ICONE19-43257 TUBE SUPPORT PLATE BLOCKAGE IN STEAM GENERATORS : INVERSE METHOD FOR THE ESTIMATION OF UNOBSERVABLE PARAMETERS IN A DEPOSIT MODEL

Abstract

An improper cleanliness in the secondary circuit of nuclear plant steam generators may result in tube support plate blockage by deposits, and tube fouling. In order to improve our understanding of those two phenomena, a model for the growth of solid deposits on the secondary side of steam generators has been developed by the R&D Division of EDF. This model has been implemented in the frame of THYC, which is the EDF reference code for the modelling of two-phase thermal-hydraulic phenomena at the steam generator scale. It aims to evaluate the localization and the growth rate of deposits, as well as the resulting additional pressure losses. The model considers that the deposits are corrosion products that are introduced in the steam generator either as suspended particles or as dissolved species. The transport of corrosion products in the secondary circuit is modelled, as well as the mass transfers between dissolved species, suspended particles and wall deposits. Several deposit mechanisms have been investigated, such as boiling, diffusion, turbulence, and deposit removal. Mass transfers being complex processes which depend on several physical and chemical local parameters, they cannot be represented by simple laws. It is therefore necessary to introduce some empirical correlations between thermalhydraulic parameters and deposit growth rates. Those correlations depend on parameters whose values can only be determined by specific experiments performed on devoted test-facilities. Authors of the domain provide different values for those parameters, depending on experimental conditions. At the steam generator scale and in Pressurized Water Reactors (PWR) conditions, their values are not determined. We hereafter present an inverse method, that could enable one to evaluate those parameters in PWR conditions, by fitting the results of simulations to the actual levels of deposits observed in some French nuclear plants. In this paper, we only discuss the case of tube support plate blockage but the method is in principle also applicable to deal with tube fouling. The limits of the model and the strategy to improve it are discussed.