• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.18 no.5
• /
• pp.37-47
• /
• 2019

Recently, as demand for various paper containers has rapidly grown, it is inevitable that paper cup-forming machines have increased their manufacturing speed. However, the faster manufacturing speed naturally brings more frequent manufacturing failures, which decreases manufacturing efficiency. As such, it is necessary to develop a system that monitors the failures in real time and diagnoses the failure progress in advance. In this research, a paper cup-forming machine diagnosis system was developed. Three major failure targets, paper deviation, temperature failure, and abnormal vibration, which dominantly affect the manufacturing process when they occur, were monitored and diagnosed. To evaluate the developed diagnosis system, extensive experiments were performed with the actual data gathered from the paper cup-forming machine. Furthermore, the desired system validation was obtained. The proposed system is expected to anticipate and prevent serious promising failures in advance and lower the final defect rate considerably.

• Proceedings of the KSME Conference
• /
• 2005.11a
• /
• pp.138.1-138.1
• /
• 2005

• Journal of the KSME
• /
• v.54 no.7
• /
• pp.28-33
• /
• 2014

최근 풍력에너지 발전기술이 해상 풍력 및 대형 풍력으로 천이되어 감에 따라 더 향상된 신뢰성 기술을 요구하고 있고, 풍력 블레이드는 풍력발전시스템에서 고장 빈도가 비교적 낮지 않고, 고장 발생 시 심각도가 상대적으로 높아 신뢰성을 고려해야 하는 부품이다. 블레이드 제작 생산 과정과 사용 중에 발생하는 손상 및 결함은 신뢰성의 심각성에 큰 영향을 미치고 있어 본 기술에서는 블레이드에서 발생할 수 있는 결함 유형, 결함을 탐지하는 기술 그리고 신뢰성을 평가하는 시험/평가 기술에 대하여 소개하고자 한다.

• Journal of the KSME
• /
• v.54 no.7
• /
• pp.40-45
• /
• 2014

풍력발전시스템은 육상용에서 해상용으로 개발되며 대형화하고 있는 추세이다. 풍력발전시스템이 해상용 등으로 대형화되면서 유지, 보수에 높은 비용이 소요되고 이에 따라 신뢰성을 확보하기 위한 요구가 더 커지고 있다. 증속기의 경우 고장 빈도도 낮지 않고, 고장 발생 시 심각도가 높아 신뢰성이 가장 높게 요구된다. 최근 증속기의 신뢰성을 확보하기 위한 연구 개발에서 수명에 가장 영항을 크게 미치는 조건으로 적용하중 영향에 대한 비토크 하중(non-torque load)과 유성기어열의 설계 인자로 취급되는 하중 분할(load sharing), 치면 하중 분포(face-load distribution) 그리고 이를 필수적으로 입증하고자 요구되는 시험/평가 기술에 대하여 소개하고자 한다.

• Proceedings of the KIEE Conference
• /
• 2005.07b
• /
• pp.984-986
• /
• 2005

변압기 권선내에서 발생되는 고장을 유한요소 전자계해석 프로그램(FLUX2D)을 이용하여 해석하였다. 50kVA, 13200/230(V) 단상 변압기의 권선간 고장시 전자계해석을 1차측 권선고장(turn-to-earth)과 2차측 권선고장(turn-to-turn)을 모의하여 해석하였다. 권선내부 고장 및 2차측 단락시 누설 자속분포, 권선간의 힘의 분포, 변압기 내부 권선의 정상시와 단락시의 전자계비교, 단락권선과 권선 상호간의 전자계 해석을 통하여 변압기에 미치는 영향을 해석하였다. 변압기 권선간 단락시의 전자력 분석결과는 변압기의 절연설계 및 단락기계력에 대한 프레임 구조 설계를 위한 자료로 활용될 수 있다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.109-112
• /
• 2006

본 논문에서는 PCA에 의한 특징추출과 k-NN과 SVM에 기반을 계층구조의 분류기에 의한 유도전동기의 고장진단 알고리즘을 제안한다. 제안된 방법은 k-NN에 의해 선형적으로 분류 가능한 고장패턴을 분류한 후, 분류가 되지 않는 부분을 커널 함수에 의해 고차원 공간으로 입력패턴을 매핑한 후 SVM에 의해 고장을 진단하는 계층구조를 갖는다. 실험장치를 구축한 후, 다양한 부하에 대하여 몇몇의 전기적 고장과 기계적 고장 하에서 획득한 데이터를 이용하여 제안된 방법의 타당성을 검증한다.

• Journal of Navigation and Port Research
• /
• v.46 no.3
• /
• pp.280-288
• /
• 2022

This study examined the diagnostics of abnormalities and faults of equipment, whose rotational speed changes even during regular operation. The purpose of this study was to suggest a procedure that can properly apply machine learning to the time series data, comprising non-stationary characteristics as the rotational speed changes. Anomaly and fault diagnosis was performed using machine learning: k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), and Random Forest. To compare the diagnostic accuracy, an autoencoder was used for anomaly detection and a convolution based Conv1D was additionally used for fault diagnosis. Feature vectors comprising statistical and frequency attributes were extracted, and normalization & dimensional reduction were applied to the extracted feature vectors. Changes in the diagnostic accuracy of machine learning according to feature selection, normalization, and dimensional reduction are explained. The hyperparameter optimization process and the layered structure are also described for each algorithm. Finally, results show that machine learning can accurately diagnose the failure of a variable-rotation machine under the appropriate feature treatment, although the convolution algorithms have been widely applied to the considered problem.

• Journal of Aerospace System Engineering
• /
• v.16 no.6
• /
• pp.129-137
• /
• 2022

Recently, the technology of aircraft drivers has been transitioning from the existing hydraulically-focused mechanical system to an all-electric one due to the high precision and ease of maintenance of electric drivers. Consequently, the failure of an aircraft's electric motor can have fatal consequences. To ensure aircraft safety, efficient and timely fault diagnosis methods are required prompting the active pursuit of research into fault diagnosis technology. This paper introduces and analyses the failure types and failure diagnosis technology trends of permanent magnet synchronous motors among electric motors.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.6
• /
• pp.629-633
• /
• 2011

Recently, systems such as aircraft, trains and ships have become larger more complex. Therefore, the reliability calculation of these systems is more difficult. This paper presents a reliability calculation algorithm for a complex system with a solution that is difficult to analyze. When the system has a very complex structure, it is very difficult to find an analytical solution. In this case, we can assess system reliability using the failure enumeration method of the reliability path. In this research, we represent the reliability block diagram by an adjacent matrix and define the reliability path. We can find any system status by the failure enumeration of the reliability path, and thus we can calculate any kind of system reliability through this process. This result can be applied to RCM (Reliability-Centered Maintenance) and reliability information-management systems, in which the system reliability is composed of the reliabilities of individual parts.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.12
• /
• pp.1379-1385
• /
• 2014

Even the rough prediction of the product test time before the lifetime test of mechanical component begins would be of use in estimating cost and deciding how to keep up with the test. The reliability predictions of mechanical components are difficult because failure or degradation mechanisms are complicated, and few plausible databases are available for lifetime prediction. Therefore, this study conducted lifetime predictions of elastomeric U seals that were respectively installed in a hydraulic actuator and a pneumatic actuator using lifetime models and a field database based on failure physics and an actual test database obtained from the NSWC handbook. To validate the results, the predicted failure rates were compared with the actual lifetime test results acquired in the lab durability tests. Finally, this study discussed an engineering procedure to determine the coefficients in the failure rate models and analyzed the sensitivity of each influential parameter on the seal lifetime.