• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.133-145
• /
• 2022

Compared to the continuously increasing dog population and industry size in Korea, systematic analysis of related data and research on breed classification methods are very insufficient. In this paper, an automatic breed classification method is proposed using deep learning technology for 14 major dog breeds domestically raised. To do this, dog images are collected for deep learning training and a dataset is built, and a breed classification algorithm is created by performing transfer learning based on VGG-16 and Resnet-34 as backbone networks. In order to check the transfer learning effect of the two models on dog images, we compared the use of pre-trained weights and the experiment of updating the weights. When fine tuning was performed based on VGG-16 backbone network, in the final model, the accuracy of Top 1 was about 89% and that of Top 3 was about 94%, respectively. The domestic dog breed classification method and data construction proposed in this paper have the potential to be used for various application purposes, such as classification of abandoned and lost dog breeds in animal protection centers or utilization in pet-feed industry.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.23 no.3
• /
• pp.258-265
• /
• 2017

This paper validates a system identification method using mathematical optimization using sea trial measurement data as a benchmark. A fast time simulation tool, SIMOPT, and a Rheinmetall Defence mathematical model have been adopted to conduct initial hydrodynamic coefficient estimation and simulate ship modelling. Calibration for the environmental effect of sea trial measurement and sensitivity analysis have been carried out to enable a simple and efficient optimization process. The optimization process consists of three steps, and each step controls different coefficients according to the corresponding manoeuvre. Optimization result of Step 1, an optimization for coefficient on x-axis, was similar compared to values applying an empirical regression formulae by Clarke and Norrbin, which is used for SIMOPT. Results of Steps 2 and 3, which are for linear coefficients and nonlinear coefficients, respectively, was differ from the calculation results of the method by Clarke and Norrbin. A comparison for ship trajectory of simulation results from the benchmark and optimization results indicated that the suggested stepwise optimization method enables a coefficient tuning in a mathematical way.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.4
• /
• pp.1508-1517
• /
• 2015

Optimal Power dispatch is the short-term decision of the optimal output of a number of power generation facilities, to meet the system demand, with the objective of Power dispatching at the lowest possible cost, subject to transmission lines power loss and operational constraints. The operational constraint includes power balance constraint, generator limit constraint, and emission dispatch constraint and valve point effects. In this paper, Opposition based Differential Evolution Algorithm (ODEA) has been proposed to handle the objective function and the operational constraints simultaneously. Furthermore, the valve point loading effects and transmission lines power loss are also considered for the efficient and effective Power dispatch. The ODEA has unique features such as self tuning of its control parameters, self acceleration and migration for searching. As a result, it requires very minimum executions compared with other searching strategies. The effectiveness of the algorithm has been validated through four standard test cases and compared with previous studies. The proposed method out performs the previous methods.

• Journal of the Korean Society for Precision Engineering
• /
• v.29 no.4
• /
• pp.417-425
• /
• 2012

This paper presents reference ZMP trajectory generation and implementation for a biped robot via linear inverted dumbbell model (LIDM), which can consider the effect of external momentum on the center of mass (COM) of robot. Based on a reference ZMP trajectory derived by using LIDM, a base trajectory is proposed not only to make the locomotion of robot similar to that of human but also to facilitate its implementation and tuning. In order to realize a dynamic walking using the proposed trajectory, compliance, impedance and ZMP tracking controllers are adopted together. Extensive experiments show that the proposed locomotion of a biped robot is stable and also, similar to that of human. Further researches on balance recovery of a biped robot will be carried out to guarantee its robust locomotion in combination with the proposed trajectory.

• Structural Engineering and Mechanics
• /
• v.26 no.5
• /
• pp.517-544
• /
• 2007

This study shows a fuzzy tuning scheme to fuzzy sliding mode controller (FSMC) for seismic isolation of earthquake-excited structures. The sliding surface can rotate in the phase plane in such a direction that the seismic isolation can be improved. Since ideal sliding mode control requires very fast switch on the input, which can not be provided by real actuators, some modifications to the conventional sliding-mode controller have been proposed based on fuzzy logic. A superior control performance has been obtained with FSMC to deal with problems of uncertainty, imprecision and time delay. Furthermore, using the fuzzy moving sliding surface, the excellent system response is obtained if comparing with the conventional sliding mode controller (SMC), as well as reducing chattering effect. For simulation validation of the proposed seismic response control, 16-floor tall building has been considered. Simulations for six different seismic events, Elcentro (1940), Hyogoken (1995), Northridge (1994), Takochi-oki (1968), the east-west acceleration component of D$\ddot{u}$zce and Bolu records of 1999 D$\ddot{u}$zce-Bolu earthquake in Turkey, have been performed for assessing the effectiveness of the proposed control approach. Then, the simulations have been presented with figures and tables. As a result, the performance of the proposed controller has been quite remarkable, compared with that of conventional SMC.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2005.11a
• /
• pp.407-410
• /
• 2005

