Fire Probabilistic Safety Assessment Research Articles

Fire probabilistic safety assessment of control room in TRIGA PUSPATI research reactor

An internal Fire Probabilistic Safety Assessment (FPSA) study was conducted in the control room of the TRIGA PUSPATI (RTP) research reactor. This study aims to determine the likelihood of a fire occurring in the control room of RTP and the probability of fire occurring after an accident loss of instrumentation and control (LOIC) to the core damage frequency (CDF) value. The quantification results of the CDF are compared between LOIC with fire-induced (LOIC-F) and without fire-induced (LOIC). A methodical strategy to identify potential fire hazards that includes the precise identification of the design area, the size of any opening compartments, the location of electrical cabinets and fire suppression systems, as well as the placement of any safety-related equipment was applied. Included is Failure Mode and Effect Analysis (FMEA), which determines failure modes and designates repercussions for each scenario. This research will contribute on providing essential information on status of fire in RTP’s control room as an insight of identifying any risk of fire scenario that may occur including investigation of current fire protection system sufficiency. In addition, a core damage end state of consequences of fire after an LOIC is also presented. The result of this research can be beneficial and creating awareness to the management as well as reactor operators in case of fire in the control room of RTP.

Practical modeling and quantification of a single-top fire events probabilistic safety assessment model

In general, an internal fire events probabilistic safety assessment (PSA) model is quantified by modifying the pre-existing internal event PSA model. Because many pieces of equipment or cables can be damaged by a fire, a single fire event can lead to multiple internal events PSA initiating events (IEs). Consequently, when the fire events PSA model is quantified, inappropriate minimal cut sets (MCSs), such as duplicate MCSs, may be generated. This paper shows that single quantification of a hypothetical single-top fire event PSA model may generate the following four types of inappropriate MCSs: duplicate MCSs, MCSs subsumed by other MCSs, nonsense MCSs, and MCSs with over-counted fire frequencies. Among the inappropriate MCSs, the nonsense MCSs should be addressed first because they can interfere with the right interpretation of the other MCSs and prevent the resolution of the issues related to the other inappropriate MCSs. In addition, we propose a resolution process for each of the issues caused by these inappropriate MCSs and suggest an overall procedure for resolving them. The results of this study will contribute to the understanding and resolution of the inappropriate MCSs that may appear in the quantification of fire events PSA models.

Study on Importance Analysis and Uncertainty Analysis of Fire Probabilistic Safety Assessment in Domestic Nuclear Power Plants

In Korea, the first of fire Probabilistic Safety Assessment (PSA) was conducted for Hanbit nuclear power plant in 1992. The fire probabilistic safety assessment of a domestic power plant was applied to the methodology and data presented in TR-105928 published in 1995 by EPRI, which separately addresses accident scenarios and conditional core damage frequency. In terms of importance analysis, they performed only a rough assessment of fire area and detailed scenarios that can have a significant impact on the Core Damage Frequency (CDF), but they do not perform uncertainty analysis. In this paper, we present a method for performing importance analysis and uncertainty analysis through a combination of the Minimum Cut Set (MCS) resulting in core damage to nuclear power plants and basic events that form an accident scenario. Based on this, we also presented insights into the basic events that were significantly derived from the fire event probabilistic safety assessment through analysis of switchgear rooms in reference nuclear power plants, and conducted a uncertainty analysis of the CDF. This paper is expected to be available for future Risk-Informed Decision Making (RIDM).

Behaviour of electrical cables under fire conditions

Abstract A Fire Probabilistic Safety Assessment – called the Fire PSA – is being carried out by the French Institute of Radiological Protection and Nuclear Safety (IRSN) to be used in the framework of the safety assessment of operating 900 MWe PWRs. The aim of this study is to evaluate the core damage conditional probability which could result from a fire. A fire can induce unavailability of safety equipment, notably damaging electrical cables introducing a significant risk contributor. The purpose of this paper is to present the electrical cable fire tests carried out by IRSN to identify the failure modes and to determine the cable damage criteria. The impact of each kind of cable failure mode and the methodology used to estimate the conditional probability of a failure mode when cable damage occurred is also discussed.

