Severe Accident Phenomena Research Articles

00/02535 Analyses of loss of vacuum accident (LOVA) in ITER

The aim of this work is the comparative study of hydrogen generation and the associated parameters in a simulated severe accident of a short-term station blackout (STSBO) in a typical BWR-5 with Mark-II containment. MELCOR (v.1.8.6) and SCDAP/RELAP5 (Mod.3.4) codes were used to understand the main phenomena in the STSBO event through the results comparison obtained from simulations with these codes. Due that the simulation scope of SCDAP/RELAP5 is limited to failure of the vessel pressure boundary, the comparison was focused on in-vessel severe accident phenomena; with a special interest in the vessel pressure, boil of cooling, core temperature, and hydrogen generation. The results show that at the beginning of the scenario, both codes present similar thermal-hydraulic behavior for pressure and boil off of cooling, but during the relocation, the pressure and boil off, present differences in timing and order of magnitude. Both codes predict in similar time the beginning of melting material drop to the lower head. As far as the hydrogen production rate, SCDAP/RELAP5 predicts 15.8% lower production than MELCOR.

Development of Molten Core Relocation Analysis Module MCRA in the Severe Accident Analysis Code SAMPSON

IMPACT is a complex software system under development at the Nuclear Power Engineering Corporation, that includes the severe accident analysis code (SAMPSON). SAMPSON is an integration of twelve modules and will be capable of simulating hypothesized severe accidents in a nuclear power plant in the final phase of the IMPACT project. As one of these modules, the Molten Core Relocation Analysis (MCRA) module simulates the relocation behavior of a molten core during a severe accident. MCRA adopts a multi-phase, multi- component, multi-velocity field model to simulate severe accident phenomena mechanistically. Herein, we describe flow regimes, interfacial area modeling and physics models such as the momentum exchange and phase change models. Two separate effect analyses are presented as verification of MCRA's models following the model descriptions. First, the fluid dynamics of the multi-velocity field model was verified in the calculation of nitrogen gas bubbling through water in Leung's experiment. Second, data of the JRC-Ispra KROTOS-37 experiment were adopted for verification of the multi-velocity field model and vaporization/condensation model. Calculation results of both analyses compared well with the experimental data. Finally, MCRA's integrated function involving the melting/freezing model was demonstrated by calculating two-phase flow with a fuel rod and molten fuel.

Development of Molten Core Relocation Analysis Module MCRA in the Severe Accident Analysis Code SAMPSON.

IMPACT is a complex software system under development at the Nuclear Power Engineering Corporation, that includes the severe accident analysis code (SAMPSON). SAMPSON is an integration of twelve modules and will be capable of simulating hypothesized severe accidents in a nuclear power plant in the final phase of the IMPACT project. As one of these modules, the Molten Core Relocation Analysis (MCRA) module simulates the relocation behavior of a molten core during a severe accident. MCRA adopts a multi-phase, multi- component, multi-velocity field model to simulate severe accident phenomena mechanistically. Herein, we describe flow regimes, interfacial area modeling and physics models such as the momentum exchange and phase change models. Two separate effect analyses are presented as verification of MCRA's models following the model descriptions. First, the fluid dynamics of the multi-velocity field model was verified in the calculation of nitrogen gas bubbling through water in Leung's experiment. Second, data of the JRC-Ispra KROTOS-37 experiment were adopted for verification of the multi-velocity field model and vaporization/condensation model. Calculation results of both analyses compared well with the experimental data. Finally, MCRA's integrated function involving the melting/freezing model was demonstrated by calculating two-phase flow with a fuel rod and molten fuel.

Accident sequence precursor analysis: level 2/3 model development

The US Nuclear Regulatory Commission’s Accident Sequence Precursor (ASP) program currently uses simple level 1 models to assess the conditional core damage probability for operational events occurring in commercial nuclear power plants (NPP). Not all accident sequences leading to core damage will result in the same radiological consequences. Therefore, it is necessary to develop simple level 2/3 models that can be used to analyze the response of the NPP containment structure in the context of a core damage accident, estimate the magnitude of the resulting radioactive releases to the environment, and calculate the consequences associated with these releases. This level 2/3 model development work was initiated in 1995, and several prototype models have been completed. Two versions of these prototype level 2/3 models have been developed. Simple level 2/3 models have been linked to the simple level 1 models to provide risk perspectives for operational events. Additionally, very complex and detailed models have been developed that take provide much greater flexibility in accommodating a much wider range of level 1 core damage accident sequences. These detailed models also make possible assessments that are integrated with the level 1 model, on the importance of different severe accident phenomena and containment performance characteristics. This paper describes the development and capabilities of these level 2/3 ASP models, and the linkage to the level 1 models.

