Evaluating the uncertainty bound of the multiple scenario CDF (Core Damage Frequency) by the distribution sampling of the basic events.
Probabilistic Risk Assessment (PRA) developed in order to investigate the factors that cause threat to nuclear power plants and improve the safety of nuclear power plants. (eg seismic, flooding, multiple components failure, human error). As a result, in addition to the Core Damage Frequency (CDF) obtained by probabilistic analysis, the main reason leading to the core meltdown and their importance can also be known. In some possible hazards (such as floods, fires, etc.), multiple scenarios may be used to develop and analyze the safety assessment of the nuclear power plant, and then the CDFs calculated from multiple scenarios will be summed up as the CDF of the power plant in these scenarios. However, this value is only a point estimate with unknowing uncertainty bound. In order to obtain more statistically significant analysis results, this paper uses the method of randomly sampling basic event values to quantify multi-scenario events, in order to obtain the CDF with statistically significant information.
A PSAM Profile is not yet available for this author.
Paper 2 FR84
Lead Author: Francesco Di Dedda Co-author(s): Anders Olsson, anders.olsson@vysusgroup.com
Thomas Augustsson, thomas.augustsson@okg.uniper.energy
PSA implementation of the Independent Core Cooling and new EOPs/SAMGs at Oskarshamn 3
An independent core cooling function (OBH) has been installed in unit 3 at the Oskarshamn NPP as one of the post-Fukushima actions. In conjunction with the OBH project an update of the Emergency Operating Procedures (EOPs) has been made and a set of the Severe Accident Management Guidelines (SAMGs) has been developed. A complete review and validation of both level 1 and level 2 PSA sequences for all operating modes was performed to achieve the goal of implementing the new system functions and new procedures in the full scope analysis.
Given the new system configuration, requirements on some systems affected the requirements on dependent systems functions. This interaction between different systems was challenging to assess in specific cases. Dedicated workshop activities with operating personnel were crucial to determine accident progression, especially when considering the new system functions and the manual actions according to the new EOPs/SAMGs. Changes in the electrical systems as well as the adoption of a new version the integral code MAAP and a new version of RiskSpectrum PSA software were also undertaken in the implementation.
The modelling detail and realism has generally been increased in the "base-line PSA" (full power level 1 and 2, internal events) while other assumptions and simplifications are still present for other operating modes.
The presentation will focus on the encountered challenges, how those were solved and on giving an overview on the updated results.