The operation of a passive system is more sensitive to initial and boundary conditions with respect to an active system with a similar mission; this is due to the dependence of the system upon the driving thermal-hydraulic phenomena (e.g. natural circulation). Moreover, during the start-up and the operation phases, the system passes through several different operating conditions that may in turn affect its behavior. For these reasons, the evaluation of the reliability of a passive system needs specific methodologies that consider also the reliability of the involved thermal-hydraulic phenomena. The objective of the activity carried out by UNIPI and POLITO, summarized in this section, is to show the results of the application of an assessment method, consistent with the technology available nowadays, considering not only the behavior of a stand-alone passive system as mostly done in the past, see e.g. D’Auria & Galassi, 2000, but also its interaction with the reactor system during the evolution of a selected DBA. The activity aims to show the capability of the procedure rather than to assess the reliability of the system. The elements for the proposed activity are presented here after: 1. Method to evaluate the reliability of passive systems – A simplified application of the REPAS methodology was pursued. Only deterministically selected cases, so called "extreme cases” have been performed. Examples of full application of the methodology, which include also cases selected by a probabilistic analysis, are described in D’Auria et al., 2000, and IAEA, 2014b. 2. Best estimate SYS-TH code – The well-known RELAP5/MOD3.3 code, RELAP5/MOD3.3, 2003, has been selected. 3. Qualified model of a passive system – A TH model of a passive system, which has the key characteristic of a typical Isolation Condenser, has been derived and validated based on the PERSEO benchmark activity carried out in the framework of this Task Group. The configuration of this passive system has some differences from PERSEO (described in section 4.2.2.5 and Appendix 3). Namely the heat exchanger is submerged under water as initial condition and the activating valve is located on the line connecting the system with the RPV. However, the result of the benchmark calculations can be used to validate the heat transfer model of the RELAP5 code for the considered heat exchanger geometry in similar operating conditions. 4. Model of a full-scale reactor system – A model of an SBWR reactor developed by UNIPI Breghi et al., 1992, and Barbucci et al., 1995, has been used to derive an input deck suitable for the adopted release of the RELAP5 code. It must be mentioned that, even though the model was developed following a rigorous procedure, e.g. see Bonuccelli et al., 1993, a full validation was not possible. 5. A scaling method – The full-scale model of the SBWR has been scaled following a procedure consistent with the OECD/NEA-CSNI/WGAMA State of the Art Report on Scaling, OECD/NEA/CSNI, 2017. The original model has been downscaled by a factor 6 following the power-to volume method. This was needed because, for simplicity, the analysis has been conducted considering only one heat exchanger module over the six that constitute the 3 independent trains available in the reference reactor, USNRC, 2016, or NUREG/CR-6309. In element 1, “extreme cases” are transient scenarios for the concerned passive systems: the related probability of occurrence (PM in Chapter 1) is not established, though it is expected to have a ‘low’ value and to fall in a region outside typical PSA studies; the related evolution (transient conditions) may cover parameter space regions not covered by typical DSA studies and may challenge the target function of the concerned passive system (TM in Chapter 1). Boundary conditions for extreme cases are fixed based on the expertise of scientists performing reliability analysis. The results from extreme cases analyses shall contributeto the overall value of reliability determined for the passive system, see e.g. D’Auria & Galassi, 2000. The analysis of extreme cases also contributes to the deep understanding of the passive system performance.
Reliability of Passive Systems: The 'extreme' cases analysis
D'Auria F.
Ultimo
Supervision
2019-01-01
Abstract
The operation of a passive system is more sensitive to initial and boundary conditions with respect to an active system with a similar mission; this is due to the dependence of the system upon the driving thermal-hydraulic phenomena (e.g. natural circulation). Moreover, during the start-up and the operation phases, the system passes through several different operating conditions that may in turn affect its behavior. For these reasons, the evaluation of the reliability of a passive system needs specific methodologies that consider also the reliability of the involved thermal-hydraulic phenomena. The objective of the activity carried out by UNIPI and POLITO, summarized in this section, is to show the results of the application of an assessment method, consistent with the technology available nowadays, considering not only the behavior of a stand-alone passive system as mostly done in the past, see e.g. D’Auria & Galassi, 2000, but also its interaction with the reactor system during the evolution of a selected DBA. The activity aims to show the capability of the procedure rather than to assess the reliability of the system. The elements for the proposed activity are presented here after: 1. Method to evaluate the reliability of passive systems – A simplified application of the REPAS methodology was pursued. Only deterministically selected cases, so called "extreme cases” have been performed. Examples of full application of the methodology, which include also cases selected by a probabilistic analysis, are described in D’Auria et al., 2000, and IAEA, 2014b. 2. Best estimate SYS-TH code – The well-known RELAP5/MOD3.3 code, RELAP5/MOD3.3, 2003, has been selected. 3. Qualified model of a passive system – A TH model of a passive system, which has the key characteristic of a typical Isolation Condenser, has been derived and validated based on the PERSEO benchmark activity carried out in the framework of this Task Group. The configuration of this passive system has some differences from PERSEO (described in section 4.2.2.5 and Appendix 3). Namely the heat exchanger is submerged under water as initial condition and the activating valve is located on the line connecting the system with the RPV. However, the result of the benchmark calculations can be used to validate the heat transfer model of the RELAP5 code for the considered heat exchanger geometry in similar operating conditions. 4. Model of a full-scale reactor system – A model of an SBWR reactor developed by UNIPI Breghi et al., 1992, and Barbucci et al., 1995, has been used to derive an input deck suitable for the adopted release of the RELAP5 code. It must be mentioned that, even though the model was developed following a rigorous procedure, e.g. see Bonuccelli et al., 1993, a full validation was not possible. 5. A scaling method – The full-scale model of the SBWR has been scaled following a procedure consistent with the OECD/NEA-CSNI/WGAMA State of the Art Report on Scaling, OECD/NEA/CSNI, 2017. The original model has been downscaled by a factor 6 following the power-to volume method. This was needed because, for simplicity, the analysis has been conducted considering only one heat exchanger module over the six that constitute the 3 independent trains available in the reference reactor, USNRC, 2016, or NUREG/CR-6309. In element 1, “extreme cases” are transient scenarios for the concerned passive systems: the related probability of occurrence (PM in Chapter 1) is not established, though it is expected to have a ‘low’ value and to fall in a region outside typical PSA studies; the related evolution (transient conditions) may cover parameter space regions not covered by typical DSA studies and may challenge the target function of the concerned passive system (TM in Chapter 1). Boundary conditions for extreme cases are fixed based on the expertise of scientists performing reliability analysis. The results from extreme cases analyses shall contributeto the overall value of reliability determined for the passive system, see e.g. D’Auria & Galassi, 2000. The analysis of extreme cases also contributes to the deep understanding of the passive system performance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
