DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Seong, Poong-Hyun | - |
dc.contributor.advisor | 성풍현 | - |
dc.contributor.author | Kim, Seung-Geun | - |
dc.contributor.author | 김승근 | - |
dc.date.accessioned | 2015-04-23T07:10:04Z | - |
dc.date.available | 2015-04-23T07:10:04Z | - |
dc.date.issued | 2014 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=592361&flag=dissertation | - |
dc.identifier.uri | http://hdl.handle.net/10203/197293 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 원자력및양자공학과, 2014.8, [ vi, 60 p. ] | - |
dc.description.abstract | To protect nuclear power plants (NPPs) from dangerous situations, operators are trained for various transient types and many procedures are prepared. However, operators could commit mistakes and make wrong decisions since there are so many kind of plant status variables to check and working memory is limited. In order to solve this kind of problem, many kind of operation support systems (OSSs) such as fault diagnosis system and computerized procedure system were developed. Although the transient mitigation failure probability is very low, still there is a possibility of failure that leads to severe accident occurrence. Also, recent occurrence of Fukushima accident shows that severe accident could happen even if there are many safety related systems. Therefore, support systems related to monitoring and prediction of severe accident are necessary in order to enhance the safety of NPPs. However, the researches about support systems which conduct severe accident monitoring and prediction are not sufficient. In this research, occurrence time of several events that could happen in the process of severe accident (time when maximum core temperature exceeds 1200℃, reactor vessel failure time, containment failure time) were predicted by using support vector classification (SVC) and support vector regression (SVR). And for preliminary step, break location and break size of loss of coolant accident (LOCA) were identified. SVC and SVR are included in support vector machine (SVM), which is a machine-learning algorithm that has been successfully used for solving classification problem and conducting regression analysis. Firstly, break location of LOCA is identified by using SVC which is capable to classifies cold leg LOCA and hot leg LOCA with 13 kinds of plant status variables (broken side steam generator pressure, level, and temperature, unbroken side steam generator pressure, level, and temperature, pressurizer pressure and level, core water temperature and... | eng |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Support Vector Classification | - |
dc.subject | 냉각재 상실 사고 | - |
dc.subject | 중대사고 | - |
dc.subject | 서포트 벡터 회귀 | - |
dc.subject | 서포트 벡터 분류 | - |
dc.subject | Loss of Coolant Accident | - |
dc.subject | Support Vector Regression | - |
dc.subject | Severe Accident | - |
dc.title | Severe accident occurrence time prediction by using support vector classification and support vector regression | - |
dc.title.alternative | 서포트 벡터 분류 및 서포트 벡터 회귀를 이용한 중대사고 진입시간 예측 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 592361/325007 | - |
dc.description.department | 한국과학기술원 : 원자력및양자공학과, | - |
dc.identifier.uid | 020133115 | - |
dc.contributor.localauthor | Seong, Poong-Hyun | - |
dc.contributor.localauthor | 성풍현 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.