The use of Computational Fluid Dynamics (CFD) for industrial applications often implies the capability of dealing with geometries which are large with respect to the characteristic dimensions of the involved physical phenomena.
Such situations arise for instance when computing the pressure loads generated by a hydrogen-air combustion occurring in a nuclear reactor containment during a postulated Loss of Coolant Accident. The free volume of the European Pressurized Reactor building is about 75000 m3 while the characteristic physical lengths of the combustion flame are much smaller: the reaction zone in a laminar deflagration at atmospheric condition can vary from about 1 mm to 10 mm; consequently, the direct simulation of flame propagation and deflagration-to-detonation transition (DDT) in such a large-scale geometry requires prohibitively large mesh sizes. Alternatively, the direct simulation can be avoided by considering the flame as infinitely thin and by modeling the diffusion effects through phenomenological laws for the flame speed. The combustion-induced pressure loads can then be correctly predicted provided the flame speed is correctly estimated.
In this work we present the strategy we follow to compute combustions in large geometries, with partic-ular emphasis to the numerical approach (based on the so-called Reactive Discrete Equation Method [1, 2]). Numerical experiments are analyzed to investigate the accuracy and the robustness of the numerical method. Conclusions are drawn and some perspectives are outlined.
Tuesday 23, 15.30 -16.30, room G
Wednesday 24, 15.30-17.30, room F
Strada le Grazie 15
Partita IVA 01541040232
Codice Fiscale 93009870234
© 2017 | Università degli studi di Verona | Credits