Quantitative Analysis of Valve Contribution to the Failure Probability of Carbon Dioxide Suppression System

Carbon dioxide suppression systems are used in nuclear power plants to extinguish oil fires and ensure integrity of critical equipment. In this study, the contributions of the valves in the carbon dioxide suppression system to the failure probability of suppression were quantitatively analyzed, and the failure probability of the fire suppression system applied to the fire Probabilistic Safety Assessment (PSA) was evaluated for appropriateness. Then, a reliability evaluation model was developed in the form of a fault tree, and the reliability data were analyzed for the major component. The failure probability of the carbon dioxide suppression system with early air-release equipment was assessed as 6.22 × 10<sup>-3</sup>/demand, which is 2.67 times that of the system without such equipment. The results were attributed to the addition of the diverted line isolation valves and quick exhaust valve to release residual air into the atmosphere before supplying carbon dioxide to the fire area. It was also confirmed that the failure probability of the carbon dioxide suppression system with the early air-release equipment was 15.6% that of the suppression provided by the fire PSA and that the failure probability of suppression by the fire PSA was conservative. There are no reported guidelines in literature for analyzing the reliabilities of carbon dioxide fire suppression systems, and the fire PSA currently use the failure probability of suppression recommended by the NSAC-179L.

Internal fire probabilistic safety assessment preliminary work at research reactor TRIGA PUSPATI

An internal fire analysis Level-I probabilistic safety assessment (PSA) for TRIGA PUSPATI (RTP) nuclear research reactor was initiated. Previous study on half-scope mode Level-I PSA, focusing only on internal initiating event, were developed for both full power and shutdown operational mode. Hence, in ensuring PSA remains relevant, a subsequent study to kept PSA up to date is a compulsory. The aim of this study is to present the step-by-step work done and the insights gained in identifying and screening significant compartments that are potential to had the worst consequences as an ignition sources. This preliminary study will only focus on RTP’s basement which consist of 3 compartments: switch room, battery room and basement hall. Procedure in data collection were based on IAEA document, Specific Safety Guide No. 3: Development and Application of Level 1 Probabilistic Safety Assessment for Nuclear Power Plants (SSG-3). Microsoft Excel was used as data collection media and internal fire hazard database development. A brief qualitative analysis using failure mode and effect analysis is also presented.

Construction and quantification of a fire probabilistic safety assessment model using support system initiating event fault trees

This study investigated the effect of the fire-induced initiating event fault trees (IE FTs) related to the support systems of a reference plant, Hanul 3,4, on the risk quantification results by comparing the results of two (2) different kinds of fire probabilistic safety assessment (PSA) models: one based on the FT approach (with IE FTs) and the other based on the conventional approach (without IE FTs). The results show that the FT modeling of fire-induced IEs decreases fire-induced core damage frequencies (CDFs) if the IE FTs involve relatively many AND logic gates, because the fire-induced component failure events, the probabilities of which are less than '1′(TRUE), constitute the fire-induced IE FTs. The results also show that the CDFs increase if the fire-induced IE FTs involve relatively many OR logic gates. This is because a rare event approximation leads to overestimation of the real fire-induced risks.

Study on construction of loss of component cooling water system initiating event fault tree for fire event PSA

ABSTRACT In this study, the construction of the loss of component cooling water system (LOCCWS) initiating event (IE) fault tree (FT) for an actual fire event probabilistic safety assessment (PSA) model of the Korean reference nuclear power plant considering only IE initiators was validated. The quantification results of the LOCCWS accident sequences obtained using an LOCCWS IE FT model with only initiators are similar to that with initiators and enabling events. This confirmed that the LOCCWS IE FT for an actual fire event PSA model could be constructed by considering only IE initiators. In addition, the same LOCCWS accident sequences were quantified assuming that fire triggering only the LOCCWS IE leads to reactor shutdown. Compared with the quantification result obtained based on the assumption that any fire included in the fire event PSA leads to reactor shutdown, the core damage frequency (CDF) can be reduced. Thus, it can be concluded that there is a possibility of underestimation of CDF when the LOCCWS IE FT model with only initiators is used and the assumption that fire triggering only the LOCCWS IE results in reactor shutdown is employed for the quantification of LOCCWS accident sequences.

Comparative study on the construction of support system initiating event fault trees for a fire probabilistic safety assessment

This study compared two different fire-induced support system initiating event fault tree (SSIE FT) models: one with only fire IE initiators, and the other with both fire IE initiators and enabling events. Enabling events place a system in a state such that it transitions from the operating state to the failed state when the initiating event occurs. To develop the fire event probabilistic safety assessment (PSA) models for these two different SSIE FT models, hypothetical internal event accident scenarios were used based on the assumption that a fire included in the fire event PSA causes a reactor shutdown. The results obtained indicate that SSIE FT models with only fire IE initiators, in terms of quantification results of fire event PSA models, are equivalent to those with both fire IE initiators and enabling events under the assumption that a fire included in the fire event PSA results in a reactor shutdown. For the case where a fire in a nuclear power plant (NPP) does not lead to a reactor shutdown, on the other hand, this study showed that SSIE FT models with only fire IE initiators might underestimate the real fire-induced risk, whereas SSIE FT models with both fire IE initiators and enabling events could lead to reasonable quantification results for risk due to fires.

