• 대한기계학회논문집A
• /
• 제35권8호
• /
• pp.965-970
• /
• 2011

본 연구를 통하여 개발하고자 하는 진단시스템은 자동차 엔진 어셈블리라인에서 발생될 수 있는 특정 조립 불량유형을 검사하는 시스템이다. 대상으로 하는 불량 유형은 엔진 고장의 유형 중 커다란 충격성신호가 발생한 후, 보상적인 작은 충격파가 주기적으로 발생되는 형태이다. 이러한 불량유형을 기존의 시간영역분석 진단, 주파수분석, 통계적분석등에 의해 진단할 경우 한계점이 존재한다. 이에 웨이블릿 잡음 제거 전처리 방법, 피크검지 알고리즘, 불순도 최소값 선택 분할 방법을 이용한 새로운 유형의 이상진단 방법을 개발하는 연구를 진행하였다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2012년도 추계학술대회 논문집
• /
• pp.394-399
• /
• 2012

최근 MEMS 센서는 기계상태감시에 있어서 전력소모, 크기, 비용, 이동성, 응용 등에 있어서 각광을 받고 있다. 특히, MEMS 센서는 스마트센서와 통합가능하고, 대량생산이 가능하여 가격이 저렴하다는 장점이 있다. 이와 관련한 기계상태감시를 위한 많은 실험적 연구가 수행되고 있다. 이 논문은 MEMS 센서들을 3 가지 인공지능 분류기 성능평가를 위한 비교연구에 대해 설명하고 있다. 회전기계에 MEMS 가속도와 전류센서들을 부착하여 데이터를 취득했고, 특징추출과 파라미터 최적화를 위해 Cross validation 기법을 사용하였다. MEMS 센서를 이용한 결함분류기 적용은 적합하다고 판단된다.

• Tribology and Lubricants
• /
• 제30권6호
• /
• pp.356-363
• /
• 2014

Marine diesel engines operate in environments in which damage easily occurs from corrosion. Recently, damage to cylinder liners has increased from corrosion wear caused by increased engine power. This damage can cause serious problems in the economy. Thus, many researchers have treated and studied damaged cylinder liners. However, a method is necessary for real-time monitoring of damage to cylinder liners during operation of the engine, before serious damage can occur. This study carries out reciprocating friction and wear tests on a cast iron specimen under various corrosion atmospheres and verifies the variations of friction coefficient and friction surface. Additionally, the friction coefficient and friction status are predicted by using a neural network that learns the vibration and frequency spectrum data from an acceleration sensor. According to our conclusions, amplitude is distributed highly at high frequencies, and values of standard deviation and kurtosis are high when damage to the friction surface is serious. The accuracy rate of the friction coefficient predicted by the neural network is over 80% of the real measured value without NaCl, and application of the neural network is very effective for diagnosing the friction condition and damage to the cylinder liner.

• 한국정밀공학회지
• /
• 제21권5호
• /
• pp.73-82
• /
• 2004

This paper presents the study on damage diagnosis of an intelligent cantilevered beams using PZT actuator and PVDF sensor This study provides the theoretical and experimental verification to examine structural damage. Time domain analysis for the non-destructive detection of damage is presented by parameterized partial differential equations and Galerkin approximation techniques. The time histories of the vibration response of structure were used to identify the presence of damage. Furthermore, this systematic approach permits one to use the piezomaterials to both excite and sense the vibration of structures. We also carried out the experimental verification about reliability of theoretical methods fur detecting the damage of a composite beam with PZT actuator and PVDF sensor. Experimental results are presented from tests on cantilevered composite beams which is damaged at different location and different dimensions. The results were compared with the simulation results. Good agreement between the results was found for the time shifts and amplitude difference in transients response of the cantilevered beam.

• 한국기계가공학회지
• /
• 제20권9호
• /
• pp.28-34
• /
• 2021

The Fourth Industrial Revolution has led to the development of drones for commercial and private applications. Therefore, the malfunction of drones has become a prominent problem. Failure mode and effect analysis was used in this study to analyze the primary cause of drone failure, and blade breakage was observed to have the highest frequency of failure. This was tested using a vibration sensor placed on drones along the breakage length of the blades. The data exhibited a significant increase in vibration within the drone body for blade fracture length. Principal component analysis was used to reduce the data dimension and classify the state with machine learning algorithms such as support vector machine, k-nearest neighbor, Gaussian naive Bayes, and random forest. The performance of machine learning was higher than 0.95 for the four algorithms in terms of accuracy, precision, recall, and f1-score. A follow-up study on failure prediction will be conducted based on the results of fault diagnosis.

• 한국기계가공학회지
• /
• 제20권9호
• /
• pp.35-41
• /
• 2021

A rotor is a crucial component in various mechanical assemblies. Additionally, high-speed and high-efficiency components are required in the automotive industry, manufacturing industry, and turbine systems. In particular, the failure of high-speed rotating bearings has catastrophic effects on auxiliary systems. Therefore, bearing reliability and fault diagnosis are essential for bearing maintenance. In this work, we performed failure mode and effect analysis on bearing rotors and determined that corrosion is the most critical failure type. Furthermore, we conducted experiments to extract vibration characteristic data and preprocess the vibration data through principle component analysis. Finally, we applied a machine learning algorithm called support vector machine to diagnose the failure and observed a classification performance of 98%.

