• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11c
• /
• pp.2333-2336
• /
• 2002

1990년대를 지나면서 컴퓨팅 환경은 급속히 변화하였다. 과거 메인 프레임 중심의 환경에서 네트워크를 기반으로 클라이언트/서버 모델 중심의 분산 컴퓨팅 환경으로 전환되었다. 이러한 환경에서 복잡하고 유지하기 힘든 데이터 네트워크를 관리를 수행하고, 또한 유용한 정보를 줄 수 있는, 효과적인 네트워크 관리의 필요성이 대두되었다. 이러한 필요에 의해서 네트워크 관리의 개념이 등장하였다. 네트워크 관리에는, 기능적 관점에서, 장애 관리, 구성 관리, 계정관리, 성능 관리, 그리고 보안 관리로 나누어진다[1]. 그런데 네트워크가 발전함에 따라, 네트워크에서 발생할 수 있는 각종 장애를 신속히 발견하여 관리자에게 통보하고 적절한 처리를 할 수 있는 기능이 특히 요구되고 있다. 하지만 기존의 SNMP 트랩 기반의 장애 관리는 기능 면에서 부족한 점이 많다[2]. 예를 들어, 불규칙적인 트랩 메시지, 또는 SNMP 패킷의 소실 때문에 정확한 장애 발견에 문제가 생기게 된다. 본 논문에서는 SNMP 장애 관리 기능을 보완하기 위해 ICMP Echo/Reply 메시지를 이용한 장애 처리 시스템의 구조와 알고리즘에 대해 설명하고, 기존의 SNMP 기반의 NMS와 어떻게 연동될 수 있는 방안에 대해서 기술하도록 하겠다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.5
• /
• pp.105-111
• /
• 2010

In this paper the elasticity of fault-detection algorithm based on fuzzy logic is proposed through lots of experiments, justifying its validity. The four mathematical morpholgical operators was defined to detect the cracks. The cracks was detected via center of area method with ${\lambda}$-fuzzy measure of fuzzy sets. However generally favorable, the result owes to how adequate the lighting device is designed in case of the so far fine crack of pieces. In an attempt to improve the response of the system, It is designed to minimize the use of memory via LookUp table in software.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.4
• /
• pp.783-788
• /
• 2020

Surges generated by direct strike or induction of lightning, frequent opening and closing of power system breakers, etc. cause excessive overvoltage, which can interfere with normal operation of transformers, circuit breakers, and power equipment. In order to prevent such phenomena, this study goes further from the method of establishing countermeasures by installing lightning arresters on power equipment or lines in parallel, and furthermore, remotely checking the condition of the lightning arrester and preparing immediate countermeasures when an event occurs. The purpose of the study is to prevent damage in advance.

• Proceedings of the KIEE Conference
• /
• 2003.11c
• /
• pp.555-558
• /
• 2003

When a fault relating to the urgent alarm occurs, we must prevent control rods from dropping and make one of two grippers in Control Rod Drive Mechanism (CRDM) grip the drive rod laking a control rod assembly. If a gripper with any problem is ordered to grip the drive rod, the gripper which cannot latch the rod stably will fail to take the rod. On the purpose of escaping this bad case, we order two grippers to hold the drive rod and enhance the reliability of holding control rods. This action is called the double hold. In the middle of the movement of the drive rod, the latching of the drive rod can cause friction between a gripper and the drive rod. This state may give damage to both the gripper and the drive rod. In this paper, we have devised the method which can have two grippers hold the drive rod more stably, without damaging the equipment.

• Journal of Information Processing Systems
• /
• v.15 no.4
• /
• pp.833-852
• /
• 2019

In real time applications, due to their effective cost and small size, wireless networks play an important role in receiving particular data and transmitting it to a base station for analysis, a process that can be easily deployed. Due to various internal and external factors, networks can change dynamically, which impacts the localisation of nodes, delays, routing mechanisms, geographical coverage, cross-layer design, the quality of links, fault detection, and quality of service, among others. Conventional methods were programmed, for static networks which made it difficult for networks to respond dynamically. Here, machine learning strategies can be applied for dynamic networks effecting self-learning and developing tools to react quickly and efficiently, with less human intervention and reprogramming. In this paper, we present a wireless networks survey based on different machine learning algorithms and network lifetime parameters, and include the advantages and drawbacks of such a system. Furthermore, we present learning algorithms and techniques for congestion, synchronisation, energy harvesting, and for scheduling mobile sinks. Finally, we present a statistical evaluation of the survey, the motive for choosing specific techniques to deal with wireless network problems, and a brief discussion on the challenges inherent in this area of research.

