• 산업경영시스템학회지
• /
• 제45권4호
• /
• pp.157-166
• /
• 2022

In the case of a die-casting process, defects that are difficult to confirm by visual inspection, such as shrinkage bubbles, may occur due to an error in maintaining a vacuum state. Since these casting defects are discovered during post-processing operations such as heat treatment or finishing work, they cannot be taken in advance at the casting time, which can cause a large number of defects. In this study, we propose an approach that can predict the occurrence of casting defects by defect type using machine learning technology based on casting parameter data collected from equipment in the die casting process in real time. Die-casting parameter data can basically be collected through the casting equipment controller. In order to perform classification analysis for predicting defects by defect type, labeling of casting parameters must be performed. In this study, first, the defective data set is separated by performing the primary clustering based on the total defect rate obtained during the post-processing. Second, the secondary cluster analysis is performed using the defect rate by type for the separated defect data set, and the labeling task is performed by defect type using the cluster analysis result. Finally, a classification learning model is created by collecting the entire labeled data set, and a real-time monitoring system for defect prediction using LabView and Python was implemented. When a defect is predicted, notification is performed so that the operator can cope with it, such as displaying on the monitoring screen and alarm notification.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권10호
• /
• pp.4117-4135
• /
• 2020

Due to the angle of view, illumination and scene diversity, real-time dynamic face detection and recognition is no small difficulty in those unrestricted environments. In this study, we used the intrinsic correlation between detection and calibration, using a multi-task cascaded convolutional neural network(MTCNN) to improve the efficiency of face recognition, and the output of each core network is mapped in parallel to a compact Euclidean space, where distance represents the similarity of facial features, so that the target face can be identified as quickly as possible, without waiting for all network iteration calculations to complete the recognition results. And after the angle of the target face and the illumination change, the correlation between the recognition results can be well obtained. In the actual application scenario, we use a multi-camera real-time monitoring system to perform face matching and recognition using successive frames acquired from different angles. The effectiveness of the method was verified by several real-time monitoring experiments, and good results were obtained.

• 대한의생명과학회지
• /
• 제26권4호
• /
• pp.302-309
• /
• 2020

Human Coxsackievirus B5 (HuCoxV-B5) infection has been associated with various diseases such as myocarditis, aseptic meningitis, hand-foot-and mouth-disease, and insulin-dependent diabetes. HuCoxV-B5 is a virus transmitted through the fecal-oral route and is detected in clinics, aquatic environments, food, shellfish, etc. and is one of the more important viruses in public health because of its incidence rate reported worldwide. In this study, a combination of SYBR Green-based real-time PCR primers for molecular diagnosis including monitoring of HuCoxV-B5 was selected and the optimal reaction conditions were established. Compared with the previously reported TaqMan probe-based real-time PCR method, assessments including a sample applicability test were performed. Results showed that the real-time PCR method developed in this study was suitable for a molecular diagnostic technique for detecting HuCoxV-B5. This study is expected to contribute to efforts in responding to safety accidents in public health because the proposed method facilitates rapid diagnosis of clinical patients. It can also be used as a specific monitoring tool of HuCoxV-B5 in non-clinical areas such as aquatic environments among others.

• 한국멀티미디어학회논문지
• /
• 제15권1호
• /
• pp.87-92
• /
• 2012

무선 센서 네트워크는 실시간으로 객체에 대한 정보를 수집하기 위한 기술로 짧은 시간에 많은 양의 데이터가 지속적으로 발생되는 특성을 갖고 있다. 센서로부터 획득되는 많은 양의 데이터 스트림을 효과적으로 처리하기 위해서는 데이터 간의 관계를 정의하여 분석, 사용자가 주목할 만한 상황, 즉 센서 데이터들 중 이상 데이터가 포함되어 있는 것을 발견하는 것이 필수적이라 할 수 있다. 본 논문에서는 센서 데이터 스트림을 효과적으로 처리하기 위해 ECA 규칙을 사용하여 의미있는 데이터를 구성하고, 이를 통해 보다 실용적인 실시간 모니터링 시스템을 제안한다.