• Journal of Preventive Medicine and Public Health
• /
• 제29권3호
• /
• pp.507-519
• /
• 1996

The objectives of this study are to investigate the prevalence of occupation related carpal tunnel syndrome(CTS) among workers in a condom industry : to analyse the sensitivity and specificity of clinical signs or symptoms such as hand diagram, Tinel's sign and Phalen's sign in carpal tunnel syndrome : and to test vibration threshold test using audiometry as a technically easy and noninvasive method in the diagnosis of carpal tunnel syndrome in stead of nerve conduction velocity (NCV). The study group was divided into exposed group(39 cases) and non-exposed group(48 cases) based on whether or not excessive use of wrist movements exsist. 1. There are stastically significant differences in symptoms and signs of carpal tunnel syndrome such as hand diagram, Tinel's sign and Phalen's sign between exposed and non-exposed group(p<0.05). 2. Six cases(9 hands) were comfirmed as carpal tunnel syndrome by NCV. Five cases(7 hands) belonged to exposed group, 1 case(2 hands) to nonexposed group. As there are significant differences in prevalence of carpal tunnel syndrome between two groups(p<0.05), excessive use of wrist in occupation is a risk factor of carpal tunnel syndrome. 3. When we use NCV as a gold standard in the diagnosis of carpal tunnel syndrome, sensitivity and specificity of hand diagram, Tinel's sign and Phalen's sign is as followed; hand diagram , sensitivity 88.9%, specificity 84.2% Tinel's sign ; sensitivity 55.6%, specificity 72.8% Phalen's sign ; sensitivity 14.3%, specificity 88.4%. Among above clinical signs and symptoms, hand diagram is the best clinical screening test. 4. The differences of vibration threshold between median and ulnar nerve at the same time are useful in the diagnosis of carpal tunnel syndrome but the time change of vibration threshold of median nerve over time are not sensitive enough. It is concluded that vibration threshold between median and ulnar nerve at the same time can be used as a supplementary or alternative criterion to indicate that the nerve dysfunction is located in the carpal tunnel.

• 플랜트 저널
• /
• 제17권4호
• /
• pp.53-60
• /
• 2021

본 연구에서는 국내 진동감시시스템 제작업체인 (주)나다와 발전소 회전기기 진동데이터 취득 및 분석에 효과적인 휴대용 진동분석기를 개발하고자 하였다. 개발된 휴대용 진동분석기의 하드웨어는 측정기기의 교정을 위한 측정 불확도를 통해 얻은 측정값을 시스템에서 보정함으로써 측정오차를 최소화하였고, 고속 데이터 처리로 높은 분해능을 가진 신호처리 장치로 구성되었다. 소프트웨어 구조는 다양한 진동 플롯을 구현하여 상세한 분석프로그램을 실행하며, 회전기기 운전 중 외란으로 발생하는 노이즈 측정 및 제거에 효과적인 알고리즘을 적용하였다. 개발품은 사용자의 이동 편의성 증대 및 고속 데이터 처리로 분해능을 높임으로써 성능향상은 물론, 국산화 개발에 따른 구입비용 절감에도 크게 기여하였다.

• 한국산업융합학회 논문집
• /
• 제23권5호
• /
• pp.825-833
• /
• 2020

Currently, the K-Water uses various valves that can be remotely controlled for optimal water management. Valve system fault can be classified into rotor defects, stator defects, bearing defects, and gear defects of induction motors. If the valve cannot be operated due to a gear fault, the water management operation can be greatly affected. For effective water management, there is an urgent need for preemptive repairs to determine whether gear is damaged through failure prediction diagnosis.. Recently, deep learning algorithms are being applied for valve failure diagnosis. However, the method currently applied has a disadvantage of attaching a vibration sensor to the valve. In this paper, propose a new algorithm to determine whether a fault exists using a convolutional neural network (CNN) based on the voltage and current information of the valve without additional sensor mounting. In particular, a normalized Lisasjous diagram was used to maximize the fault classification performance in the CNN-based diagnostic system.

• 한국정보통신학회논문지
• /
• 제4권5호
• /
• pp.951-961
• /
• 2000

압연기의 기계적인 이상을 사전에 알아내는 압연기 베어링 고장-진단 시스템은 예측하지 못하는 압연 공정의 중단으로 인하여 발생하는 큰 피해를 사전에 막기 위해서 매우 중요한 시스템이다. 그러나 압연기의 동적 거동은 비선형 특성이 매우 강하기 때문에 압연기에서 사전에 고장 예측 정보를 제공하는 것은 매우 어렵다. 본 논문에서는 웨이브렛을 이용한 압연기의 고장 진단 방법을 제안하였으며 제안된 방법은 온라인으로 압연기에서 진동 신호를 실시간으로 측정하여 웨이브렛을 이용하여 패턴을 분석하고 분석된 결과로부터 고장 특징 정보를 얻었다. 얻어진 데이터를 이용하여 압연기 베어링을 진단하는 뉴로 퍼지 모델을 설계하고 수치적인 시뮬레이션을 통하여 그 타당성을 입증하였다.