수소 연료전지용 가스켓의 유한요소해석

천강민(Kang-Min Cheon),장종호(Jong-Ho Jang),허장욱(Jang-Wook Hur) 한국기계가공학회 2021 한국기계가공학회지 Vol.20 No.10

An analysis was conducted to predict the behavior of gasket by applying an optimal-strain energy-density function selected through a uniaxial tensile test and an analysis of the gasket used in an actual hydrogen fuel cell. Among the models compared to predict the materials" properties, the Mooney-Rivlin secondary model showed the behavior most similar to the test results. The maximum stress of the gasket was not significantly different, depending on the location. The maximum surface pressure of the gasket was higher at positions “T” and “Y” than at other positions, owing to the branch-shape effect. In the future, a jig that can measure the surface pressure will be manufactured and a comparative verification study will be conducted between the test results and the analysis results.

설명 가능한 AI를 적용한 기계 예지 정비 방법

천강민(Kang Min Cheon),양재경(Jaekyung Yang) 한국산업경영시스템학회 2021 한국산업경영시스템학회지 Vol.44 No.4

Predictive maintenance has been one of important applications of data science technology that creates a predictive model by collecting numerous data related to management targeted equipment. It does not predict equipment failure with just one or two signs, but quantifies and models numerous symptoms and historical data of actual failure. Statistical methods were used a lot in the past as this predictive maintenance method, but recently, many machine learning-based methods have been proposed. Such proposed machine learning-based methods are preferable in that they show more accurate prediction performance. However, with the exception of some learning models such as decision tree-based models, it is very difficult to explicitly know the structure of learning models (Black-Box Model) and to explain to what extent certain attributes (features or variables) of the learning model affected the prediction results. To overcome this problem, a recently proposed study is an explainable artificial intelligence (AI). It is a methodology that makes it easy for users to understand and trust the results of machine learning-based learning models. In this paper, we propose an explainable AI method to further enhance the explanatory power of the existing learning model by targeting the previously proposedpredictive model [5] that learned data from a core facility (Hyper Compressor) of a domestic chemical plant that produces polyethylene. The ensemble prediction model, which is a black box model, wasconverted to a white box model using the Explainable AI. The proposed methodology explains the direction of control for the major features in the failure prediction results through the Explainable AI. Through this methodology, it is possible to flexibly replace the timing of maintenance of the machine and supply and demand of parts, and to improve the efficiency of the facility operation through proper pre-control.

앙상블 모델 기반의 기계 고장 예측 방법

천강민(Kang Min Cheon),양재경(Jaekyung Yang) 한국산업경영시스템학회 2020 한국산업경영시스템학회지 Vol.43 No.1

There have been a lot of studies in the past for the method of predicting the failure of a machine, and recently, a lot of researches and applications have been generated to diagnose the physical condition of the machine and the parts and to calculate the remaining life through various methods. Survival models are also used to predict plant failures based on past anomaly cycles. In particular, special machine that reflect the fluid flow and process characteristics of chemical plants are connected to hundreds or thousands of sensors, so there are not many factors that need to be considered, such as process and material data as well as application of derivative variables. In this paper, the data were preprocessed through time series anomaly detection based on unsupervised learning to predict the abnormalities of these special machine. Next, clustering results reflecting clustering-based data characteristics were applied to produce additional variables, and a learning data set was created based on the history of past facility abnormalities. Finally, the prediction methodology based on the supervised learning algorithm was applied, and the model update was confirmed to improve the accuracy of the prediction of facility failure. Through this, it is expected to improve the efficiency of facility operation by flexibly replacing the maintenance time and parts supply and demand by predicting abnormalities of machine and extracting key factors.

