Matthew W Francis

Fukushima Daiichi accident study : status as of April 2012.

In response to the accident at the Fukushima Daiichi nuclear power station in Japan, the U.S. Nuclear Regulatory Commission (NRC) and Department of Energy agreed to jointly sponsor an accident reconstruction study as a means of assessing severe accident modeling capability of the MELCOR code. MELCOR is the state-of-the-art system-level severe accident analysis code used by the NRC to provide information for its decision-making process in this area. The objectives of the project were: (1) collect, verify, and document data on the accidents by developing an information portal system; (2) reconstruct the accident progressions using computer models and accident data; and (3) validate the MELCOR code and the Fukushima models, and suggest potential future data needs. Idaho National Laboratory (INL) developed an information portal for the Fukushima Daiichi accident information. Sandia National Laboratories (SNL) developed MELCOR 2.1 models of the Fukushima Daiichi Units 1, 2, and 3 reactors and the Unit 4 spent fuel pool. Oak Ridge National Laboratory (ORNL) developed a MELCOR 1.8.5 model of the Unit 3 reactor and a TRACE model of the Unit 4 spent fuel pool. The good correlation of the results from the SNL models with the data from the plants and with the ORNL model results provides additional confidence in the MELCOR code. The modeling effort has also provided insights into future data needs for both model development and validation.

STUDY OF FUKUSHIMA DAIICHI NUCLEAR POWER STATION UNIT 4 SPENT-FUEL POOL

A study on the Fukushima Daiichi nuclear power station spent-fuel pool (SFP) at Unit 4 (SFP4) is presented in this paper. We discuss the design characteristics of SFP4 and its decay heat load in detail and provide a model that we developed to estimate the SFP evaporation rate based on the SFP temperature. The SFP level of SFP4 following the March 11, 2011, accident is predicted based on the fundamental conservation laws of mass and energy. Our predicted SFP level and temperatures are in good agreement with measured data and are consistent with Tokyo Electric Power Company evaluation results.

Insight from Fukushima Daiichi Unit 3 Investigations using MELCOR

Abstract During the emergency response period of the accidents that took place at the Fukushima Daiichi nuclear power plant (NPP) in March of 2011, researchers at Oak Ridge National Laboratory (ORNL) conducted a number of studies using the MELCOR code to help understand what was occurring and what had occurred. During the postaccident period, the U.S. Department of Energy (DOE) and the U.S. Nuclear Regulatory Commission (NRC) jointly sponsored a study of the Fukushima Daiichi NPP accident with collaboration among ORNL, Sandia National Laboratories, and Idaho National Laboratory. The purpose of the study was to compile relevant data, reconstruct the accident progression using computer codes, assess the codes’ predictive capabilities, and identify future data needs. The current paper summarizes some of the early MELCOR simulations and analyses conducted at ORNL of the Fukushima Daiichi NPP Unit 3 (1F3) accident. Extended analysis and discussion of the 1F3 accident are also presented taking into account new knowledge and modeling refinements made since the joint DOE-NRC study.

Fukushima Daiichi Unit 1 Ex-Vessel Prediction: Core Melt Spreading

Abstract Lower head failure and corium-concrete interaction were predicted to occur at Fukushima Daiichi Unit 1 (1F1) by several different system-level code analyses, including MELCOR v2.1 and MAAP5. Although these codes capture a wide range of accident phenomena, they do not contain detailed models for ex-vessel core melt behavior. However, specialized codes exist for the analysis of ex-vessel melt spreading (e.g., MELTSPREAD) and long-term debris coolability (e.g., CORQUENCH). On this basis, an analysis has been carried out to further evaluate ex-vessel behavior for 1F1 using MELTSPREAD and CORQUENCH. The best-estimate melt pour conditions predicted by MELCOR v2.1 and MAAP5 were used as input. MELTSPREAD was then used to predict the spatially dependent melt conditions and the extent of spreading during relocation from the vessel. This information was then used as input for the long-term debris coolability analysis with CORQUENCH, which is reported in a companion paper.