• Journal of the Korean Society for Precision Engineering
• /
• v.22 no.7 s.172
• /
• pp.27-37
• /
• 2005

Direct writing technology on the silicon wafer surface is used to reduce the size of the chip as the miniature trend in electronic circuit. In order to improve the productivity and efficiency, the real time quality estimation is very important in each semiconductor process. In laser marking, marking quality is determined by readability which is dependant on the contrast of surface, the line width, and the melting depth. Many researchers have tried to find theoretical and numerical estimation models fur groove geometry. However, these models are limited to be applied to the real system. In this study, the estimation system for the line width during the laser marking was proposed by process monitoring method. The light intensity emitted by plasma which is produced when irradiating the laser to the silicon wafer was measured using the optical sensor. Because the laser marking is too fast to measure with external sensor, we build up the coaxial monitoring system. Analysis for the correlation between the acquired signals and the line width according to the change of laser power was carried out. Also, we developed the models enabling the estimation of line width of the laser marking through the statistical regression models and may see that their estimating performances were excellent.

• Journal of Sensor Science and Technology
• /
• v.25 no.1
• /
• pp.8-12
• /
• 2016

A planar, light-weight, and low-cost notch filter structure is required for the KSTAR ECEI (Electron Cyclotron Emission Imaging) system to protect the mixer arrays from spurious plasma heating power. Without protection, this heating power can significantly degrade or damage the performance of the mixer array. To protect mixer arrays, a frequency selective surface (FSS) structure is the suitable choice as a notch filter to reject the spurious heating power. The FSS notch filter should be located between the lenses of the ECEI system. This paper presents a 170 GHz FSS notch filter for the KSTAR ECEI sensor application. The design of such an FSS notch filter is based on the single-sided square loop geometry, because that makes it relatively insensitive to the incident angle of incoming wave. The FSS notch filter exhibits high notch rejection with low pass-band insertion loss over a wide range of incident angles. This paper also reviews the simulated and measured results. The proposed FSS notch filter might be implemented in other millimeter-wave plasma devices.

• Journal of Sensor Science and Technology
• /
• v.22 no.5
• /
• pp.364-368
• /
• 2013

Vibrational energy harvester based on piezoelectricity has been expected to be the dominant energy harvesting technology due to the advantages of high conversion efficiency, light weight and small size, night operation, etc. Its commercialization is just around the corner but the integration with power management electronics should be solved in advance. In this paper, therefore, fully-integrated design environment for piezoelectric energy harvesting systems is presented to assist co-design with the power management electronics. The proposed design environment is capable of analyzing the energy harvester including the package-induced damping effects and simulating the device and its power management electronics simultaneously. When the developed design environment was applied to the fabricated device, the simulated resonant frequency matched well with the experimental result with a difference of 2.97% only. Also, the complex transient response was completed in short simulation time of 3,001 seconds including the displacement distribution over the device geometry. Furthermore, a full-bridge power management circuit was modeled and simulated with the energy harvester simultaneously. Therefore the proposed, fully-integrated design environment is accurate and fast enough for the contribution on successful commercialization of piezoelectric energy harvester.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.7
• /
• pp.1326-1331
• /
• 2006

Weld root gap is a important fact of a falling-off weld quality in various kind of weld defect. The welding quality can be controlled by monitoring important parameters, such as, the Arc voltage, welding current and welding speed during the welding process. Welding systems use either a vision sensor or an Arc sensor, both of which are unable to control these parameters directly. Therefore, it is difficult to obtain necessary bead geometry without automatically controlling the welding parameters through the sensors. In this paper we propose a novel approach using neural networks for detecting and monitoring of weld root gap and bead shape. Through experiments we demonstrate that the proposed system can be used for real welding processes. The results demonstrate that the system can efficiently estimate the weld bead shape and detect the welding defects.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.3
• /
• pp.68-79
• /
• 2008

