공기구동 제어밸브 비정상상태 운전변수에 관한 실험적 연구

김양석(Yang-seok Kim),김대웅(Dae-woong Kim),이병오(Byoung-oh Lee),정래혁(Rae-hyuk Jeoung),이승호(Seung-ho Lee) 대한기계학회 2016 大韓機械學會論文集A Vol.40 No.6

공기구동 제어밸브는 여러 유형의 발전소에서 유체계통의 유량을 제어하거나 탱크의 수위를 조절하는 중요한 기능을 수행하며 발전소 수명기간 동안 성능이 보장되어야 한다. 공기구동 제어밸브는 공기구동기에 공기를 공급하거나 배기하여 작동시키며, 다양한 기능을 하는 부속기기로 구성된 제어설비를 이용하여 공기량을 조절함으로서 밸브 위치를 제어한다. 본 논문에서는 2인치 공기구동기에 I/P 변환기, 포지셔너 등이 장착된 제어밸브 동작모사 실험 장치를 이용하여 공기공급배관에서의 공기누설, 밸브 패킹마찰력 변화, 포지셔너 설정 불량 등을 모사하고, 약 67% 개도에서 ±2% 이하 제어구간에서의 밸브-구동기 거동을 측정하여 비교하였다. A pneumatic control valve performs a major role in controlling the flow of a system or the level of a key tank in many power plants, and its performance should be guaranteed during the plant"s lifetime. Its operation starts by supplying air to the pneumatic actuator or by exhausting the air from the actuator. To control the valve position, the amount of air supply or exhaust is adjusted by a control loop where various accessaries are equipped. In this paper, air leakage in the air supply line, changes in the valve packing force, and false adjustments of zero and the span of the positioner are simulated and analyzed using a 2-in pneumatic valve with a position control loop including an I/P converter and positioner, where the valve position is controlled within ±2% of the control pressure at 67% opening position.

시간경과에 따른 공기구동밸브의 성능저하 평가 방법

김양석(Kim Yang-seok),김대웅(Kim Dae-woong),강신철(Kang Shin-cheul),최종민(Choi Jong-min) 한국유체기계학회 2014 한국유체기계학회 학술대회 논문집 Vol.2014 No.11

결함유형별 최적 특징과 Support Vector Machine을 이용한 회전기계 결함 분류

김양석(Yang Seok Kim),이도환(Do Hwan Lee),김성국(Seong Kook Kim) 대한기계학회 2010 대한기계학회 춘추학술대회 Vol.2010 No.5

Support Vector Machine(SVM)을 이용한 회전기계 진단 연구가 많이 수행되어 왔으나 결함 분류 성능은 입력 특징과 더불어 다중 분류 방법, 이진분류기, 커널함수 등에 따라 다르다. SVM 을 이용한 대부분의 기존 연구들은 한번 입력 특징들을 선정하면 결함 분류시 동일한 특징데이터를 이용한다. 본 논문에서는 회전기계의 다양한 결함조건에서 측정한 진동신호로부터 추출한 통계적 특징들을 이용하여 각각의 결함을 분류하기 위한 최적 특징들을 선정한 후, 해당 결함상태를 분류하기 위한 SVM 학습과 분류에 각각 이용하였다. 실험자료를 이용한 검증 결과, 제안한 단계 분류 방법이 분류 성능을 개선할 수 있음을 보여주었다. Several studies on the use of Support Vector Machine (SVM) for rotating machinery diagnosis have been successfully performed, but the classification performance depends on input features as well as multi-classification scheme, binary optimizer, kernel function and parameter to be used. Once input features were selected, most of the published papers on multi-class SVM application used the same features to classify the faults to be considered. In this paper, simple statistical features are extracted from time domain vibration signals for various fault conditions and the optimal features for each fault condition are selected. Then the optimal features are applied to the SVM training and classification for each fault condition to be identified from the other fault conditions. Simulation results using experimental data show that the proposed stepwise classification approach can improve the classification performance.

