Computer codes thermal effect models

Although much as been done, much work remains. Improved material models for anisotropic materials, brittle materials, and chemically reacting materials challenge the numerical methods to provide greater accuracy and challenge the computer manufacturers to provide more memory and speed. Phenomena with different time and length scales need to be coupled so shock waves, structural motions, electromagnetic, and thermal effects can be analyzed in a consistent manner. Smarter codes must be developed to adapt the mesh and solution techniques to optimize the accuracy without human intervention. 

The initial and boundary conditions for the mechanical, thermal, and hydraulic effects are shown in Figure 1, with heating maintained for 100 years. (See Table 2 for the research teams and codes applied to this study). Figure 3 shows the different representations and simplifications in the models used by the various teams. This BMT was regarded as an excellent model for testing the capabilities of many alternative mathematical models and computer codes (Stephansson et al., 1996). 

Review of the literature was given by Boure et al. (1973), in which the physical mechanisms and mathematical models in use up to that time are discussed. The report has a summary of the computer codes that had been applied to the problem in the frequency and time domains. Saha (1974), Ishii (1976), Saha et al. (1976), and Saha and Zuber (1978) provided initial investigations into effects of thermal nonequilibrium between the liquid and vapor phases. Lahey and Drew (1980) discussed instability issues associated with LWRs. 

Accident analysis calculations in which the thermal effect of a reactivity insertion may be determined are normally carried out by means of a computer code or combination of codes in which the reactor kinetics description is coupled with the thermal-hydraulics model. For example, the previously described vapor growth models become more meaningful for accident analysis if they are coupled with a reactivity feedback model 14). 

The code Iode has been developed to calculate iodine behavior in a reactor containment and the auxiliary building it is part of the French computer system Escadre which, similar to the US Source Term Code Package, describes the whole sequence of a severe reactor accident. The general philosophy of Iode is to model the main phenomena which may influence the behavior of iodine in the reactor containment IS chemical reactions are modelled, concerning both the water phase and the gas phase (Gauvain et al., 1991). Similar to Impair, radiolysis is not described in detail but is taken into account over its global effect on iodine species. The kinetic data of the reactions were taken from the literature as far as inorganic iodine thermal reactions are concerned other kinetic data were compiled from the elaboration of the Impair 2 code. 

The TFM approach (Laureau et ah, 2015b Laureau, 2015c) has been developed specifically as a neutronic model able to take into account the precursor motion-associated phenomena and to perform coupled transient calculations with an accuracy close to that of MC calculations for the neutronics while incurring a low computational cost. This approach is based on a precalculation of the neutronic reactor response through time before the transient calculation. The results of the SERPENT MC code (Leppanen, 2013) calculations are condensed in fission matrices, keeping the time information. These hssion matrices are interpolated to take into account local Doppler and density thermal feedback effects due to temperature variations in the system. With this approach, an estimation of the neutron flux variation for any temperature and precursor distribution in the reactor can be very quickly obtained. 

---