Proceedings of the Eighth International Conference on Probabilistic Safety Assessment & Management (PSAM)
205 The Use of the GO-FLOW Methodology to Investigate the Aging Effects in Nuclear Power Plants (PSAM-0113)
Download citation file:
- Ris (Zotero)
- Reference Manager
Japan has been building nuclear power plants since 1966 and it has been 30 years since the safety assessment for the first plants was completed. The maintenance rules for the existing Japanese nuclear power plants have proven to be satisfactory and have led to reliable plant operation. While most individual component availability for the existing plants can be also maintained in the future through continued periodic surveillance, plant availability may decrease due to the aging. Aging effects for nuclear power plant components have been investigated by various probabilistic risk assessment studies, most of the studies encountered in the literature use a static unavailability. However, the static unavailability was smaller than the peak unavailability calculated from a time-dependent model. The use of a time-dependent technique was also necessary in order to be able to modify surveillance periods as a function of predicted plant unavailability, in a similar manner to the use of risk monitors today to modify plant configuration. If the unavailability of each component was decreased by shortening their surveillance intervals, the plant availability could be improved. This study used the Emergency Core Cooling System (ECCS) of BWR as an example system to investigate the feasibility of such approach. To evaluate the system availability for the ECCS, the GO-FLOW methodology was particularly suitable because it could model complex sequences in system operation. However, the GO-FLOW methodology had not yet been implemented in situations where failure rate changes depending on aging effects. Therefore, an application to substitute data of time-dependent unavailability for GO-FLOW was developed in this study. The unavailability of aging components was obtained by using time-dependent technique. According to the some trial calculations, the availability of ECCS after 30 years increased to the level before aging starts, if the surveillance of all components of ECCS was shortened from12 months to 9 months at 30 years.