Automated modelling approach for neutronic analysis of high temperature gas-cooled pebble bed reactors

Pebble bed reactors are particularly characterized by stochastic distribution of TRISO (Tri-structural Isotropic) particles and pebbles in the core. These random distributions pose a unique challenge for neutronic analysis. Generation of randomized locations for TRISO particles in fuel sphere along with pebbles in the core manually is a laborious and protracted task. This paper introduces an innovative and simple approach to model randomly distributed TRISO particles within fuel pebbles i.e., Random Vector Method. A three-dimensional grid is superimposed on fuel region of the pebble. A vector with random magnitude and direction starts from the center of grid. A TRISO particle is added if the mesh at the current position is empty. The vector is resampled if mesh is already filled or goes beyond the fuel region boundary. This repeats until required fuel loading has been achieved. In effect some of the mesh points remain empty, thereby imparting randomness. Secondly, each TRISO particle is arbitrarily displaced from the center within the margins of its cube. This acts as a twofold process resulting in random dispersion of TRISO particles in the fuel sphere. As for the pebbles, technique described in IAEA benchmark document has been applied. Fuel and moderator pebbles are stochastically distributed on a layer of hexagonal lattice while maintaining a certain fuel-to-moderator ratio. These layers are stacked vertically to achieve the desired core height. Volume map is introduced and applied to correctly account for cut spheres in the total tally of fuel/moderator balls for a given core configuration. Overall automation of the methodology proves to be highly useful in modelling pebble bed reactors with varying parameters e.g., core height, fuel to moderator (F:M) ratio, and number of TRISO particles. It also significantly reduces the time and effort required to generate these complex random distributions, making it a useful tool for various pebble bed reactor calculations. To validate the methodology, the HTR-10 reactor has been modelled using this technique, and the results are compared with various benchmarks from the IAEA TECDOC 1382 and 1694. The findings demonstrate a good concordance with the benchmark data, highlighting the suitability and efficiency of this modelling technique for neutronic analysis in pebble bed reactors.