• Proceedings of the KAIS Fall Conference
• /
• 2007.05a
• /
• pp.35-38
• /
• 2007

본 논문에서는 생산 현장에서 사람이 하고 있는 자동차 부품의 전수(全數) 검사 방법을 대체하기 위한 머신 비전 시스템의 개발 연구이다. 세계 자동차 생산업체간 활발한 합병 및 전략적 제휴로 인한 완성차 업체수의 감소는 부품 업체 간 경쟁을 심화시키고 있다. 외국 업체에 비하여 상대적으로 규모 의 영세성을 보이고 있는 국내 부품 생산 업체는 대규모 생산 설비 투자와 함께 생산 효율성을 증대시켜 국제적인 경쟁력을 갖추도록 하여야 한다. 생산 효율과 품질 향상을 위한 노력의 일환으로 플라스틱 압출 성형에서 생기는 여러 가지 불량품 유형을 PC를 기반한 머신 비전 시스템 (Machine Vision System)을 구축하여 생산된 부품을 실시간 검사하고 불량품을 분류 하는 것이 본 연구의 목적이다. 머신 비전 시스템에 사용된 소프트웨어는 NI-LabVIEW를 사용하였으며, LabVIEW Vision 이미지 함수를 사용하여 검사 프로그램의 개발하였다. 부품 검사에 걸리는 시간은 50ms 내로 완료되기 때문에 생산 부품의 실시간 검사에 적용 될 수 있으며, 검사 영역과 설정 값을 비전 시스템 운용자가 설정할 수 있도록 프로그램이 만들어 졌다.

• Journal of Applied Reliability
• /
• v.18 no.3
• /
• pp.240-249
• /
• 2018

Purpose: The purpose of this research is to find which technique, between the PAUT (Phased array ultrasonic test) that has been used widely in practice and RT (Radiographic test) that was used widely in the past, has the higher reliability as a non-destructive testing of welding points in water wall tubes. Methods: To evaluated the reliability of non-destructive testing, eleven test pieces that were fabricated intentionally, which have the most frequently occurred defect types in water wall tubes and then both the PAUT and RT were performed on those eleven test pieces to compare their reliability. Results: The differences of type of defect, length are occurred due to the characteristics of nondestructive testing. The RT could not detect the lack of fusion defect type in specimen #4 and #8 while PAUT could not detect the lateral crack and 1 mm size small porosity in specimen #11. Conclusion: It is concluded that applying both the RT and PAUT result the best reliability rather than applying only one test method, if it is possible, in nondestructive testing of weld water wall tube in thermal power plant boiler case.

• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.225-234
• /
• 2019

Deep learning has become increasingly popular in both academic and industrial areas nowadays. Various domains including pattern recognition, Computer vision have witnessed the great power of deep neural networks. However, current studies on deep learning mainly focus on quality data sets with balanced class labels, while training on bad and imbalanced data set have been providing great challenges for classification tasks. We propose in this paper a method of data analysis-based data reduction techniques for selecting good and diversity data samples from a large dataset for a deep learning model. Furthermore, data sampling techniques could be applied to decrease the large size of raw data by retrieving its useful knowledge as representatives. Therefore, instead of dealing with large size of raw data, we can use some data reduction techniques to sample data without losing important information. We group PCB characters in classes and train deep learning on the ResNet56 v2 and SENet model in order to improve the classification performance of optical character recognition (OCR) character classifier.

• Journal of Welding and Joining
• /
• v.34 no.5
• /
• pp.13-18
• /
• 2016

Toshiba has developed a laser peening system for PWRs(pressurized water reactors) as well after the one for BWRs(boiling water reactors), and applied it for BMI(bottom-mounted instrumentation) nozzles, core deluge line nozzles and primary water inlet nozzles of Ikata Unit 1 and 2 of Shikoku Electric Power Company since 2004, which are Japanese operating PWR power plants. Laser pulses were delivered through twin optical fibers and irradiated on two portions in parallel to reduce operation time. For BMI nozzles, we developed a tiny irradiation head for small tubes and we peened the inner surface around J-groove welds after laser ultrasonic testing (LUT) as the remote inspection, and we peened the outer surface and the weld for Ikata Unit 2 supplementary. For core deluge line nozzles and primary water inlet nozzles, we peened the inner surface of the dissimilar metal welding, which is of nickel base alloy, joining a safe end and a low alloy metal nozzle. In this paper, the development and the actual application of the laser peening system for PWR power plants will be described.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.05a
• /
• pp.35-36
• /
• 2006

In Nd:YAG laser welding, evaluation methods of welding flaw are various. But, the method due to fume shape is difficult to classification of welding flaw. The Nd:YAG laser process is known to have high speed and deep penetration capability to become one of the most advanced welding technologies. At the present time, some methods are studied for measurement of fume shape by using high-speed camera and photo diode. This paper describes the machining characteristics of SM45C carbon steel welding by use of an Nd:YAG laser. In spite of its good mechanical characteristics, SM45C carbon steel has a high carbon contents and suffers a limitation in the industrial application due to the poor welding properties. In this study, fume shape was measured by infrared thermal camera that is non-contact/non-destructive thermal measurement equipment through change of laser generating power, speed, focus. Weld was performed on bead-on method. Measurement results are compared as two equipments. Here, two results are composed of measurement results of fume quantities due to fume shape by infrared thermal camera and inspection results of weld bead include weld flaws by ultrasonic inspector.