Assessment of the Severe Accident Mitigation Features of the European Pressurized Water Reactor by Cooperative Research and Development

In the development of the European pressurized water reactor (EPR), a new, challenging safety goal is the requirement to restrict the consequences of even severe accidents to the immediate vicinity of the plant.To deal with the consequences of a severe accident means to deal with different phenomena of an assumed core meltdown accident. Specific controlling and mitigating measures and dedicated design features that will cope with these phenomena are intended to be incorporated into the safety design of the EPR.To prove that the assumptions made by the EPR project are in line with the results of research and development, a first cooperation between Forschungszentrum Karlsruhe, the vendor Siemens Company, and a consortium of utilities was started in 1993 and produced provisional results in 1995. Further investigations of severe accident phenomena are necessary to verify the controlling and mitigating design features of the EPR.

Simulation of Maanshan TMLB’ Sequence with MELCOR 1.8.3

The MELCOR code, developed by Sandia National Laboratories (SNL), is capable of simulating the severe accident phenomena of light water reactor nuclear power plants (NPPs).A specific station blackout accident (TMLB’ sequence) for Maanshan NPP is simulated using the MELCOR 1.8.3 code. The MELCOR input deck for Maanshan NPP is established based on Maanshan NPP design data and the MELCOR users’ guides. The initial steady-state conditions are generated with a developed self-initialization algorithm. The main severe accident phenomena and the corresponding fission product release fractions associated with the TMLB’ sequence were simulated. The sequence of events up to the vessel breach is similar to that in the SNL report simulated with MELCOR 1.8.2. The predicted results provide useful information for the probabilistic risk assessment (PRA) of Maanshan NPP. This tool will be applied to the PRA, the severe accident analysis, and the severe accident management study of Maanshan NPP in the near future.

The modeling of core melting and in-vessel corium relocation in the APRIL code

This paper is concerned with the modeling of severe accident phenomena in boiling water reactors (BWR). New models of core melting and in-vessel corium debris relocation are presented, developed for implementation in the APRIL computer code. The results of model testing and validation are given, including comparisons against available experimental data and parametric/sensitivity studies. Also, the application of these models, as parts of the APRIL code, is presented to simulate accident progression in a typical BWR reactor.

シビアアクシデントに関する熱流動研究の最近の動向

Research on severe accident for light water reactors has been initiated after the accident of Three Mile Island in 1979 and accelerated all over the world after the Chernobyl accident in 1986. For the severe accident management Nuclear Safety Commission of Japan recommended electric utilities to investigate the possible measures in 1992. In response to this they proposed accident management strategies based on their investigations in 1994 to be voluntarily implemented in several years by reflecting findings from severe accident research. Since the severe accident phenomena largely involve the thermal-hydraulic aspects, it is of great importance to evaluate the thermal-hydraulic behaviors of severe accident in order to reduce the associated uncertainties. In the present article thermal-hydraulic behaviors during severe accident are described in terms of phenomenology, severe accident management, key issues, research methodology and remaining research items. The main phenomena include core melt progression, in-vessel core melt behaviors, molten core coolant interaction, molten core concrete interaction, direct containment heating, hydrogen behaviors, and fission product behaviors.

Analysis of Core Uncovery Time in Kuosheng Station Blackout Transient with MELCOR

The MELCOR code, developed by the Sandia National Laboratories, is capable of simulating severe accident phenomena of nuclear power plants. Core uncovery time is an important parameter in the probabilistic risk assessment. However, many MELCOR users do not generate the initial conditions in a station blackout (SBO) transient analysis. Thus, achieving reliable core uncovery time is difficult. The core uncovery time for the Kuosheng nuclear power plant during an SBO transient is analyzed. First, full-power steady-state conditions are generated with the application of a developed self-initialization algorithm. Then the response of the SBO transient up to core uncovery is simulated. The effects of key parameters including the initialization process and the reactor feed pump (RFP) coastdown time on the core uncovery time are analyzed. The initialization process is the most important parameter that affects the core uncovery time. Because SBO transient analysis, the correct initial conditions must be generated to achieve a reliable core uncovery time. The core uncovery time is also sensitive to the RFP coastdown time. A correct time constant is required.

