1 and WLAN 2 game, the maximum payoff of each players was 93, 90.6 and 92.8 respectively. The experimental results show that every requested service can be selected by the network provider. Consequently, the proposed radio resource management mechanism is more effective in heterogeneous wireless networks.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.8 no.4
• /
• pp.283-289
• /
• 2014

This paper presents a design and optimization of the fully-active transfemoral prosthesis leg system. As it has one degree of freedom in knee joint, this prosthesis leg can imitate the human's gait. The weight of system, which makes the users more comfortable due to less tiredness, and the knee joint torque to rise stability of the system are major factors of prosthesis leg system. Thus the mechanism of prosthesis changes from 3-linkage type to geared type. The sensorized foot is also designed to effectively determine human's gait by measuring deformation of the foot during gait. Topology optimization is carried out for the sensorized foot to remove its unnecessary weight. The safety of optimized foot is verified by carrying out finite element analysis.

• Polymer(Korea)
• /
• v.34 no.1
• /
• pp.69-73
• /
• 2010

Effects of reaction time and temperature on the isomerization and dehydrobromination reactions of brominated butyl rubber were investigated. The structural composition of brominated butyl rubber was determined by Fourier transform infrared spectroscopy (FT-IR) and proton nuclear magnetic resonance spectroscopy($^1H$-NMR), Density functional theory (DFT) was used to study on the isomerization and dehydrobromination mechanisms of model compounds. The geometries for model compounds of 3-bromo-5,5,7,7-tetramethyl-2(2',2',4',4'-tetramethyl)pentyl-1-octylene (3BrOE), 1-bromo-5,5,7,7-tetramethyl-2(2',2',4',4'-tetramethyl)pentyl-2-octylene (1Br2OE) and 5,5,7,7-tetramethyl-2(2',2', 4',4'-tetramethyl)pentyl-1,3-octadiene (CD) had been optimized by using density functional theory at B3LYP/3-21G and B3LYP/6-31G levels. The predicted energy of 3BrOE lies higher than that of 1Br2OE which suggests that 1Br2OE configuration is more stable than the 3BrOE configuration. Compared with the energy barrier, the pathway of dehydrobromination is less competitive than that of isomerization. This is qualitatively consistent with the experimental results.

• The Journal of Society for e-Business Studies
• /
• v.9 no.3
• /
• pp.57-77
• /
• 2004

The main purpose of this study is to develop XML schemata and related STEP model for sharing construction drawings information. To achieve this purpose, the authors have developed a drawing information model based on STEP/AP202, the data searching mechanism based on STEP, and XML schemata for sharing and exchanging information between vector data and non-shape attribute information. Finally, the authors have suggested the way of sharing drawing information through linked STEP data and a XML schema using test cases of construction material information and code checking. The study shows a way of optimized managing and sharing construction information through the drawing information and external data for the whole building life-cycle, from early design stage to the construction stage.

• Journal of the Microelectronics and Packaging Society
• /
• v.22 no.1
• /
• pp.51-54
• /
• 2015

An electro-plating technology on a cured isotropic conductive pattern with a hybrid Cu paste composed of resin matrix, copper, and solder powders has been developed. In a conventional technology, Ag paste was used to perform a conductive pattern on a PCB or silicon substrate. From previous research, the electrical conductive mechanism and principle of the hybrid Cu paste were concisely investigated. The isotropic conductive pattern on the PCB substrate was performed using screen-printing technology. The optimum electro-plating condition was experimentally determined by processing parameters such as the metal content of the hybrid Cu paste, applied current density, and time for the electroplating in the plating bath. The surfaces and cross-sections were observed using optical and SEM photographs. In conclusion, the optimized processing conditions for Cu electro-plating technology on the conductive pattern were a current density of $40mA/cm^2$ and a plating time of 20min on the hybrid Cu paste with a metal content of 44 vol.%. More details of the mechanical properties and processing conditions will be investigated in further research.