• Journal of Korean Society of Water and Wastewater
• /
• v.29 no.1
• /
• pp.97-106
• /
• 2015

A CCTV inspection method has been widely used to assess sewer condition and performance, but Korea lacks a proper decision support system for prioritizing sewer repair and rehabilitation (R&R). The objective of this paper is to introduce the results that we have developed in the Sewer Condition Assessment and Rehabilitation Decision-making (SCARD) Program using MS-EXCEL. The SCARD-Program is based on a standardized defect score for sewer structural and hydraulic assessment. Priorities are ranked based on risk scores, which are calculated by multiplying the sewer severity scores by the environmental impacts. This program is composed of three parts, which are decision-making for sewer condition and performance assessment, decision-making for sewer R&R priority assessment, and decision-making for optimal budget allocation. The SCARD-Program is useful for decision-makers, as it enables them to assess the sewer condition and to prioritize sewer R&R within the limited annual budget. In the future, this program logic will applied to the GIS-based sewer asset management system in local governments.

• Journal of the Korean Society of Industry Convergence
• /
• v.17 no.3
• /
• pp.136-144
• /
• 2014

This paper proposes a new approach to the designed of visual feedback control system based on visual servoing method. The main focus of this paper is presents how it is effective to use many features for improving the accuracy of the visual feedback control of industrial articulated robot for assembling and inspection of parts. Some rank conditions, which relate the image Jacobian, and the control performance are derived. It is also proven that the accuracy is improved by increasing the number of features. The effectiveness of redundant features is verified by the real time experiments on a SCARA type robot(FARA) made in samsung electronics company.

• Polymer(Korea)
• /
• v.38 no.5
• /
• pp.613-619
• /
• 2014

The polypropylene/polyamide elastomer (PP/PAE) blends were prepared by melt mixing. PP and PAE in PP/ PAE were immiscible completely. The size of PAE domains was large and the clear gap in the interface between PP and PAE existed, which did not meet the conditions enhancing toughness of polymers by elastomer. Therefore, maleic anhydride grafted polypropylene (MP) was used to improve the miscibility between PP and PAE. The miscibility between PP and PAE was improved and the size of dispersed phase PAE decreased by introducing MP. The crystallization of PP became easier by introducing PAE as a nucleating agent. With the increase of PAE content, the melt-crystallization temperatures of PP components in PP/PAE/MP blends increased gradually. The melt-crystallization of the polytetramethylene oxide segment of PAE component in PP/PAE blends were hampered by PP component. In addition, PAE can enhance significantly the toughness of PP, and the tensile strength and modulus did not decrease.

• Journal of the Korea Safety Management & Science
• /
• v.12 no.4
• /
• pp.255-263
• /
• 2010

The research developes short-run standardized control charts(SSCC) and short-run acceptance control charts(SACC) under the various demand patterns. The demand patterns considered in this paper are three types such as high-variety and repetitive low-volume pattern, extremely-high-variety and nonrepetitive low-volume pattern, and high-variety and extremely-low-volume pattern. The short-run standardized control charts developed by extending the long-run ${\bar{x}}$-R, ${\bar{x}}$-s and I-MR charts have strengths for practioners to understand and use easily. Moreover, the short-range acceptance control charts developed in the study can be efficiently used through combining the functions of the inspection and control chart. The weighting schemes such as Shewhart, moving average (MA) and exponentially weighted moving average (EWMA) can be considered by the reliability of data sets. The two types according to the use of control chart are presented in the short-range standardized charts and acceptance control charts. Finally, process capability index(PCI) and process performance index(PPI) classified by the demand patterns are presented.

• Advanced Composite Materials
• /
• v.16 no.2
• /
• pp.151-166
• /
• 2007

High-cycle fatigue characteristics of quasi-isotropic carbon fiber reinforced plastic (CFRP) laminates [-45/0/45/90]s up to $10^8$ cycles were investigated. To assess the fatigue behavior in the high-cycle region, fatigue tests were conducted at a frequency of 100 Hz, since it is difficult to investigate the fatigue characteristics in high-cycle at 5 Hz. Then, the damage behavior of the specimen was observed with a microscope, soft X-ray photography and a 3D ultrasonic inspection system. In this study, to evaluate quantitative characteristics of both transverse crack propagation and delamination growth in the high-cycle region, the energy release rate associated with damage growth in the width direction was calculated. Transverse crack propagation and delamination growth in the width direction were evaluated based on a modified Paris law approach. The results revealed that transverse crack propagation delayed under the test conditions of less than ${\sigma}_{max}/{\sigma}_b$ = 0.3 of the applied stress level.