THE DEVELOPMENT OF A PREPROCESSOR TO GENERATE INPUT FOR MONTE-CARLO CODES AND ITS APPLICATION TO FAST-REACTOR ANALYSIS

For many calculations in Reactor Physics. Monte Carlo methods have sig nificant advantages over standard deterministic techniques. However, b ecause very accurate geometric models can be generated, input data to Monte Carlo codes is difficult, tedious and extremely time consuming t o prepare. This paper describes a preprocessor which reduces considera bly the size and complexity of user generated data. In fact only dimen sional and composition information is required by the preprocessor for it to produce the code input file and it is believed that this is a s ignificant development in facilitating Monte Carlo data preparation. A fter a description of the preprocessor an indication of the reduction in effort required to generate the Monte Carlo input for the heterogen eous pin structure of MASURCA by using the preprocessor is given follo wed by comparisons of calculated and experimental k(eff) values. Final ly future developments and applications of the preprocessor are descri bed. In this paper only applications to Fast Reactor Physics will be c onsidered. However the approach is quite general and could be extended to Thermal studies.