Development and assessment of fire-related risk unavailability matrices to support the application of the maintenance rule in a PWR nuclear power plant

ABSTRACTTwo methods are presented which serve to incorporate the fire-related risk into the current practices in nuclear power plants with respect to the assessment of configurations. The development of these methods is restricted to the compulsory use of fire probabilistic safety assessment (PSA) models. The first method is a fire protection systems and key safety functions unavailability matrix which is developed to identify structures, systems, and components significant for fire-related risk. The second method is a fire zones and key safety functions (KSFs) fire risk matrix which is useful to identify fire zones which are candidates for risk management actions. Specific selection and quantification methodologies have been developed to obtain the matrices. The Monte Carlo method has been used to assess the uncertainty of the unavailability matrix. The analysis shows that the uncertainty is sufficiently bounded. The significant fire-related risk is localized in six KSF representative components and one fire protection system which should be included in the maintenance rule. The unavailability of fire protection systems does not significantly affect the risk. The fire risk matrix identifies the fire zones that contribute the most to the fire-related risk. These zones belong to the control building and electric penetrations building.

Fire risk analysis based on one-dimensional model in nuclear power plant

Fire probabilistic safety assessment (fire PSA) is developed to give the insight of nuclear power plant risk induced by fire accident and the main contributors, and fire accident scenario analysis is one of the important parts in the work to get the key factors such as time to target damage and time to detection. The thermal–hydraulic model simulating fire accident should have enough accuracy and high speed to satisfy the request of fire PSA. Many researches indicate that in a large volume the hot fluid will concentrate in the upper part while the cold fluid will be in the lower part because of density difference under fire condition, that is, the gradients of temperature and of some other parameters in vertical direction are much greater than in horizontal direction. Based on such thermal stratification theory a one-dimensional model is developed, and the buoyant jet is used to simulate the process of heated air flowing up. In this paper a fire in a compartment is analyzed based on one-dimensional model and the temperature distribution is obtained, the results are compared with those of commercial software such as FDT, CFAST and FDS, then the time to target damage is evaluated based on results of different models and the fire non-suppression probabilities are evaluated. The results illustrate that one-dimensional model has better accuracy than FDT and CFAST since such a model can simulate the thermal stratification and natural circulation which exist in the volume simultaneously. Moreover when the fire power is low, the thermal stratification is apparent and air temperature in the hot upper layer is much lower than the critical value of target damage, the one-dimensional model has enough accuracy to be used directly in fire PSA. While the thermal stratification will be weakened when the fire power increases because of the effects of radiation heat transfer and the entrainment by the jet, so more detailed model such as FDS is needed for such situations, however the results of one-dimensional model can give the advice for the proper simulation time of FDS to improve the calculation efficiency when the upper part temperature is close to or higher than the critical value.

Development of the IPRO-ZONE for internal fire probabilistic safety assessment

Fire probabilistic safety assessment (PSA) analysts have been manually determining fire-induced component failure modes and modeling them into PSA logics. These are time-consuming tasks, and may be difficult as they require a lot of information and many events should be modeled. KAERI has been developing the IPRO-ZONE (interface program for constructing zone effect table) to facilitate fire PSA works for identifying and modeling fire-induced component failure modes, and to construct a one-top fire event PSA model. A one-top fire event PSA model is automatically constructed with the output of the IPRO-ZONE, the AIMS-PSA, and internal event one-top PSA model. The output of the IPRO-ZONE includes information on fire scenarios, fire propagation areas, equipment failure modes affected by a fire, internal PSA basic events corresponding to fire-induced equipment failure modes, and the fire events to be modeled. This paper introduces the IPRO-ZONE, and its application results to fire PSAs for Ulchin Unit 3 and SMART (System-integrated Modular Advanced Reactor).

The Performance-Based Fire Protection in the Nuclear Power Plant Design

The process of the performance-based fire protection method is introduced. And the method of the fire probabilistic safety assessment (PSA) and the fire protection deign in the nuclear power plant are described. The combination of the performance-based fire protection and the fire protection deign in the nuclear power plant are discussed, and the fire PSA will be used to validate to improve the design and the fire safety level in the nuclear power plant. The process of the combined method is also described. Although there are some improvements in the future, but the combination is possible and useful in the nuclear power plant fire protection design. The advantages and disadvantages of the combination are also discussed.

