Probabilistic Safety Assessment (PSA) Of A Reference 10 Mw Water-Water Research Reactor (VVR) with Emphasis on Nuclear Safety Application

Abstract

Fault trees (FT) and event trees (ET) are widely used in industries to model and calculate the reliability of safety systems. Detailed analysis in nuclear installations require the combination of these two techniques. This research work uses FT (fault tree) and ET (Event Tree) to study PSA (Probabilistic Safety Assessment) in a 10 MW Russian water water research reactors (VVR). This dissertation focuses studies on level 1 PSA intent on the search for and the acquisition of knowledge for consolidation of methodologies for future studies of reliability. The reference 10 MW Water Water research reactor was used as a case example. Seven initiating events (IEs) were chosen for this work and an example of the IEs is LOCA (Loss of Coolant Accident) and from there the possible consequences of accidents (sequence of events) which could cause damage to the core were developed, using ET. In addition, for each of the affected systems involved in the accident, FTs were built and used to calculate the probability of each event top of the FT. Estimates of the importance of basic events we re also presented. This research was conducted using the commercial computational tool SAPHIRE suit code. The idea was to assess the frequency of the reactors hazard states fir st and see whether or not the range 1.0E-5/ yr to 1 .0E-4/ yr can be met. In this case, it can certainly be concluded that the smaller frequency of damage states is adequate and ther efore results obtained for performance or failure of the systems analyzed, were considered satisfactory.