결함유형별 최적 특징과 Support Vector Machine 을 이용한 회전기계 결함 분류

김양석(Yang Seok Kim),이도환(Do Hwan Lee),김성국(Seong Kook Kim) 대한기계학회 2010 大韓機械學會論文集A Vol.34 No.11

Support Vector Machine(SVM)을 이용한 회전기계 진단 연구가 많이 수행되어 왔으나 결함 분류성능은 입력 특징과 더불어 다중 분류 방법, 이진분류기, 커널함수 등에 따라 다르다. SVM 을 이용한 대부분의 기존 연구들은 한번 입력 특징들을 선정하면 결함 분류시 동일한 특징데이터를 이용한다. 본 논문에서는 회전기계의 다양한 결함조건에서 측정한 진동신호로부터 추출한 통계적 특징들을 이용하여 각각의 결함을 분류하기 위한 최적 특징들을 선정한 후, 해당 결함상태를 분류하기 위한 SVM 학습과 분류에 각각 이용하였다. 실험자료를 이용한 검증 결과, 제안한 단계 분류 방법이 상대적으로 적은 학습시간으로 단일 다중 분류 방법과 유사한 분류 성능을 얻을 수 있었다. Several studies on the use of Support Vector Machines (SVMs) for diagnosing rotating machinery have been successfully carried out, but the fault classification depends on the input features as well as a multi-classification scheme, binary optimizer, kernel function, and the parameter to be used in the kernel function. Most of the published papers on multiclass SVM applications report the use of the same features to classify the faults. In this study, simple statistical features are determined on the basis of time domain vibration signals for various fault conditions, and the optimal features for each fault condition are selected. Then, the optimal features are used in the SVM training and in the classification of each fault condition. Simulation results using experimental data show that the results of the proposed stepwise classification approach with a relatively short training time are comparable to those for a single multi-class SVM.

공기구동 제어밸브 비정상상태 운전변수에 관한 실험적 연구

김양석(Yang-seok Kim),김대웅(Dae-woong Kim),이병오(Byoung-oh Lee),정래혁(Rae-hyuk Jeoung),이승호(Seung-ho Lee) 대한기계학회 2015 대한기계학회 춘추학술대회 Vol.2015 No.11

A pneumatic control valve performs a major role to control the flow of a system or level of a key component in many power plants and its performance should be guaranteed during the plant’s life time. In particular, flow or level control failure may lead to plant shutdown so that maintaining control function is very important to both plants and operators. The operation of the pneumatic valve starts by supplying the air to the pneumatic actuator or exhausting the air from the actuator. To control the valve position, the amount of air supply or exhaust are adjusted by the control loop where various accessaries are equipped. In this paper, air leakage in the air supply line, change of the valve packing force, and false adjustment of zero and span of the positioner are simulated by a 2-inch pneumatic valve with a position control loop including I/P converter, positioner, volume booster etc. The measured data from each condition are compared to drive an idea for prognostics and to validate the currently developing prognostic model.

서포트 벡터 머신을 이용한 볼 베어링의 결함 정도 진단

김양석(Yang-Seok Kim),이도환(Do-Hwan Lee),김대웅(Dae-Woong Kim) 대한기계학회 2012 대한기계학회 춘추학술대회 Vol.2012 No.5-2

서포트 벡터 머신(Support Vector Machine, SVM)은 학습용 데이터 집합이 확보되어 있을 경우, 매우 강력한 분류 알고리즘이다. 따라서 패턴인식은 물론 기계학습 분야에서 결함 진단 도구의 하나로 이용되고 있다. 본 논문에서는 최적 특징과 SVM을 이용하여 볼 베어링의 결함유형과 결함 정도를 진단한 결과를 기술하였다. SVM 학습용 특징데이터에는 12개의 시간영역 특징과 9개의 주파수영역 특징들이 포함되어 있으며 이들 특징들은 다양한 베어링 결함조건에서 측정된 진동신호와 진동신호의 이산 웨이블렛 변환신호로부터 추출하였다. Support Vector Machine (SVM) is a very powerful classification algorithm when a set of training data, each marked as belonging to one o f several categories is given. So, SVM technique has been used as one of the diagnostic tools in machine learning as well as pattern recognition. In this paper, we present the results of classifying ball bearing fault types and severities with an optimized feature set based on the minimum distance rule. A feature set as input for SVM includes twelve time-domain and nine frequency-domain features which extracted from the measured vibration signals and their decomposed details and approximations with discrete wavelet transform. The vibration signals were obtained from a test rig to simulate various bearing fault conditions.

