A.W. Shiver

Robustness of an uncertainty and sensitivity analysis of early exposure results with the MACCS reactor accident consequence model

Abstract Uncertainty and sensitivity analysis techniques based on Latin hypercube sampling, partial correlation analysis and stepwise regression analysis were used in an investigation with the MACCS model of the early health effects associated with a severe accident at a nuclear power station. The following results were obtained in tests to check the robustness of the analysis techniques: two independent Latin hypercube samples produced similar uncertainty and sensitivity analysis results; setting important variables to best-estimate values produced substantial reductions in uncertainty, while setting the less important variables to best-estimate values had little effect on uncertainty; similar sensitivity analysis results were obtained when the original uniform and loguniform distributions assigned to the 34 imprecisely known input variables were changed to left-triangular distributions and then to right-triangular distributions; and analyses with rank-transformed and logarithmically-transformed data produced similar results and substantially outperformed analyses with raw (i.e., untransformed) data.

Uncertainty and sensitivity analysis of chronic exposure results with the MACCS reactor accident consequence model

Abstract Uncertainty and sensitivity analysis techniques based on Latin hypercube sampling, partial correlation analysis and stepwise regression analysis are used in an investigation with the MACCS model of the chronic exposure pathways associated with a severe accident at a nuclear power station. The primary purpose of this study is to provide guidance on the variables to be considered in future review work to reduce the uncertainty in the important variables used in the calculation of reactor accident consequences. The effects of 75 imprecisely known input variables on the following reactor accident consequences are studied: crop growing-season dose, crop long-term dose, water ingestion dose, milk growing-season dose, long-term groundshine dose, long-term inhalation dose, total food pathways dose, total ingestion pathways dose, total long-term pathways dose, total latent cancer fatalities, area-dependent cost, crop disposal cost, milk disposal cost, population-dependent cost, total economic cost, condemnation area, condemnation population, crop disposal area and milk disposal area. When the predicted variables are considered collectively, the following input variables were found to be the dominant contributors to uncertainty: dry deposition velocity, transfer of cesium from animal feed to milk, transfer of cesium from animal feed to meet, ground concentration of Cs-134 at which the disposal of milk products will be initiated, transfer of Sr-90 from soil to legumes, maximum allowable ground concentration of Sr-90 for production of crops, fraction of cesium entering surface water that is consumed in drinking water, groundshine shielding factor, scale factor defining resuspension, dose reduction associated with decontamination, and ground concentration of I-131 at which disposal of crops will be initiated due to accidents that occur during the growing season. Reducing the uncertainty in the preceding variables was found to substantially reduce the uncertainty in the predicted variables under consideration. For total number of latent cancer fatalities, the dominant variable was dry deposition velocity, with small effects indicated for a large number of additional variables.

The NUREG-1150 probabilistic risk assessment for the Surry nuclear Power station

Abstract This paper summarizes the findings of the probabilistic risk assessment for Unit 1 of the Surry Power Station performed in support of NUREG-1150. The emphasis is on the “back-end” analyses, that is, the accident progression, source term, and consequence analyses, and the risk results obtained when the results of these analyses are combined with the accident frequency analysis. The results show that the risk from internal initiators is well below the safety goals and is somewhat lower than estimated by Reactor Safety Study (RSS) over a decade ago. The risk from internal initiators is dominated by bypass accidents in the current analysis. The risk from seismic initiators is comparable to or greater than the risk from internal initiators, but still less than that estimated for internal initiators in the RSS. The uncertainty band for both types of initiators is considerably greater than that estimated in the RSS.

The NUREG-1150 probabilistic risk assessment for the Peach Bottom Atomic Power Station

Abstract This paper summarizes the findings of the probabilistic risk assessment (PRA) for Unit 2 of the Peach Bottom Power Station performed in support of NUREG-1150. The emphasis is on the “back-end” analyses, that is, the accident progression, source term, and consequence analyses, and the risk results obtained when the results of these analyses are combined with the accident frequency analysis. The results show that the annual risk from internal initiators is much lower than estimated by the Reactor Safety Study (RSS) over a decade ago. The risk from fire initiators is about an order of magnitude higher than the risk from internal initiators, but is still less than the risk from internal initiators estimated by the RSS. The risk from seismic initiators at Peach Bottom is much greater than the risk from internal initiators. The uncertainty band for all types of initiators is considerably greater than that estimated in the RSS.

