• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.613-616
• /
• 2007

현재 국내 외 제조 산업은 기업 시스템의 노후화 등의 많은 문제들이 발생하고 있다. 현재의 자동차 부품공장에서의 불량품에 대한 처리방법은 제품의 생산이 완료된 후 테스트 단계를 거쳐 양품과 불량품을 분류하고 불량품이 발생하면 생산을 중단하고 생산라인의 상태를 점검하는 방식이다. 본 연구에서는 자동차 부품공장의 생산라인에서 불량품 생산을 줄이고 생산라인 가동시간의 지연을 줄이기 위한 불량 예방 모니터링 시스템에 대해 제안한다. 불량 예방 모니터링 시스템은 제품 조립의 각 단계 마다 테스트를 통해 데이터를 수집하고, 수집한 데이터에서 불량이 예상되면 알람 기능을 이용해서 경고를 하도록 설계하였다. 경고 메시지를 통해 불량이 예상되는 곳에 대해 조기에 조치하여 불량품이 나올 확률을 최소한으로 하고 제품의 생산지연 시간을 줄이는 것을 목표로 한다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.8 no.5
• /
• pp.1020-1025
• /
• 2007

EMS(energy management system) and SCADA(supervisory control and data acquisition) systems are used for reliable and efficient operation of electrical power systems. Various functions in EMS such as power flow, contingency analysis, security analysis essentially need accurate data set for reliable operation. State estimation can be a tool for providing these data. In this paper, programs for observability analysis and bad data processing are developed. Fundamental algorithms are introduced and validity of the proposed techniques is inspected with test cases.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1305-1307
• /
• 2022

자동차 부품의 결함은 시스템 전체의 성능 저하 및 인적 물적 손실이 발생할 수 있으므로 생산라인에서의 불량 검출은 매우 중요하다. 따라서 정확하고 균일한 결과의 불량 검출을 위해 딥러닝 기반의 고장 진단 시스템이 다양하게 연구되고 있다. 하지만 제조현장에서는 정상 샘플보다 비정상 샘플의 발생 빈도가 현저히 낮다. 이는 학습 데이터의 클래스 불균형 문제로 이어지게 되고, 이러한 불균형 문제는 고장을 판별하는 분류 모델의 성능에 영향을 끼치게 된다. 이에 본 연구에서는 모터의 동작음으로부터 불량 모터를 판별하는 불량 검출 시스템 설계를 위한 데이터 불균형 해결 방법을 제안한다. 자동차 사이드 미러 모터의 동작음을 학습 및 테스트를 위한 데이터 셋으로 사용하였으며 손실함수 계산 시 학습 데이터 셋의 클래스별 샘플 수 가 반영되는 label-distribution-aware margin(LDAM) loss 와 Inception, ResNet, DenseNet 신경망 모델의 비교 분석을 통해 불균형 데이터를 처리할 수 있는 가능성을 보여주었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.459-462
• /
• 2021

본 연구는 반도체 제조 공정중 발생하는 센서 데이터를 활용하여 딥러닝기반으로 불량을 예측하는 모델을 제안한다. 반도체 공장에서는 FDC((Fault Detection and Classification)라는 불량을 예측하는 시스템이 있지만, 공정의 복잡도가 높고 센서의 종류가 많아 공정 관리자가 모든 센서의 기준을 설정 및 관리하는데 한계가 있다. 이를 해결하기 위해 공정 설비의 센서 데이터를 딥러닝을 활용하여 학습시켜 센서 기준정보로 임계치를 제공하고, 가공중 발생하는 센서 데이터가 입력되면 정상 여부를 판정하는 모델을 제안한다.

• Journal of Internet of Things and Convergence
• /
• v.9 no.3
• /
• pp.81-86
• /
• 2023

SSD (solid state disk), which is flash memory-based storage device, has the advantages of high density and fast data processing. Therefore, it is being utilized as a storage device for high-capacity data storage systems that manage rapidly increasing big data. However, flash memory, a storage media, has a physical limitation that when the write/erase operation is repeated more than a certain number of times, the cells are worn out and can no longer be used. In this paper, we propose a method for converting defective multi-bit cells into single-bit cells to reduce the defect rate of flash memory and extend its lifetime. The proposed idea distinguishes the defects and treatment methods of multi-bit cells and single-bit cells, which have different physical characteristics but are treated as the same defect, and converts the expected defective multi-bit cells into single-bit cells to improve the defect rate and extend the overall lifetime. Finally, we demonstrate the effectiveness of our proposed idea by measuring the increased lifetime of SSD through simulations.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.201-204
• /
• 2007

