The severe accident scenario propagation studies of nuclear power plants (NPPs) have been one of the most critical factors in deploying nuclear power for decades. During an NPP accident, the accident scenario can change during its propagation from the initiating event to a series of accident sub-scenarios. Hence, having time-wise updated information about the current type of accident sub-scenario can help plant operators mitigate the accident propagation and underlying consequences. In this work, we demonstrate the capability of machine learning (Decision Tree) to help researchers and design engineers in finding distinctive physical insights between four different types of accident scenarios based on the pressure vessel's maximum external surface temperature at a particular time. Although the four accidents we included in this study are considered some of the most extensively studied NPPs accident scenarios for decades, our findings shows that decision tree classification could define remarkable distinct differences between them with reliable statistical confidence.

In a severe accident scenario of a nuclear power plant involving core meltdown and relocation to the lower head of the reactor pressure vessel (RPV), the vessel may undergo serious deformation and even failure due to extreme thermo-mechanical loads from the relocated core melt. Proper material models and detailed structural analysis are paramount in predicting the timing and mode of possible vessel failure.This paper presents a strain hardening creep model with optimal parameters to simulate the material behavior of the reactor steel 16MND5 under extreme thermo-mechanical loads. First, validations against two experiments, a tensile-creep test and the EU-REVISA RUPTHER #14 test, show that the proposed model is best overall compared to three previous models. Next, the creep model is implemented for the thermo-mechanical analysis of an ablated RPV under a severe accident scenario with external vessel cooling as a mitigation strategy. The effect of internal pressures from 3 to 50 bars is investigated with the assumption that the corners of the ablated part of the vessel have sharp corners. In this case, we found that the vessel fails above 40 bars. However, if we model the corners with varying smoothness or fillet sizes, we found significant delay in failure time and an increase in failure internal pressure.

In a postulated severe accident, the thermo-mechanical loads from the corium debris that has relocated to the lower head of the reactor pressure vessel (RPV) can pose a credible threat to the RPV's structural integrity. In case of a vessel breach, it is vital to predict the mode and timing of the vessel failure. This affects the ex-vessel accident progression and plays a critical role in the development of mitigation strategies. We propose a methodology to assess RPV failure based on MELCOR and ANSYS Mechanical APDL simulations. A Nordic-type boiling water reactor (BWR) is considered with two severe accident scenarios: i) SBO (Station Blackout) and ii) SBO + LOCA (Loss of Coolant Accident). In addition, the approach considers the dynamic ablation of the vessel wall due to a high-temperature debris bed with the use of the element kill function in ANSYS. The results indicate that the stress failure mechanism is the major cause of the RPV failure, compared to the strain failure mechanism. Moreover, the axial normal stress and circumferential normal stress make the dominant contributions to the equivalent stress sigma at the lower head of RPVs. As expected, the region with high ablation is most likely the failure location in both SBO and SBO + LOCA. In addition, comparisons of the failure mode and timing between SBO and SBO + LOCA are described in detail. A short discussion on RPV failure between ANSYS and MELCOR is also presented.

The reactor pressure vessel (RPV) of a nuclear reactor is one of the key safety barriers preventing radioactive environmental releases during a severe accident. One of the promising strategies of severe accident management (SAM) is to retain the molten core having continuous decay heat inside the RPV by natural water cooling of the external vessel surface. The feasibility of such a strategy relies on complex safety analyses including accurateprediction of vessel thermo-mechanical behavior which can be assessed by mechanical stresses and strains. In this paper, we present the stress–strain response of an ablated RPV of a Nordic boiling water reactor (BWR) to dynamic thermomechanical loads set by expanding volumetrically heated molten pool inside the RPV cooled by water at the external surface. MELCOR 2.2.9541 severe accident code is used to simulate the in-vessel behavior and provides the input conditions for dedicated structural analysis of the RPV using ANSYS® Mechanical APDL 19.2. A creep model of the SA533B1 vessel steel is validated against uniaxial creep tests carried out by INEL (Idaho National Engineering Laboratory) and creep tests performed at CEA (French AlternativeEnergies and Atomic Energy Commission) as part of the OLHF (OECD Lower Head Failure) Project. Two generic severe accident scenarios are considered: (i) Station Blackout (SBO) and (ii) Station Black-out and Loss-of-coolant Accident (SBO + LOCA). In both scenarios, we found that the RPV has maintained structural integrity considering two failure criteria: stress-based and strain-based.

Vessel penetrations are important features of both pressurized water reactors and boiling water reactors. The thermal and structural behaviour of instrumentation guide tubes (IGTs) and control rod guide tubes (CRGTs) during a severe accident is vital in the assessment of the structure integrity of the reactor pressure vessel. Penetrations may fail due to welding failure, nozzle rupture, melt-through, etc. It is thus important to assess the failure mechanisms of penetrations with sufficient details. The objective of this paper is to assess the timing and failure modes of IGTs at the lower head during a severe accident in a Nordic boiling water reactor. In this study, a three-dimensional local finite element model was established using Ansys Mechanical that includes the vessel wall, the nozzle, and the weld joint. The thermo-mechanical loads of the finite element model were based on MELCOR results of a station blackout accident (SBO) combined with a large-break loss-of-coolant accident (LBLOCA) including an external vessel cooling by water as a severe accident management strategy. Given the temperature, creep strain, elastic strain, plastic strain, stress and displacement from the ANSYS simulations, the results showed the timing and failure modes of IGTs. Failure of the IGT penetration by nozzle creep is found to be the dominant failure mode of the vessel. However, it was also found that the IGT is clamped by the flow limiter before the nozzle creep, which means that IGT ejection is unlikely.

In a severe accident scenario of a nuclear power plant involving core meltdown, the instrumentation guide tubes (IGTs) that exist in some reactor designs may fail and lead to melt leakage through the RPV lower head. Therefore, failure analysis on IGTs plays a significant role in RPV failure analysis. In this study, a 3D global Finite Element (FE) model with three IGT structures is considered. A detailed analysis of three IGT structures illustrates the effect of location on the behaviour of weld joints. In addition, the failure of the welds is predicted with three failure criteria: melt-through failure, stress-based failure, and strain-based failure. Finally, a possible ejection failure for three IGTs at different locations on the RPVs is also discussed.