CANDU Fire Probabilistic Risk Assessment

Abstract

Advancement of fire risk analysis methods has led to a widespread development of detailed fire probabilistic risk assessments (PRA) at nuclear power plants. The Fire PRA assesses the possibility of a fire at critical plant locations and evaluates the fire damage. Fire PRA also evaluates the effect of the fire on safety-related cables and equipment. The scope of the Fire PRA is limited to demonstrating that the fire safe shutdown objectives and performance criteria are met. Hence, the Fire PRA is only used for plant areas where fires may have a potential impact on systems, structures, and components (SSCs) that are required to perform the fire safe shutdown functions. A CANDU Fire Database was developed by the Canadian Nuclear Safety Commission (CNSC) to collect and maintain data of all CANDU related fires. There are 19 nuclear power plants operating in Canada today, with 75 fires reported from 1981 to 2020. The main objective of this thesis is to develop a Fire PRA for Canadian CANDU nuclear reactors. For this, a fuel load survey of all 1,230 Fire Safe Shutdown System (FSSA) rooms in CANDU reactors in Canada was carried out, and found that the average fuel density for the 1,230 general fire zones was 170.1 MJ/m2, and electric faults were the most likely ignition source. The results of the fuel load survey were used to group fire zones based on their areas, heights and combustible loads and to produce a list of critical fire scenarios. Fire Dynamics Simulator (FDS) was used to simulate these critical fire scenarios and the associated consequences on FSSA cables. The fire scenarios' simulation results were used to create a qualitative screening method (decision tree) for CANDU reactors, which is an essential step in the CANDU Fire PRA. A CANDU Fire PRA methodology for CANDU reactors was developed, and two fire zones were selected to demonstrate the use of the CANDU Fire PRA methodology. In addition, High Energy Arc Fault (HEAF) risks in CANDU reactors were examined and analyzed, and recommendations were given to mitigate the risks and consequences of any potential HEAF events.