• Journal of Korean Society for Quality Management
• /
• v.47 no.1
• /
• pp.75-86
• /
• 2019

Purpose: It is essential for the steel industry to produce steel products without unexpected downtime to reduce costs and produce high quality products. A hot strip rolling mill consists of many mechanical and electrical units. In condition monitoring and diagnosis, various units could fail for unknown reasons. Methods: In this study, we propose an effective method to detect units with abnormal status early to minimize system downtime. The early warning problem with various units was first defined. An autoencoder was modeled to detect abnormal states. An application of the proposed method was also implemented in a simulated field-data analysis. Results: We can compare images of original data and reconstructed images, as well as visually identify differences between original and reconstruction images. We confirmed that normal and abnormal states can be distinguished by reconstruction error of autoencoder. Experimental results show the possibility of prediction due to the increase of reconstruction error from just before equipment failure. Conclusion: In this paper, hot strip roughing mill monitoring method using autoencoder is proposed and experiments are performed to study the benefit of the autoencoder.

• Tribology and Lubricants
• /
• v.36 no.6
• /
• pp.297-306
• /
• 2020

This paper reviews studies on the types and functions of oil sensors used for machine condition monitoring. Machine condition monitoring is essential for maintaining the reliability of machines and can help avoid catastrophic failures while ensuring the safety and longevity of operation. Machine condition monitoring involves several components, such as compliance monitoring, structural monitoring, thermography, non-destructive testing, and noise and vibration monitoring. Real-time monitoring with oil analysis is also utilized in various industries, such as manufacturing, aerospace, and power plants. The three main methods of oil analysis are off-line, in-line, and on-line techniques. The on-line method is the most popular among these three because it reduces human error during oil sampling, prevents incipient machine failure, reduces the total maintenance cost, and does not need complicated setup or skilled analysts. This method has two advantages over the other two monitoring methods. First, fault conditions can be noticed at the early stages via detection of wear particles using wear particle sensors; therefore, it provides early warning in the failure process. Second, it is convenient and effective for diagnosing data regardless of the measurement time. Real-time condition monitoring with oil analysis uses various oil sensors to diagnose the machine and oil statuses; further, integrated oil sensors can be used to measure several properties simultaneously.

• Tribology and Lubricants
• /
• v.37 no.6
• /
• pp.203-207
• /
• 2021

This study presents the development of dielectric constant sensors to measure lubricant properties. The lubricant oil sensor is used to measure oil properties and machine conditions. Various condition monitoring methods are applied to diagnose machine conditions. Machine condition monitoring using oil sensors has advantage over other machine condition monitoring methods. The fault conditions can be noticed at the early stages by the detection of wear particles using oil sensors. Therefore, it provides an early warning in the failure procedure. A variety of oil sensors are applied to check the machine condition. Among all oil sensors, only one sensor can measure the tendency of several properties such as acidity and water content. A dielectric constant sensor is also used to measure various oil properties; therefore, it is very useful. The dielectric constant is the ratio of the capacitance of a capacitor using that material as a dielectric to that of a similar capacitor using vacuum as its dielectric. The dielectric constant has an effect on water content, contaminants, base oil, additive, and so forth. In this study, the dielectric constant sensor is fabricated using MEMS process. In the fabrication process, the shape, gap of the electrode array, and thickness of the insulation material are considered to improve the sensitivity of the sensor.

• 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.

• The KIPS Transactions:PartC
• /
• v.8C no.5
• /
• pp.657-666
• /
• 2001

In this paper, it is proposed the management items for voice traffic using CDRs so that global carriers can treat and manage the voice traffic for a customer, and defined computational expressions to produce the management items. From them, we have designed the management system, which is composed of web interface module, analysis module, data collection module and database management module, and have improved the availability and convenience of the system using web technologies. In addition, we have tested these items using CDRs in real environments that are collected by the global carrier in order to verify their validity. It is expected that the proposed web based voice traffic management system provide a global carrier with network information collection, fault detection/trouble-shooting and high quality of service through analyzing the characteristics of subscribers.