Prediction of operating time of the nuclear power plant : multiple linear regression analysis = 원자력발전소의 가동시간 예측 : 다중선형회귀분석 multiple linear regression analysis

In Reliability/Availability(R/A) analysis, operating time prediction plays one of the major key roles to availability forecasts which are based solely on the past operating experience. Statistical prdiction of operating time of the nuclear reactor and the knowledge of the interrelationships betwen the explanatory variables and the response variable are essential to the evaluation of availability of the existing nuclear power plant. In addition, these informations are usdful for an evaluation of maintenance policy, for component replacement policy and for readiness of the nuclear power plant. Multiple linear regression analysis enables a decision maker to predict future operating times under given conditions. In addition, it is possible to assess various factors influencing nuclear power plant operating characteristics. In this paper, it is investigatexplanatory variables an the response variable form the operating history data and seven multiple linear regression models are developed in sequence to see the trend of model explanability. This approach was verified over the past data and compared with time series analysis. The results showed that until observation number around 60, it is well predicted but thereafter the deviation increased. The correlation coefficient between the observed and the predicted values was 0.6992. This implies tht multiple linear regression analysis is an effective approach to operating time prediction.
Advisors
Lee, Byong-Whiresearcher이병휘researcher
Publisher
한국과학기술원
Issue Date
1990
Identifier
67585/325007 / 000881037
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 핵공학과, 1990.2, [ v, 60 p. ]

Keywords

원자로 운전

URI
http://hdl.handle.net/10203/49252
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=67585&flag=t
Appears in Collection
NE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.
  • Hit : 166
  • Download : 0
  • Cited 0 times in thomson ci

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0