Flownex Validation for Steady-State Temperature Calculation of the HTR-10 Benchmark Problem

Abstract

The International Atomic Energy Agency (IAEA) initiated the framework of a Coordinated Research Project (CRP) on the evaluation of high temperature gas-cooled reactor performance. This project defined the calculation of the steady-state temperature distribution of the 10MW High Temperature Gas- Cooled Reactor Test Module (HTR-10) under its initial full power experimental operation as one of the benchmark problems. A Flownex validation exercise was performed using this HTR-10 benchmark problem, which demonstrated that Flownex can predict the temperature distribution in the HTR-10 reactor core within ±15% prediction error for steady-state operation at full power. The discretization error for a coarse node model of the HTR-10 reactor core model was estimated to be approximately 9°C (1%) for predicting fuel pebble centerline temperature (coarse grid underprediction of temperature). The code bias for the side reflector temperatures was estimated to be 29.8°C (overprediction), and -37.6°C (underprediction) for the top and bottom regions of the core. The Euclidian difference between the experiment and simulation results is calculated to be 12.8%.

The Flownex HTR-10 benchmark study was compared with the previous two benchmark study results (performed by NET and CEA): the Chinese Institute of Nuclear and New Technology (INET) employed the modular code THERMIX, while the French Atomic Energy Commission (CEA) completed the benchmark problem using the CFD code ARCTURUS. Overall, the predicted temperatures by all three codes agreed well with the measured temperatures within ±15% prediction errors. The code bias and its variation in all three benchmark studies were similar.