서포트 벡터 머신을 이용한 볼 베어링의 결함 정도 진단

김양석(Kim Yang-seok),이도환(Lee Do-hwan),김대웅(Kim Dae-woong) 대한기계학회 2013 大韓機械學會論文集B Vol.37 No.6

서포트 벡터 머신(Support Vector Machine, SVM)은 학습용 데이터 집합이 확보되어 있을 경우, 매우 강력한 분류 알고리즘이다. 따라서 패턴인식은 물론 기계학습 분야에서 결함진단 도구의 하나로 이용되고 있다. 본 논문에서는 최적 특징과 SVM 을 이용하여 볼 베어링의 결함유형과 결함의 정도를 진단한 결과를 기술하였다. SVM 학습용 특징데이터에는 12 개의 시간영역 특징과 9 개의 주파수영역 특징들이 포함되어 있으며 이들 특징들은 다양한 베어링 결함조건에서 측정된 진동신호와 진동신호의 이산 웨이블렛 변환신호로부터 추출되었다. A support vector machine (SVM) is a very powerful classification algorithm when a set of training data, each marked as belonging to one of several categories, is given. Therefore, SVM techniques have been used as one of the diagnostic tools in machine learning as well as in pattern recognition. In this paper, we present the results of classifying ball bearing fault types and severities using SVM with an optimized feature set based on the minimum distance rule. A feature set as an input for SVM includes twelve time-domain and nine frequency-domain features that are extracted from the measured vibration signals and their decomposed details and approximations with discrete wavelet transform. The vibration signals were obtained from a test rig to simulate various bearing fault conditions.

스카치-요크형 공기구동기의 효율 변화 연구

김양석(Kim Yang-seok),김대웅(Kim Dae-woong),최종민(Choi Jong-min) 한국유체기계학회 2014 한국유체기계학회 학술대회 논문집 Vol.2014 No.11

스윙형 역지 밸브 개도 예측 모델 개선

김양석(Yang-seok Kim),송석윤(Seok-yoon Song),김대웅(Dae-woong Kim),박성근(Sung-keun Park) 한국유체기계학회 2004 유체기계 연구개발 발표회 논문집 Vol.- No.-

Swing check valves are the most common type of check valve in nuclear power plant and need to be operated properly to perform their functions and to keep the valve internals stable. However, for a swing check valve disc to remain stable, the opening characteristics should be identified and the upstream flow velocity should be enough to hold the disc fully open and without motion. Thus it is necessary to develop a model for predicting the flow velocity for a given disc opening. In the present study, the disc positions with mean flow velocity were measured for 3 inch and 6 inch swing check valves. Comparison of the measurements with the existing models showed that the models underestimate the mean flow velocity for a given disc position. Therefore, the existing model for predicting swing check valve disc position was improved with the realistic disc impingement area perpendicular to the flow stream and the experimental data. The result showed that the improved model with the best estimate of kb = 0.04 predicts well the disc openings of 6 inch swing check valve, especially in the low velocity region. For better prediction of the disc opening at high flow velocity, however, it is recommended to develop a kb correlation with the disc angle.

초등학교 일반교실 색채계획 개선에 관한 연구

김양석 ( Yang Seok Kim ),김유지 ( You Ji Kim ) 조선대학교 공학기술연구원 2009 공학기술논문지 Vol.2 No.2

This study compares, analyzes and reevaluates color conditions and images before and after the execution of the 7th curriculum with elementary schools in Gwang-ju Metropolitan City based on the fact that color among physical environments of elementary classrooms has an important influence on children``s emotion and draws problems. Consequently, according to color conditions, major colors on ceiling and wall were high brightness/high chroma of Y and YR line regardless of division of grades and other color factors of classroom showed that wall and ceiling used similar colors, but brightness was higher after change than before it. In evaluation of color images, lower grades showed negative color images such as ``common,`` ``rustic,`` and ``strange`` before change, but they showed positive color images after change such as ``clean,`` ``unique; and ``soft``. But, most of higher grades showed negative reactions before change such as ``common: ``strange,`` and ``rustic`` and they showed positive reactions after change only in such items as ``clean,`` ``warm,`` and ``unique``, but indicated negative reactions i n remaining items such as ``strange,`` ``rustic,`` ``stiff,`` and ``artificial``. Accordingly, in using colors for lower grade classroom, arrangement of colors through adjustment of brightness and chroma between ceiling, wall and other components based on coloring of high brightness and low chroma of similar colors should be considered and for higher grade classroom, arrangement of colors of unique images of mid-brightness/mid-chroma among ..an... n color lines of similar colors should be examined.