The NUREG-1150 probabilistic risk assessment for the Grand Gulf Nuclear Station☆

Abstract This paper summarizes the findings of the probabilistic risk assessment (PRA) for Unit 1 of the Grand Gulf Nuclear Station performed in support of NUREG-1150. The emphasis is on the “back-end” analyses, that is, the accident progression, source term, consequence analyses, and risk results obtained when the results of these analyses are combined with the accident frequency analysis. The offsite risk from internal initiating events was found to be quite low, both with respect to the safety goals and to the other plants analyzed in NUREG-1150. The offsite risk is dominated by short-term station blackout plant damage states. The long-term station blackout group and the anticipated transients without scram (ATWS) group contribute considerably less to risk. Transients in which the power conversion system is unavailable are very minor contributors to risk. The low values for risk can be attributed to low core damage frequency, good emergency response, and plant features that reduce the potential source term.

The NUREG-1150 probabilistic risk assessment for the Sequoyah Nuclear Plant☆

Abstract This paper summarizes the findings of the probabilistic risk assessment (PRA) for Unit 1 of the Sequoyah Nuclear Plant performed in support of NUREG-1150. The emphasis is on the “back-end” analyses, that is, the accident progression, source term, and consequence analyses, and the risk results obtained when the results of these analyses are combined with the accident frequency analysis. The results of this PRA indicate that the offsite risk from internal initiating events at Sequoyah are quite low with respect to the safety goals. The containment appears likely to withstand the loads that might be placed upon it if the reactor vessel fails. A good portion of the risk, in this analysis, comes from initiating events which bypass the containment. These events are estimated to have a relatively low frequency of occurrence, but their consequences are quite large. Other events that contribute to offsite risk involve early containment failures that occur during degradation of the core or near the time of vessel breach. Considerable uncertainty is associated with the risk estimates produced in this analysis. Offsite risk from external initiating events was not included in this analysis.

Uncertainty and sensitivity analysis of food pathway results with the MACCS reactor accident consequence model

Uncertainty and sensitivity analysis techniques based on Latin hypercube sampling, partial correlation analysis and stepwise regression analysis are used in an investigation with the MACCS model of the food pathways associated with a severe accident at a nuclear power station. The primary purpose of this study is to provide guidance on the variables to be considered in future review work to reduce the uncertainty in the important variables used in the calculation of reactor accident consequences. The effects of 87 imprecisely-known input variables on the following reactor accident consequences are studied: crop growing-season dose, crop long-term dose, milk growing-season dose, total food pathways dose, total ingestion pathways dose, total long-term pathways dose, area dependent cost, crop disposal cost, milk disposal cost, condemnation area, crop disposal area and milk disposal area. When the predicted variables are considered collectively, the following input variables were found to be the dominant contributors to uncertainty: fraction of cesium deposition on grain fields that is retained on plant surfaces and transferred directly to grain, maximum allowable ground concentrations of Cs-137 and Sr-90 for production of crops, ground concentrations of Cs-134, Cs-137 and I-131 at which the disposal of milk will be initiated due to accidents that occur during the growing season, ground concentrations of Cs-134, I-131 and Sr-90 at which the disposal of crops will be initiated due to accidents that occur during the growing season, rate of depletion of Cs-137 and Sr-90 from the root zone, transfer of Sr-90 from soil to legumes, transfer of Cs-137 from soil to pasture, transfer of cesium from animal feed to meat, and the transfer of cesium, iodine and strontium from animal feed to milk.

Computational implementation of a systems prioritization methodology for the Waste Isolation Pilot Plant: a preliminary example

Abstract A systems prioritization methodology (SPM) is under development for the Waste Isolation Pilot Plant (WIPP). The SPM is based on a large numerical integration problem that must be repeatedly evaluated to determine compliance probabilities associated with different experimental programs and design modifications. Due to the complexity and computational cost of the underlying integration problem, the implementation of the SPM must be planned very carefully. This presentation describes a preliminary application of the SPM, designated SPM-l, performed to provide insights to facilitate the development and implementation of the methodology. Topics illustrated by SPM-1 include definition of probability spaces on which the SPM is based, use of Latin hypercube sampling and simple random sampling to integrate over different probability spaces, selection of mechanistic calculations to be performed, efficient use of the limited number of mechanistic calculations that can be performed, and assembly of many individual calculations into a complete analysis.