• 한국컴퓨터정보학회논문지
• /
• 제21권6호
• /
• pp.1-7
• /
• 2016

An uninterruptible power supply (UPS) allows small companies and domestic users to cope with power outages; but existing designs lack flexibility of control and require expensive battery maintenance, with a cost proportional to the outage compensation time. We combine a compact synchronous generator with a battery, with 10% of the capacity that would otherwise be required, to obtain a UPS with reduced maintenance costs for the same performance. Any UPS must respond immediately to a power loss, and our uninterruptible generator supply (UGS) is therefore built around real-time scheduling of its internal operations; this also makes it suitable for integration into the industrial gateway. The UGS is based on a real-time operating system, with an integrated wireless module providing connectivity to a web server, for monitoring and management, which can be performed remotely on a mobile device.

• 비파괴검사학회지
• /
• 제27권6호
• /
• pp.556-562
• /
• 2007

This paper presents the use of a novel fiber optic accelerometer system to monitor ambient vibration (both wind-induced one and vehicle-induced) of a real bridge structure. This sensor system integrates the $Moir\'{e}$ fringe phenomenon with fiber optics to achieve accurate and reliable measurements. A low-cost signal processing unit implements unique algorithms to further enhance the resolution and increase the dynamic bandwidth of the sensors. The fiber optic accelerometer has two major benefits in using this fiber optic accelerometer system for monitoring civil engineering structures. One is its immunity to electromagnetic (EM) interference making it suitable for difficult applications in such environments involving strong EM fields, electrical spark-induced explosion risks, and cabling problems, prohibiting the use of conventional electromagnetic accelerometers. The other is its ability to measure both low- and high-amplitude vibrations with a constantly high resolution without pre-setting a gain level, as usually required in a conventional accelerometer. The second benefit makes the sensor system particularly useful for real-time measurement of both ambient vibration (that is often used for structural health monitoring) and strong motion such as earthquake. Especially, the semi-strong motion and the small ambient one are successfully simulated and measured by using the new fiber optic accelerometer in the experiment of the structural health monitoring of a real bridge.

• 한국철도학회논문집
• /
• 제19권4호
• /
• pp.445-455
• /
• 2016

실시간 철도안전 관제시스템은 철도시스템을 구성하는 열차, 신호, 전력 및 설비 등으로 분산되어 감시되던 시스템의 안전관련 데이터를 통합하여 안전감시 효율을 향상시키고 사고를 예방하는 것이 목적으로, 기존 개별 감시 시스템과 달리 데이터의 대용량 처리와 실시간 처리 성능을 동시에 요구하고 있다. 기존 관제시스템에서 주로 활용되는 디스크 기반 데이터베이스는 실시간 및 빅 데이터 처리기능이 없고, 최근 도입되는 메모리 기반 데이터베이스는 빅데이터 처리기능이 없으며, 시계열 데이터베이스는 실시간 처리 기능이 없다. 이에 따라, 실시간 안전관제에서 요구되는 빅 데이터 처리와 실시간 처리를 동시에 제공하는 새로운 솔루션이 필요하다. 본 연구에서는 기존 관제의 데이터 처리 사례를 분석하고, 빅 데이터 처리와 실시간 처리를 동시에 제공하는 새로운 데이터 처리 방안을 제안하였으며, 이를 검증하였다.

• 한국정보통신학회논문지
• /
• 제25권1호
• /
• pp.124-133
• /
• 2021

스마트팩토리는 공장 내 장비에 센서를 부착하여 각종 데이터를 실시간으로 수집, 분석 및 모니터링함으로써 스스로 공정을 제어 및 통제하는 미래형 공장이다. 스마트팩토리에서 장비의 상태 및 이력을 현장에서 실시간으로 조회 및 생성하는 것은 매우 중요한 일이며 다양한 스마트 기기의 등장은 이것을 보다 효율적으로 수행할 수 있게 하고 있다. 본 논문은 실시간으로 장비의 상태 및 이력을 생성, 조회, 삭제 할 수 있는 다중장치 기반 시스템을 제안한다. 제안 시스템은 공장이라는 특수한 환경을 고려하여 안드로이드 시스템과 스마트 글래스 시스템을 동시에 사용한다. 스마트 글래스 시스템은 장비 인식을 위해 QR코드를 이용하였으며 음성 인식 기능을 이용하여 보다 효율적인 작업환경을 제공한다. 다중장치기반 실시간 장비 모니터링을 위한 시스템 구조를 설계하였으며 안드로이드 시스템, 스마트 글래스 시스템 및 웹 애플리케이션 서버를 구현함으로써 실용성을 보이겠다.