Ground vehicle borne system which is named RoSSAV(Road Safety Survey and Analysis Vehicle) developed in KICT(Korea Institute of Construction Technology) can collect road geometric data. This system therefore is able to evaluate the road safety and analyze road deficient sections using data collected along the roads. The purpose of this study is to extract road geometric data for 3D road modeling in dangerous road section and The system should be able to quickly provide more accurate data. Various sensors(circular laser scanner, GPS, INS, CCD camera and DMI) are installed in moving object and collect road environment data. Finally, We extract 3d road geometry(center, boundary), road facility and slope using integrated multi-sensor data.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.7A
• /
• pp.697-705
• /
• 2010

In Wireless Sensor Networks (WSNs), blanket (area) coverage analysis is generally carried to find the minimum number of active sensor nodes required to cover a monitoring interest area with the desired fractional coverage-threshold. Normally, the coverage analysis is performed using the stochastic geometry as a tool. The major component of such coverage analysis is the assumed sensing model. Hence, the accuracy of such analysis depends on the underlying assumption of the sensing model: how well the assumed sensing model characterizes the real sensing phenomenon. In this paper, we review the coverage analysis for different deterministic and probabilistic sensing models like Boolean and Shadow-fading model; and extend the analysis for Exponential and hybrid Boolean-Exponential model. From the analytical performance comparison, we demonstrate the redundancy (in terms of number of sensors) that could be resulted due to the coverage analysis based on the detection capability mal-characterizing sensing models.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1302-1308
• /
• 2019

This paper presents an approach for classifying image made by acquired position data from a 2D Lidar sensor with a convolutional neural network (CNN). Lidar sensor has been widely used for unmanned devices owing to advantages in term of data accuracy, robustness against geometry distortion and light variations. A CNN algorithm consists of one or more convolutional and pooling layers and has shown a satisfactory performance for image classification. In this paper, different types of CNN architectures based on training methods, Gradient Descent(GD) and Levenberg-arquardt(LM), are implemented. The LM method has two types based on the frequency of approximating Hessian matrix, one of the factors to update training parameters. Simulation results of the LM algorithms show better classification performance of the image data than that of the GD algorithm. In addition, the LM algorithm with more frequent Hessian matrix approximation shows a smaller error than the other type of LM algorithm.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.32 no.1
• /
• pp.47-55
• /
• 2012

A structural health monitoring technique for locating impact position in a plate structure is presented in this paper. The method employs a single sensor and spatial focusing of time reversal (TR) acoustics. We first examine the TR focusing effect at the impact position and its surroundings through simulation and experiment. The imaging results of impact points show that the impact source location can be accurately estimated in any position of the plate. Compared to existing techniques for locating impact or acoustic emission source, the proposed method has the benefits of using a single sensor and not requiring material properties and geometry of structures. Furthermore, it does not depend on a particular mode of dispersive Lamb waves that is frequently used in other ultrasonic testings of plate-like structures.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.12
• /
• pp.119-126
• /
• 2004

Permanent magnetic gears are magneto-mechanical devices that are widely used to replace the ordinary mechanical gear and to transmit torque without the mechanical contact. This study investigates the characteristics of touch free permanent magnetic gear according to the employing systems. The effect of the magnetic torque is analyzed by using 3 dimensional Finite Element Method (FEM). To estimate the transmission torque of FEM model, the numerical results are compared with the experimental results. The influences of geometry size, magnet number on transmission torque are obtained. As results of this paper, it is confirmed that the transmission torque behavior is associated with the configuration of the magnet numbers and the air gap between the two permanent magnetic gears.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.17 no.1
• /
• pp.98-105
• /
• 2008

In this paper, the system for monitoring process of aluminum laser welding was developed using the light signal emitted from the plasma which comes from interaction between material and laser. Photodiode for monitoring system was selected based on the spectrum analysis of light from plasma and keyhole. Behavior of plasma and keyhole was analyzed through the sensor signals. Value of sensor signal represented the light intensity and fluctuation of signal indicated the stability of plasma and keyhole. For the relation between welding condition and sensor signals, the input power and weld geometry greatly effected on the average of each sensor signals. Using the feature values of signals, estimation model for tensile strength of weld was formulated with neural network algorithm. Performance of this model was verified through coefficient of determination and average error rate.