MHTGR-350 원지로의 제작오차에 의한 임계도 불확실도 연구

Abstract

Recently, there has been an increasing demand for the verifications of the High-Temperature Gas-cooled Reactor (HTGR) design using the reliable code and analysis system. Uncertainty analysis should be properly evaluated for the design optimization and safety features assessment. The criticality is one of the main reactor characteristics affected by the uncertainties in systematic parameters in addition to mathematical approximations and numerical errors in a model. The best estimated keff uncertainty can provide opportunities to further optimize designs and operations while simultaneously maintaining safety margins. In this study, a Monte Carlo simulation method of the systematic parameters is proposed for the criticality bias calculation. For the criticality estimation, MCNP modeling method was proposed and optimized by employing cell dividing approach method. For the application of the mechanical difference into the MCNP simulation, the sampling of systematic parameters was obtained from probability distribution functions. The analysis program of criticality was developed with C++ language with MCNP5 code. The analysis method is based on the following procedure: (a) sampling the positions of the fuel particles randomly distributed in the cylindrical compact; (b) sampling the data set of systematic parameters from the normal or right triangular distribution; (c) creating the series of MCNP input files with applying the distributions of systematic parameters and performing the criticality calculations using the MCNP5; (d) collecting the series of results with statistical sampling parameter and evaluating criticality biases. As the result of this study, it is analyzed that criticality bias in the geometrical uncertainty is most caused by fuel kernel radius. It is also found that impurity variation in graphite significantly affects the criticality analysis. The method proposed in this study can identify the key systematic parameters whose uncertainties contribute most to keff and assist the design optimization to make the system less sensitive to identified key sources of uncertainties. It is expected that the uncertainty data can be provided to nuclear fuel cycle industries, regulatory agencies, and research laboratories for the decision-making support for criticality safety assessments.