Extension of the lumped parameter code COCOSYS and a model for the simulation of momentum dominated radiation during severe accidents

The aim of this work is the optimisation of the current simulations to calculate the distribution of (inflammable) gases by jets that may occur during severe accidents in buildings e.g. in the containment of a nuclear power plant (npp). Therefore the lumped parameter code COCOSYS is extended by a model to simulate momentum dominated jets. COCOSYS is a mechanistic code designed for fast simulations of all essential phenomena of normal operation and severe incidents in containments of npp. The code is under development by the Gesellschaft fuer Anlagen- und Reaktorsicherheit (GRS) mbH. lt is the aim of this thesis to improve the prediction accuracy of concentration accumulations of e.g. hydrogen. Up to now, there is no jet model in COCOSYS available. In order to simulate the spreading of fluid jets a fan system model is used. The results of fan systems are usable but suffer from disregarded atmospherical interactions. Moreover, the eligible user influence represents a potential error source. The new developed and al ready in COCOSYS implemented jet model considers atmospheri interactions, reduces the user influence and simplifies the appliance of the code. The improvements by the jet model were investigated in a successful model validation on two HY JET tests and one THAI test that are compared to the simulation results using fan systems. The HY JET tests investigated the release of helium as hydrogen substitute under increased steam influence at jet velocities between 2 and 100 m/s (test J x 4) and 530 to 900 m/s (test J x 8.1) respectively. By means of the test THAI HM-4 the lower scope of the jet model can be validated. All results document a high performance of the jet model compared to the experimental values. Especially the calculated convection and the concentration distribution could be enhanced considerably in comparison to the simulation results using fan systems. To conclude the model validation, a hypothetical scenario of a severe accident with damage of the system and a resulting hydrogen and steam release has been assumed in order to assess the simulation ability. In this case no real/experimental values exist, but the simulation provided physically reasonable results. The aim to improve the prediction accuracy of concentration accumulations of hydrogen in accident scenarios has been achieved. lt could be shown that the model can also contribute to the improvement of accident management measures e.g. by precautionary identification of potential gas sink areas which then can. be equipped with autocatalytic recombiners as a means of hydrogen degeneration.