A reduced order model for investigating the dynamics of the Gen-IV LFR coolant pool

In the control field, the study of the system dynamics is usually carried out relying on lumped-parameter or one-dimensional modelling. Even if these approaches are well suited for control purposes since they provide fast-running simulations and are easy to linearize, they may not be sufficient to deeply assess the complexity of the systems, in particular where spatial phenomena have a significant impact on dynamics. Reduced Order Methods (ROM) can offer the proper trade-offbetween computational cost and solution accuracy. In this work, a reduced order model for the spatial description of the Gen-IV LFR coolant pool is developed for the purpose of being employed in a control-oriented plant simulator of the ALFRED reactor. The spatial modelling of the reactor pool is based on the POD-FV- ROM procedure, previously developed with the aim of extending the literature approach based on Finite Element to the Finite Volume approximation of the Navier–Stokes equa- tions, and building a reduced order model capable of handling turbulent flows modelled through the RANS equations. The mentioned approach is employed to build a ROM-based component of the ALFRED simulator for the coolant pool. The possibility of varying the in- put variables of the model has been also undertaken. In particular, the lead velocity at the Steam Generator outlet has been considered as a parametrized boundary condition since it can be a possible control variable. The results have turned out to be very satisfactory in terms of both accuracy and computational time. As a major outcome of the ROM model, it has been proved that its behaviour is more accurate than a 0D-based model without requiring an excessive computational cost.