Nuclear Science and Engineering / Volume 177 / Number 2 / June 2014 / Pages 126-140
Technical Paper / dx.doi.org/10.13182/NSE13-39
Measurement and monitoring of the degree of subcriticality of accelerator-driven systems (ADSs) are essential safety requirements to ensure that such systems remain subcritical during operation and shutdown. In recent years, a number of methods for measuring and monitoring subcriticality in ADSs have been studied around the world. Many low-power experiments have been performed, and still others are planned. Similar experiments are being planned at the Bhabha Atomic Research Centre. One general class of these techniques is based on neutron noise theory. As a part of the experimental planning, we have carried out simulations of the proposed noise experiments using a Monte Carlo–based neutron diffusion code developed for this purpose. These simulations have provided us with valuable information about the feasibility of the proposed experiments and the kind of accuracy that can be expected from such measurements. Since a diffusion theory–based Monte Carlo code has its own limitations, a more accurate description will be provided by transport theory–based analog Monte Carlo. The present paper discusses the development of such a code specifically intended for simulating the noise-based experiments, such as Rossi-alpha and Feynman-alpha. The code is based on the delta neutron tracking method (also called the Woodcock and Coleman method), which results in fast and relatively simple handling of complex geometries. The code has been validated with a few criticality and noise benchmark problems. The paper also presents results of simulations of the proposed ADS noise experiments at the Purnima facility obtained using the code.