• Journal of the Institute of Convergence Signal Processing
• /
• v.21 no.3
• /
• pp.148-153
• /
• 2020

In order to obtain high accuracy with an convolutional neural network(CNN), it is necessary to set the optimal hyperparameters. However, the exact value of the hyperparameter that can make high performance is not known, and the optimal hyperparameter value is different based on the type of the dataset, therefore, it is necessary to find it through various experiments. In addition, since the range of hyperparameter values is wide and the number of combinations is large, it is necessary to find the optimal values of the hyperparameters after the experimental design in order to save time and computational costs. In this paper, we suggest an algorithm that use the design of experiments and grid search algorithm to determine the optimal hyperparameters for a classification problem. This algorithm determines the optima values of the hyperparameters that yields high performance using the factorial design of experiments. It is shown that the amount of computational time can be efficiently reduced and the accuracy can be improved by performing a grid search after reducing the search range of each hyperparameter through the experimental design. Moreover, Based on the experimental results, it was shown that the learning rate is the only hyperparameter that has the greatest effect on the performance of the model.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.11
• /
• pp.514-520
• /
• 2002

n industrial computer vision systems, the image quality is dependent on the parameters such as light source, illumination method, optics, and surface properties. Most of them are related with the lighting system, which is designed in heuristic based on the designer's experimental knowledge. In this paper, a design code by which the optimal lighting method and light source for computer vision systems can be found are suggested based on experimental results. The design coed is applied to the design of the lighting system for the transistor marking inspection system, and the overall performance of the machine vision system with the lighting system show the effectiveness of the proposed design code.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2000.11a
• /
• pp.17-24
• /
• 2000

Hardening agent has been the traditional material for surface soil stabilization of soft ground. This study aims at determining optimal mixture ratio of hardening agent in accordance with the required design specifications. Hardening agent is properly mixtured with Fly ash, Gypsum, Slag and Cement for the ettringite hydrates which is effective for early stabilization of unconsolidated soil. The treated soil is the clay which are widely found here and there in Korea. In this study, preliminary tests were performed to get optimal mixture ratio of stabilizer ingredient, and marine clay in Jin-Hae was used to get physical and chemical properties. Laboratory tests of 50 stabilized soil were peformed to get optimal mixture ratio for 16 stabilizer material of 6 type, and stabilizer mixing was determined.

• International journal of advanced smart convergence
• /
• v.5 no.4
• /
• pp.66-70
• /
• 2016

When tooth grinding occurs, electrical stimulation is given at the same time, and tooth grinding stops on such stimulation. Electromyography signals are used as control signals of electrical stimulation to disturb tooth grinding. However because of the electrical stimulation, the M-waves are generated and mixed with spontaneous electromyogram. In this study, we designed an optimal filter to remove M-wave and conserve spontaneous electromyogram simultaneously. The inverse power method (IPM) showed that the optimal filter coefficient is the eigenvector corresponding to the minimum eigenvalue of the input covariance matrix. In order to evaluate the performance of the optimal filter, we compared using a conventional band pass filter and adaptive filter using least mean square algorithm. The experimental results show that the optimal filter can effectively remove the M-wave compared to the previously studied prediction error filter.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.17 no.5
• /
• pp.44-51
• /
• 2008

In this paper, the optimal tuning of a cross-coupled controller linked with the feedforward controller is studied to reduce contouring and tracking errors of a bi-axial servomechanisms by using the previously developed integrated tuning method. The CCC system for an arbitrary curve, which is combined with the feedforward controller, is formulated by a state-space based on a series of linear motion trajectories. An optimal tuning problem is formulated as a nonlinear constrained optimization problem including relevant controller parameters of the servo. To verify the effectiveness of the proposed optimal tuning procedure, linear and circular motion experiments are performed on the xy-table. Experimental results confirm that both tracking and contouring errors are significantly reduced by applying the proposed control and tuning system.

• Electrical & Electronic Materials
• /
• v.9 no.8
• /
• pp.799-805
• /
• 1996

In short channel MOSFET, it is very important to establish optimal process conditions because of variation of device characteristics due to the process parameters. In this paper, we used process simulator and device characteristics caused by process parameter variation. From this simulation, it has been ' derived to the dependence relations between process parameters and device characteristics. The experimental result of fabricated short channel device according to the optimal process parameters demonstrate good device characteristics.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.840-843
• /
• 2003

The Paper presents theoretical and experimental investigation of plunger type water hydraulic pump (plunger pump). An analysis of Crank shaft for plunger pump is carried out. With this results, the optimal dimensions of plunger pump are determined. And performance test results about slide resistance, hydraulic leak test, efficiency-pressure. and flow-speed and flow-pressure are presented.

• Structural Engineering and Mechanics
• /
• v.19 no.6
• /
• pp.721-747
• /
• 2005

The strong need of verifying theories formulated for semi-active control through applications to real structures is due to the fact that theoretical research on semi-active control systems is not matched by a corresponding satisfactory experimental activity. This paper shows how a smart system including magnetorheological devices as damping elements can be implemented in a large-scale structural model, by describing in detail the kind of electronics (dedicated hardware and software) adopted during the experimental campaign. It also describes the most interesting results in terms of reduction of the seismic response (either experimental or numerical) of the semi-actively controlled structure compared to a passive operating control system, and in terms of the evaluation criteria proposed in the benchmark for seismically excited controlled buildings. The paper also explains how to derive from the classical theory of optimal control the adopted control logic, based on a clear physical approach, and provides an exhaustive picture of the time delays characterizing the control sequence.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.19 no.5
• /
• pp.92-103
• /
• 2015

Box-Wilson experimental design method, known as central composite design, is the design of any information-gathering exercises where variation is present. This method was devised to gather as much data as possible in spite of the low design cost. This method was employed to model the effect of mixing factors on several performances of 60 MPa high strength self compacting concrete and to numerically calculate the optimal mix proportion. The nonlinear relations between factors and responses of HSSCC were approximated in the form of second order polynomial equation. In order to characterize five performances like compressive strength, passing ability, segregation resistance, manufacturing cost and density depending on five factors like water-binder ratio, cement content, fine aggregate percentage, fly ash content and superplasticizer content, the experiments were made at the total 52 experimental points composed of 32 factorial points, 10 axial points and 10 center points. The study results showed that Box-Wilson experimental design was really effective in designing the experiments and analyzing the relation between factor and response.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.27 no.8
• /
• pp.1042-1050
• /
• 2003

A heating flow discharged from a 4-way ceiling type indoor unit has been investigated to determine the design parameters for the optimal flow control. The flow was measured by a PIV(particle image velocimetry) system and an experimental model of 1/10 scale with a transparent room was devised by satisfying the Archimedes number. This similarity is generally used in cases where the forced convection has similar magnitude of the natural convection. To optimize the heating flow, several vane angles and vane control algorithms of cross and right angle controls were considered. Regarding the vane angle, experimental results show that 30$^{\circ}$is an optimal angle to avoid re-suction flows without significant increase in flow noise. Temperature distribution measured in the environmental chamber ensures the increased thermal comfort when compared to the case, 60$^{\circ}$angle. At the optimal angle, applying open/close control gives rise to more uniform distribution of the heating flow than without control. Especially, the cross-control seems to be satisfactory for thermal comfort.