An approach to the construction of a one top fire event PSA model

An internal fire event probabilistic safety assessment (PSA) model has been generally quantified by modifications of a pre-developed internal events PSA model. New accident sequence logics not covered in the internal events PSA model are separately developed to incorporate them into the fire PSA model. Previous studies on the changes of the one top internal event PSA model for the one top fire event PSA model have been limited to the equipment failures affected by the fire. In addition, they assumed that the probabilities of basic events associated with equipment or cables impacted by the fire are one. However, the probabilities of spurious operation events and human failure events affected by the fire might not be estimated as one. In this study, new modification rules were proposed for the construction of a one top PSA model for fire events by using a one top internal event PSA model. The proposed new modification rules can be applied to all the fire damage events for the fire-induced equipment failure events and spurious operation events, human error events impacted by a fire, regardless of whether they are estimated as one or not. Applications of the proposed modification rules to the compartment and scenario-level fires for the hypothetical plants were performed for demonstrating their appropriateness to the changes of the one top internal event PSA model to the one top fire event PSA model. In addition, quantification procedure with the one top fire event PSA model was presented and discussed.

Development of a new quantification method for a fire PSA

For an internal fire analysis, fire scenarios are developed carefully and quantified in a sequential and iterative way in a traditional fire Probabilistic Safety Assessment (PSA). However, there has been no proven explicit method to avoid these iterative quantifications till now. This study presents the Jung's Single Top And Run (JSTAR) method that facilitates a simultaneous single quantification of all fire scenarios. The JSTAR method could be employed at the fire PSA phases of a quantitative screening or detailed analysis. Using the JSTAR method, accurate fire risks of a fault tree that has many negates could be calculated by avoiding the frequent house event propagations of the fire scenario conditions. Furthermore, the proposed JSTAR method is a simple and explicit method to build a single-top external event PSA model for a risk-monitoring system. The JSTAR method could be implemented easily by developing a small automatic conversion tool. Depending on the maintenance policy of a fire PSA model, a single-top fire PSA model that is created by the conversion tool could be maintained permanently or it could be temporarily generated and discarded. The use of the JSTAR method is recommended for all external event PSAs such as an internal flooding risk analysis.

Insights from fire PSA for enhancing NPP safety

This paper presents the findings of an effort to gain insights from fire probabilistic safety assessment (FPSA) conducted in nuclear power plants. Using probabilistic models, the fire PSA takes into account the possibility of a fire at specific plant locations and its propagation, detection and suppression of the fire; and also helps to assess the effect of the fire on safety-related cables and equipment. The results of FPSA contributed to design modifications in plant to enhance the safety and thereby reduce its contribution to core damage frequency. It also highlights the sources of uncertainty while conducting and suggesting values of risk parameter in FPSA study.

Fire PSA of the French PWR 900 MWe series

Abstract The Institute for Radioprotection and Nuclear Safety (IRSN) has developed a Fire Probabilistic Safety Assessment of the operating French 900 MWe PWR in order to organize into a hierarchy the items of equipments and the compartments, impacting the core melt frequency following a fire, and to reinforce deterministic safety analysis with functional analysis and realistic fire scenarios. The IRSN Fire PSA is composed of two stages: the first one concerns the selection of critical zones and the second one concerns the detailed study of the critical compartments, corresponding to the development of the fire scenarios and of the fire-induced core damage sequences. The first version of the IRSN Fire PSA has been updated by integrating the specificities of the Fire Action Plan, implemented by the utility on the French 900 MWe PWR. This study shows that four compartments are preponderant. The integration of the Fire Action Plan allowed decreasing the number of preponderant compartments. The integration of the symptom-oriented procedures and the removal of conservatisms should lead to a lower core damage frequency for these preponderant compartments. The results show that fire is not a negligible risk but the implementation of the Fire Action Plan has a significant positive impact on safety.

Fire Probability Safety Analysıs in France for 900 MWe nuclear power plants / Probabilistische Sicherheitsanalyse für Brandschutz (Brand PSA) bei französischen 900-MWe-Druckwasserreaktoren

This paper describes the methodology implemented by the Institute for Nuclear Safety and Protection (IPSN) to carry out the Fire Probabilistic Safety Assessment (Fire PSA) for French 900 MWe pressurised water reactors. The initial results obtained are presented. Additional research and development activities are indicated which IPSN carried out or decided to perform in order to reduce the amount of uncertainty associated with the data or to confirm hypotheses that can impact significantly the study results.