클러스터링 알고리즘을 이용한 배관의 부식 상태 분류

천강민(Kang-Min Cheon),신건호(Geon-Ho Shin),허장욱(Jang-Wook Hur) 한국기계가공학회 2022 한국기계가공학회지 Vol.21 No.7

다구찌 기법을 이용한 수소 연료전지용 가스켓 설계

천강민(Kang-Min Cheon),안준현(Jun-Hyeon An),허장욱(Jang-Wook Hur) 한국기계가공학회 2022 한국기계가공학회지 Vol.21 No.1

The Mooney-Rivlin second order optimal strain energy function derived through uniaxial tensile test and analysis was applied to a gasket to confirm the internal stress and surface pressure generated during compression. The Taguchi method, a statistical technique, was used to design the optimum shape of the gasket, and through characteristic evaluation, the optimum shape of the gasket was obtained when the reference plane (T: 0.15 mm), contact surface (W: 1.00 mm), and curvature (R: 0.30 mm) were used. It was determined that the optimum shape yields a von Mises stress of 4.83 MPa, and the contact pressure stress is 20.14 MPa, which satisfies breakage and sealing requirements. In the future, we plan to manufacture a jig that can measure surface pressure to conduct comparative verification studies between the test results and analysis results.

레이저 변위 센서를 활용한 배관 표면 상태분류

천강민(Kang-Min Cheon),신백천(Baek-Cheon Shin),신건호(Geon-Ho Shin),고정일(Jeong-Il Go),이준혁(Jun-Hyeok Lee),허장욱(Jang-Wook Hur) 한국기계가공학회 2022 한국기계가공학회지 Vol.21 No.5

Although pipe performs various functions in industrial sites and residential spaces, if it is damaged due to corrosion caused by the external environment, it may cause equipment failure or a major accident. For this reason, various studies for safety management are being conducted, but studies on detecting corrosion or cracks on the pipe surface using a laser displacement sensor have hardly been conducted. Therefore, in this study, the corrosion degree of the pipe surface was compared and classified into 4 corrosion conditions, and inspection equipment using a laser scanner was manufactured. The corrosion height was calculated from the four surface data obtained from the measuring equipment and applied to various CNN algorithms, and 91% accuracy was obtained during training using the Modified VGGNet16 code with reduced number of parameters.

앙상블모델을 활용한 기계 고장 예측 및 주요 인자 선별 방법

천강민(Cheon Kang Min),양재경(Jaekyung Yang) 한국산업경영시스템학회 2019 한국산업경영시스템학회 학술대회 Vol.2019 No.추계

설비의 이상(고장)을 예측하는 방법들은 과거에도 많은 연구들이 있었고 최근 역시 유사한 방법들을 통해 설비와 부품의 물리적 상태 진단을 통해 잔존 수명을 산출하는 사례가 많으며, 생존모형을 활용하여 과거 이상 주기 기반의 설비 수명을 예즉하기도 한다. 그러나 단독 설비의 이상을 예측하는 것과 순방향 및 역방향 공정 프로세스가 혼합되어 있는 특수 설비의 수명을 예측하거나 이상 진단을 위해 단순 생존 모형만을 가지고는 설명하기 어려운 점이 많다. 또한 단독 설비가 아닌 연계설비(다수의 연계 공정), 특히 화학공장의 유체 흐름과 공정 특성이 반영하는 특수 설비는 수백 또는 수천 개의 센서와 연결되어 있기 때문에 설비 관련 데이터뿐만 아니라 공정 및 재료 데이터 및 파생 변수 적용 등 고려되어야 할 요소들이 적지 않다. 본 논문에서는 이러한 특수 설비의 이상을 예즉하기 위해 비지도학습 기반의 시계열 이상탐지 방법을 통해 데이터를 필터링하였다. 다음으로 클러스터링 기반의 데이터 특성을 반영한 군집요소를 추가 변수 적용하였으며 과거 설비 이상 히스토리를 기반으로 학습 데이터 셋을 생성하였다. 마지막으로 지도학습 알고리즘 기반의 예측 방법론을 적용하였으며 모델 업데이트를 통해 설비 이상 예측의 정확도가 향상되는 것을 확인하였다. 이를 통해 설비의 이상을 예측하고 주요인자를 추출함으로써 설비의 정비 시점과 부품 수급에 대해 유연하게 대체함으로써 설비 운영의 효율성 향상을 기대한다. Research for predicting abnormalities of equipment has been introduced. Recently, there have been also cases estimating the remaining life from the physical status diagnosis of equipment and parts of a machine by using similar methods to the past. In addition, the method using a duration model based on the history data of an abnormal cycle is one of such methods. However, by using simple duration model, it has more limitation on predicting the remaining life and abnormalities of special equipment in the forward and backward mixing process than doing those of independent equipment. Since the special equipment in the interconnected processes such as fluid flow processes in a chemical plant communicates with hundreds or thousands of sensors, there is no lack of factors to consider process and material data, derived variable application, equipment related data, etc. This paper proposes an ensemble model with multiple algorithms such as training dataset generation, anomaly detection, clustering xgboost, and survival model to predict and identify the major factors equipment anomalies.

