Benchmark solutions for the assessment of computer codes based on the shallow water equations

TASSI PA; VIONNET CA

Paraná, Prov. Entre Ríos

Congreso; MECOM 2002; 2002

AMCA - Asociación Argentina de Mecánica Computacional

Several finite element codes for solving the two-dimensional shallow water model are currently available for engineering applications. While most of these codes are capable of computing the water surface elevation and the depth-averaged velocity of free-surface flows in any given geometry, the computing process itself is sometimes hampered with unpleasant consequences for the inexperience modeler. Some of these unwelcome problems may be triggered by the algorithms built into the codes (i.e., depending on the method used to incorporate the required amount of 'upwinding' to cope with the convective terms), or by the uncertainty present in the values of the model parameters that may influence the flow physics. In this communication, the performance of some codes is tested. To that aim, we shall present the discussion of few benchmark solutions that can be used to measure the performance of any computational code based on the shallow water equations. Some of them correspond to rather unrealistic situations that have the purpose of illustrating the capabilities and/or the accuracy/inaccuracy achieved by the codes being tested. Other benchmark solution is related to a real engineering problem that is ambiguously modeled with some of the shallow water codes.