본 논문은 소각재 재활용 모니터링 시스템 구현을 위한 에이전트 기반의 RFID 모델을 설계한다. RFID를 이용한 모니터링 시스템은 상태관리 에이전트, 위치관리 에이전트, 불량관리 에이전트, 상황관리 에이전트 등의 데이터 관리 에이전트를 이용해 데이터를 자동 관리하고, 대용량의 데이터를 처리하기 위해 대용량 데이터 처리 에이전트를 이용한다. 안정적인 소각재 재활용을 위한 에이전트 기반 데이터 모니터링 시스템의 개발은 산업체 전반에 걸쳐있는 기계화, 수작업화 된 공정을 실시간 자동화 공정으로 개발하는 획기적인 방법이 될 것이다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.5
• /
• pp.1053-1059
• /
• 2012

Barcode labels are being widely used for identifying products since they are cheap and easy to use. As the barcode labels are massively produced by seal printing, some labels have defects such as poor printing quality or data mismatch between barcode and the text. Barcode read errors and business errors caused by defected barcodes result in delay in logistics and increased processing costs. In this paper, we propose an on-line barcode verification system that uses image processing technique to verify the quality of labels at the production stage. The proposed system captures label images through the vision camera and then checks the print quality and verifies the combination of barcodes and texts in a label. If any defected label is found, the proposed system gives alarm signals and marks the defected labels so that they are removed at early stage of the production.

• Smart Media Journal
• /
• v.6 no.4
• /
• pp.58-64
• /
• 2017

Modern society is characterized by rapid increase in world population, aging of the rural population, decrease of cultivation area due to industrialization. The food problem is becoming an important issue with the farmers and becomes rural. Recently, the researches about the field of the smart farm are actively carried out to increase the profit of the rural area. The existing smart farm researches mainly monitor the cultivation environment of the crops in the greenhouse, another way like in the case of poor quality t is being studied that the system to control cultivation environmental factors is automatically activated to keep the cultivation environment of crops in optimum conditions. The researches focus on the crops cultivated indoors, and there are not many studies applied to the cultivation environment of crops grown outside. In this paper, we propose a method to improve the harvestability of poor areas by monitoring the areas with bad harvests by using big data analysis, by precisely predicting the harvest timing of fruit trees growing in orchards. Factors besides for harvesting include fruit color information and fruit weight information We suggest that a harvest correlation factor data collected in real time. It is analyzed using the Apache Spark engine. The Apache Spark engine has excellent performance in real-time data analysis as well as high capacity batch data analysis. User device receiving service supports PC user and smartphone users. A sensing data receiving device purpose Arduino, because it requires only simple processing to receive a sensed data and transmit it to the server. It regulates a harvest time of fruit which produces a good quality fruit, it is needful to determine a poor harvest area or concentrate a bad area. In this paper, we also present an architectural model to determine the bad areas of fruit harvest using strong data analysis.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.2
• /
• pp.65-70
• /
• 2015

In the PCB(Printed Circuit Board) manufacturing industry, the yield is an important management factor because it affects the product cost and quality significantly. In real situation, it is very hard to ensure a high yield in a manufacturing shop because products called chips are made through hundreds of nano-scale manufacturing processes. Therefore, in order to improve the yield, it is necessary to analyze main fault process and equipment that cause low PCB yield. This paper proposes a systematic approach to discover fault-causing processes and equipment by using a logistic regression and a stepwise variable selection procedure. We tested our approach with lot trace records of real work-site. A lot trace record consists of the equipment sequence that the lot passed through and the number of faults for each fault type in the lot. We demonstrated that the test results reflected the real situation of a PCB manufacturing line.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.149-155
• /
• 2021

In this study, we intend to develop a defective road surface object recognition model that automatically detects road surface defects that restrict the movement of the transportation handicapped using electric mobile devices with deep learning. For this purpose, road surface information was collected from the pedestrian and running routes where the electric mobility aid device is expected to move in five areas within the city of Busan. For data, images were collected by dividing the road surface and surroundings into objects constituting the surroundings. A series of recognition items such as the detection of breakage levels of sidewalk blocks were defined by classifying according to the degree of impeding the movement of the transportation handicapped in traffic from the collected data. A road surface object recognition deep learning model was implemented. In the final stage of the study, the performance verification process of a deep learning model that automatically detects defective road surface objects through model learning and validation after processing, refining, and annotation of image data separated and collected in units of objects through actual driving. proceeded.