• 국제초고층학회논문집
• /
• 제1권1호
• /
• pp.37-51
• /
• 2012

New generation of tall and complex buildings systems are now introduced that are reflective of the latest development in materials, design, sustainability, construction, and IT technologies. While the complexity in design is being overcome by the availability and advances in structural analysis tools and readily advanced software, the design of these buildings are still reliant on minimum code requirements that yet to be validated in full scale. The involvement of the author in the design and construction planning of Burj Khalifa since its inception until its completion prompted the author to conceptually develop an extensive survey and real-time structural health monitoring program to validate all the fundamental assumptions mad for the design and construction planning of the tower. The Burj Khalifa Project is the tallest structure ever built by man; the tower is 828 meters tall and comprises of 162 floors above grade and 3 basement levels. Early integration of aerodynamic shaping and wind engineering played a major role in the architectural massing and design of this multi-use tower, where mitigating and taming the dynamic wind effects was one of the most important design criteria established at the onset of the project design. Understanding the structural and foundation system behaviors of the tower are the key fundamental drivers for the development and execution of a state-of-the-art survey and structural health monitoring (SHM) programs. Therefore, the focus of this paper is to discuss the execution of the survey and real-time structural health monitoring programs to confirm the structural behavioral response of the tower during construction stage and during its service life; the monitoring programs included 1) monitoring the tower's foundation system, 2) monitoring the foundation settlement, 3) measuring the strains of the tower vertical elements, 4) measuring the wall and column vertical shortening due to elastic, shrinkage and creep effects, 5) measuring the lateral displacement of the tower under its own gravity loads (including asymmetrical effects) resulting from immediate elastic and long term creep effects, 6) measuring the building lateral movements and dynamic characteristic in real time during construction, 7) measuring the building displacements, accelerations, dynamic characteristics, and structural behavior in real time under building permanent conditions, 8) and monitoring the Pinnacle dynamic behavior and fatigue characteristics. This extensive SHM program has resulted in extensive insight into the structural response of the tower, allowed control the construction process, allowed for the evaluation of the structural response in effective and immediate manner and it allowed for immediate correlation between the measured and the predicted behavior. The survey and SHM programs developed for Burj Khalifa will with no doubt pioneer the use of new survey techniques and the execution of new SHM program concepts as part of the fundamental design of building structures. Moreover, this survey and SHM programs will be benchmarked as a model for the development of future generation of SHM programs for all critical and essential facilities, however, but with much improved devices and technologies, which are now being considered by the author for another tall and complex building development, that is presently under construction.

• 한국위성정보통신학회논문지
• /
• 제7권1호
• /
• pp.134-138
• /
• 2012

본 논문에서는 자동화 생산 시설물의 객체모니터링을 위해서 생산라인 현장의 상태 감시 모니터링이 가능한 시스템을 구현하고자하며, 효율성을 향상시키기 위해 영상추적필터를 이용한 CCTV 영상추적 기술을 제안하고자 한다. 이 시스템에서는 자동화 생산 시설물의 객체모니터링을 위해 기존에 일반적으로 사용되었던 영상모니터링 방식이 아닌 영상추적필터를 기반으로 소프트웨어를 구축하여 효율적이며, 신뢰성 있는 데이터 전달을 함으로써 PC 기반의 모니터링이 가능하게 한다. 그리고 실시간 상태 확인이 가능하도록 함으로써 관리자의 접근성과 편의성을 향상시켰다. 또한 제안한 모니터링 시스템에 영상추적필터를 적용한 시뮬레이션을 수행함으로써 성능개선효과를 확인하였으며, 제안한 시스템의 효율성과 유용성을 확인하였다.