The design of the TRIGA Mark II reactor at the “Jožef Stefan” Institute allows pulse mode operation. In a pulse experiment one control rod is quickly removed out of the reactor core and therefore the power begins to increase. As fuel heating increases the absorption of neutrons in uranium 238U and the thermal neutron spectrum shifts to higher energies. Both phenomena lead to the instant decrease in reactivity and consequently decreases power. In pulsed operation mode high and short pulses of power are obtained.
All the pulse experiments performed at the TRIGA reactor at “Jožef Stefan” Institute were collected and analyzed. The theoretical predictions (Fuchs-Hansen and Nordheim-Fuchs model) and an improved computational model of pulse experiments were compared. At an improved computational model of pulse experiments the six group point kinetics equations were solved taking into account the temperature dependence of the prompt negative temperature reactivity coefficient of the fuel and specific heat, the final time of the ejection of the transient control rod, whose value of the reactivity changes in height and the heat exchange from the fuel. With the PULSTRI-1 program was analyzed the maximal fuel temperature in the reactor core during a pulse experiment, depending on the power peaking factor and the inserted reactivity. The major part of the master's thesis is intended to evaluate the experimental and computational uncertainties of the physical parameters of the pulse, where the uncertainty of the limit expressions (maximum power, total released energy, width at half maximum) is estimated according to the theoretical model for correlated and uncorrelated parameters. Since it was found that the correlation does not significantly contribute to the estimated uncertainties, a sensitivity analysis of the theoretical model is carried out in the case of uncorrelated parameters. In the last part of the master thesis the theoretical model and the improved computational model of pulse experiments were compared with experimental data. Theoretical prediction in most cases matches experimental data within the estimated uncertainty, as well is good matching between experimental data and an improved computational model of pulse experiments. The deviation occurs in the analysis where is taken into account the number of fuel elements in the reactor core, where the theoretical predictions and the improved computational model of pulse experiments predict trend in the opposite direction as experimental data.
|