Development of a MELCOR Self-Initialization Algorithm for Boiling Water Reactors

The MELCOR code, developed by Sandia National Laboratories, is suitable for calculating source terms and simulating severe accident phenomena of nuclear power plants. Prior to simulating a severe accident transient with MELCOR, the initial steady-state conditions must be generated in advance. The current MELCOR users` manuals do not provide a self-initialization procedure; this is the reason users have to adjust the initial conditions by themselves through a trial-and-error approach. A MELCOR self-initialization algorithm for boiling water reactor plants has been developed, which eliminates the tedious trial-and-error procedures and improves the simulation accuracy. This algorithm adjusts the important plant variable such as the dome pressure, downcomer level, and core flow rate to the desired conditions automatically. It is implemented through input with control functions provided in MELCOR. The reactor power and feedwater temperature are fed as input data. The initialization work of full-power conditions of the Kuosheng nuclear power station is cited as an example. These initial conditions are generated successfully with the developed algorithm. The generated initial conditions can be stored in a restart file and used for transient analysis. The methodology in this study improves the accuracy and consistency of transient calculations. Meanwhile, the algorithm provides all MELCOR usersmore » an easy and correct method for establishing the initial conditions.« less

Aspects of the application of Level 2 probabilistic safety assessment / Aspekte der Anwendung probabilistischer Sicherheitsanalysen der Stufe 2

Level 2 probabilistic safety assessments start from those plant states in which severe core damage is inevitably imminent. The objective is to assess severe accident sequences in order to determine radiological source terms with regard to the environment and their associated frequencies. In most countries using nuclear power, Level 2 PSAs are being considered as an important investigative tool to apply state-of-the-art knowledge on severe accident phenomena in a systematic way. Based on the results of Level 2 PSA, plant vulnerabilities have been identified and plant modifications as well as accident management measures have been proposed and implemented for further reduction and control of the consequences of severe accidents. A research project concerning a generic safety assessment of severe accidents in boiling water reactors was started in Germany in 1993. The study covers full-power states as well as shut-down states. The focus is on the development of a method for the analysis of severe accidents with an integrated probabilistic assessment. It also includes the assessment of accident management measures as well as uncertainty analyses

Towards an enhanced quality in pressurized water reactor safety / Zu einer höheren Qualität der Sicherheit von Druckwasserreaktoren

An improved safety approach for future light water reactors has been derived from the reinforced defense-in-depth concept. It comprises measures of severe accident prevention and measures of mitigation of accident consequences. Severe accident phenomena will be considered already in the stage of the plant's design. The overall objective is that even in the case of a core melt accident there will be no necessity for off-site emergency measures, such as evacuation or relocation of the population from the vicinity of the damaged plant, and that large-scale land contamination can be excluded. This new approach is being followed in the development of the European Pressurized Water Reactor (APR)

Uncertainties remaining in severe accident phenomena and related research activities at JAERI

Abstract In utilizing PSAs to safety-related issues such as provision of accident management, it is essentially important to understand the uncertainties in the results. However, large uncertainties still exist in level 2 PSA results even though worldwide research programs have been carried out to understand and analyze severe accidents. This paper summarizes the uncertainties due to lack of knowledge on severe accident phenomena and discusses their effects on the estimated severe accident progression and fission product source terms, mostly based on the researches conducted at the Japan Atomic Energy Research Institute (JAERI). Furthermore, the paper also discusses the uncertainties caused by incompleteness and simplification in modeling. Since most of the uncertainties in physical models will not be significantly reduced in the near future, the methodology is expected to effectively utilize the PSA results in decision making even with such large uncertainties, while continuous research efforts to reduce the uncertainties are required.

Review of Experimental Results on Light Water Reactor Core Melt Progression

This paper reports on results from integral-effects core melt progression experiments and from the examination of the damaged core of the Three Mile Island Unit 2 (TMI-2) reactor which are reviewed to gain insight on key severe accident phenomena. The experiments and the TMI-2 accident represent a wide variety of conditions and physical scales, yet several important phenomena appear to be common to core melt progression. Eutectic interactions between core materials cause the formation of liquids and loss of original core geometry at low temperatures ({approximately}1500 K) in a severe accident. The first liquids to form are metallic in nature, and they relocate to lower elevations in the core, where they may freeze into a crust that forms a partial flow blockage. At temperatures above {approximately}2200 K, fuel liquefaction causes fuelbearing debris to accumulate in the core above the metallic lower crust. The liquefied material oxidizes in steam as it relocates, and the accumulated melt can incorporate unmelted fuel rod debris.

What we now know (and don't know) about reactor safety

This paper deals with the likelihood and consequences of major reactor accidents. Our understanding has advanced markedly in the past decade, largely because of the use of probabilistic risk assessment. There is now a good understanding of accident sequences, despite some important gaps. The insights achieved have in many ways revolutionized the reactor safety community's view of the relative importance of various safety issues. Understanding of severe accident phenomena is still inadequate, but experiments and development of analytical methods will soon enable us to calculate better the phenomena involved in core degradation, core melting, containment behavior, and the ultimate fate of fission products released from a melted core. Also, we are now gaining important insights into analysis of human errors and external initiating events. Thus the next decade will see another quantum jump in our understanding, rivaling the progress of the last decade.