Developmental assessment of RELAP5/MOD3.1 with separate-effect and integral test experiments: model changes and options

A summary of modifications and options introduced in RELAP5/MOD3.1 (R5M3.1) is presented and it is shown that the predicting capabilities of the modified version of the code are greatly improved, while the general philosophy we followed in arriving at these modifications is also outlined. These changes which are the same ones we implemented in the past in the version 7j of the code, include 2 different heat transfer packages (one of them activated during reflooding), modification of the low mass-flux Groeneveld CHF look-up table and of the dispersed flow interfacial area (and shear) as well as of the criterion for transition into and out from this regime, almost complete elimination of the under-relaxation schemes of the interfacial closure coefficients etc. The modified R5M3.1 code is assessed against a number of separate-effect and integral test experiments and in contrast to the frozen version, is shown to result in physically sound predictions which are much closer to the measurements, while almost all the predicted variables are free of unphysical spurious oscillations. The modifications introduced solve a number of problems associated with the frozen version of the code and result in a version which can be confidently used both for SB-LOCA and LB-LOCA analyses. (author) 7 figs., 20 refs