Time-dependent nuclear reactor analysis with dynamic Monte Carlo techniques

In this study, a detailed non-analog Dynamic Monte Carlo (DMC) methodology is provided and used to investigate the transient analysis of nuclear systems. The validity of the given methodology is demonstrated by solving a set of time-dependent benchmark problems. We also developed a new technique to generate time dependent Green's functions using DMC to perform transient analysis of source-driven systems (SDSs). The equilibrium and transient responses of SDSs are determined by using these generated Green's functions. This novel method is validated with comparison to different transient benchmark problem of SDSs. In the conventional Point Reactor Kinetics Models (PRKMs), the time evolution of both neutron population and power are taken proportional to the weighted neutron density (amplitude function). The weight function must be chosen in such a way that the calculated kinetics parameters and amplitude function are as accurate as possible to characterize actual system. In this work, it is shown that, there are cases where the time dependency of power and neutron population differ. To overcome this problem, we derived the general forms of one- and two- point reactor kinetics models by using the actual neutron population and power, which are different from the conventional PRKMs. For different weight functions, the derived PRKMs are tested for transient analysis of one speed reflected slab reactors. Thus, obtained results are compared with the exact analytical solutions given by the Eigenfunction Expansion Method (EEM). Furthermore, for different amount of reactivity insertions, taking into account the reactivity feedback, both non-analog dynamic Monte Carlo method named Point Kinetics Monte Carlo (PKMC) and Stochastic Point Reactor Kinetics Model (SPRKM) are developed, to simulate one- and two-point reactor kinetic models of the reflected reactors. Finally, a non-analog stochastic kinetics model is developed to simulate the PRKMs and compared with analog stochastic point kinetics model.