Acoustic coupling between combustion chamber and gas-liquid scheme injectors are studied numerically in liquid rocket engine adopting linear acoustic analysis. The injectors can play a role as half-wave resonators. The combustion chamber with numerous injectors shows peculiar acoustic coupling with the injectors. As the injector length approaches a half wavelength or the original tuning length, new injector-coupled acoustic modes show up in the chamber and thereby, the acoustic-damping effect of the tuned injectors is appreciably degraded.

• Wind and Structures
• /
• v.7 no.4
• /
• pp.251-264
• /
• 2004

The effects of a class of nonlinear Tuned Mass Dampers on the aeroelastic behavior of SDOF systems are investigated. Unlike classical linear TMDs, nonlinear constitutive laws of the internal damping acting between the primary oscillator and the TMD are considered, while the elastic properties are keept linear. The perturbative Multiple Scale Method is applied to derive a set of bifurcation equations in the amplitude and phase and a parametric analysis is performed to describe the postcritical scenario of the system. Both cubic- and van der Pol-type dampings are considered and the dependence of the limit-cycle amplitudes on the system parameters is studied. These new results, compared with the previously obtained bifurcation scenario of a SDOF aeroelastic oscillator equipped with a linear TMD, show a detrimental effect on the maximum limit-cycle amplitude reduction of the nonlinear TMD. However, the analyses evidence that in the parameter region away from the perfect tuning condition the nonlinear connection can be used to tune the system with an enhancement of the limit-cycle amplitude reduction.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2008.10a
• /
• pp.1097-1100
• /
• 2008

We demonstrate that the performance of amorphous indium-gallium-zinc-oxide (IGZO) thin-film transistors (TFT) can be optimized by controlling the interfaces between IGZO and sandwiching insulators and by proper deposition of IGZO layer. Specifically, contact and channel resistances are decreased by reducing IGZO bulk resistance and optimizing dry-etch process, respectively. Field-effect mobility ($\mu_{FE}$) and subthreshold gate swing (S) are further enhanced by fine-tuning IGZO deposition condition.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.22 no.5
• /
• pp.429-437
• /
• 2009

This study addresses the optimal shape of a LCVA maximizing its vibration control effect through numerical parametric study. Various LCVAs having the same total mass and tuning frequency are designed with constraints on the dimensions and water level, and one obtaining the highest equivalent damping ratio of the controlled system is chosen as an optimal solution. As a result, it was found that the limit on the variation of the water level in the vertical liquid column plays an important role constraining the shape of the LCVA. As the LCVA width perpendicular to the plane of liquid motion increases, the equivalent damping ratio rises with slowdown so that determination of the proper width is important in design of the LCVA shape.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 1999.08a
• /
• pp.412-422
• /
• 1999

Scale defects generated on the strip surface in a tandem finishing mill line are collected from the strip trapped among the production mills by freezing the growing scale on the strip by the melt glass coating and shutting down the line simultaneously. The samples observed of its cross sectional figure showed the process of scale defect formation where the defects are formed at the base metal surface by thicker oxidized scale during each rolling passes. The properties of the oxidized layer growth both at rolling and inter-rolling are detected down sized rolling test simulating carefully the rolling condition of the production line. The thickness of the oxidized layer at each rolling pass are simulated numerically. The critical scale thickness to avoid the defect formation is determined through the expression of mutual relation between oxidized layer thickness and the lanks of the strip called quality for the scale defects. The scale growth of scale less than the critical thickness and also to keep the bulk temperature tuning the water flow rate and cooling time appropriately. Two units of Inerstand Cooler are designed and settled among the first three stands in the production line. Two units of scale defect is counted from the recoiled strip and the results showed distinct decrease of the defects comparing to the conventionaly rolled products.