The goal of this thesis is to implement another method than the widely used Monte Carlo method to do sensitivity analysis studies. For nominal calculations the computational cost is often not prohibitive, but when sensitivity analyses are done, many samples are needed, often of the order of 102 − 104. In this case Monte Carlo methods will require hundreds of days, even with state-of-the-art computers. This is why for this thesis the method of characteristics is used. A deterministic method (as opposed to a stochastic method) is expected to provide faster simulations; nevertheless, all this with a lower accuracy. The intention is to use these kinds of simulations as first indications. For definitive, more accurate simulations based on Monte Carlo techniques as the neutron transport driving tool, could then be used afterwards. A major challenge of the MYRRHA reactor is the production of the radiotoxic polonium. Especially at times when the reactor would be opened like for maintenance, it is important to know the concentration of polonium. Besides having an estimation of the concentration of polonium, it is also important to have an indication on the possible error on this estimation, which can be done through sensitivity analyses. The polonium is produced by neutronic capture in bismuth 209. MYRRHA uses a lead-bismuth eutectic as coolant and therefore, bismuth 209 is present in large quantities providing the problem of polonium production. In this thesis a literature review is provided about the method of characteristics, multiple energy self-shielding methods and on SANDY, a code developed at SCK-CEN to sample nuclear data libraries. Next to this literature review, a sensitivity analysis is done on the effect of a varying bismuth capture cross section on the polonium concentration after a period of 90 days, after these 90 days the reactor is scheduled to open for maintenance.
|State||Published - 25 Jun 2018|