American Nuclear Society
Home

Home / Store / Journals / Electronic Articles / Nuclear Technology / Volume 161 / Number 1 / Pages 1-7

Reliability Prediction of Passive Systems Based on Bivariate Probability Distributions

Luciano Burgazzi

Nuclear Technology / Volume 161 / Number 1 / Pages 1-7

January 2008

Format:

Price:$30.00
Member Price:$27.00
Member Savings:$3.00

The focus of the present study is passive system reliability assessment, with reference to the thermal-hydraulic passive systems (i.e., relying on natural circulation). An approach based on system-relevant performance parameters is introduced to provide system-significant availability and reliability figures, within a reliability physics framework.

The method exploits the fact that for thermal-hydraulic passive systems to perform as expected to accomplish the required mission, parameters must lie between certain limits according to defined safety criteria. Some relevant physical parameters are worth considering as significant indicators of thermal-hydraulic passive system performance, for instance coolant flow or exchanged thermal power. Within this methodology, the selected representative parameters defining the system performance are properly modeled through the construction of joint probability functions in order to assess the correspondent functional reliability. The application of the methodology to a realistic passive system design is illustrated.

The results are shown to point out the relevance of the passive system functional reliability aspects with respect to the classical mechanical component malfunctions, serving as a foundation for continuous improvement of the passive system reliability assessment process.

This paper aims to remedy some of the limitations following on from applying the functional reliability approach to the passive system reliability problem, as highlighted in an earlier paper [Nuclear Technology, Vol. 144, p. 145 (Nov. 2003)]. This concerns essentially the assumption of independence between the marginal distributions to construct the joint probability distributions to evaluate system reliability.

 
 
 
Questions or comments about the site? Contact the ANS Webmaster.
advertisement