데이터마이닝과 RSM 혼합모델을 이용한 스마트폰카메라부품 공정 개선

양재경(Jaekyung Yang),천강민(Kang-min Cheon),변용완(Yong-Wan Byun) 한국산학기술학회 2015 한국산학기술학회 학술대회 Vol.2015 No.1

본 논문은 스마트폰 카메라 바디 제조 빅데이터 분석의 정확성을 높이고자 제조 공정 개선에 이용할 수 있는 데이터마이닝과 RSM(Response surface methodology) 혼합 모델을 제안하고 있다. 정밀 사출성형기로부터 수집된 제조 빅데이터는 혼합모델의 입력으로 이용되었으며, 생산된 제품의 품질 결과를 보여주는 제품 치수 데이터는 혼합 모델의 품질 결과로 사용하였다. 데이터 전처리 단계를 거쳐 속성선택을 통해 선정된 품질에 영향을 미치는 변수를 최종적으로 선정한 후 데이터마이닝 학습 알고리즘을 제품 품질에 영향을 평가할 수 있는 학습모델을 최종적으로 도출하였다. 여기서 데이터마이닝 학습모델 결과에서 나타난 품질 특성 변수 및 관계를 적용하여 RSM 인자를 약 1/8로 줄였으며 최종 실험의 횟수를 최초 실험에 비하여 약 1/12로 줄였다. 최종적으로 도출된 RSM 모델을 통해 양품 기준 제품 치수를 만족하는 최적 생산 조건의 공정변수 특성치를 제시하였다.

국가연구개발사업의 학술적 성과의 시차효과에 관한 실증적 연구

정병호(Byung-Ho Jeong),천강민(Kang-Min Cheon),양재경(Jaekyung Yang) 한국산업경영시스템학회 2012 한국산업경영시스템학회지 Vol.35 No.1

This study examines the relationship between R&D investment and subsequent outputs of the research activity. Usually, there is some time difference between the production of research outputs, such as academic papers and application or registration of patents, and the investment of R&D expenditure. The time lag for producing this kind of research outputs should be considered to evaluate the performance of research activity exactly. The purpose of this study is to identify time lag effect between the times of input and output of a R&D activity and to derive the degree of time lag using the data set of a long term R&D program supported by Korean government. A modified Almon model is suggested to identify the time lag effect between input and output of research activities performed by this program. Time-series cross-section data from 16 research centers between 2001 and 2009 are used to find time lag effect.

LSR과 EPDM 고무의 노화 전 · 후의 거동 분석 및 평가

박수현(Soo-Hyun Park),전형렬(Hyeong Ryeol Jeon),주우정(Woo-Jeong Joo),천강민(Kang-Min Cheon),허장욱(Jang-Wook Hur) 한국기계가공학회 2021 한국기계가공학회 춘추계학술대회 논문집 Vol.2021 No.12