INTELIGENT FUEL CORE MANAGMENT FOR PWR REACTORS
This work presents a methodology applied to perform an in-core fuel management (ICFM) in PWR-type reactors using genetic algorithms (AG's). ICFM consists of defining a recharge pattern during the operational cycles of the reactor, that is, finding the best arrangement of new and partially burned fuel elements (EC's), which optimizes the performance of the reactor according to its safety criteria. For this, the proposed methodology is based on developing an interface in which the GA can interact with the simulation code of reactor physics, which holds the neutron characteristics of each EC and which must be reliable and fast. Therefore, it was necessary to develop, through a technique already consolidated in the Literature, a thick mesh node code that numerically solves the multigroup diffusion equation for 2 groups of energy, fast and thermal neutrons, in two dimensions. In this type of code, it is necessary for each EC, represented by a node, to be homogenized and represented by its multigroup constants, for each burnup of the reactor. For this, the SCALE system, developed by the Reactor and Nuclear Systems Division (RNSD) of the Oak Ridge National Laboratory (ORNL) was used. All the qualification and validation of the results obtained from the homogenization of the EC’s by the SCALE together with the nodal code were performed comparing them with the Benmarmark of the Almaraz Nuclear Power Plant provided by the IAEA and other benchmarks found in the literature.