Evaluation of the proper coherence representation in random flow generation based methods

The interest in synthetic methods relies in the fact that reliability of methodologies such as large eddy simulations (LES) or direct numerical simulations strongly depends on how well the developed turbulence is characterized, which generally leads to computationally expensive simulations. Turbulence generation methods allow the initialization and up-keeping of the velocity fluctuations field to promote the observed/needed turbulence in the flow. In this work the methodology known as ‘modified discretizing and synthesizing random flow generation’ (MDSRFG) jointly with a LES method is analyzed for its use in the study of tall building aerodynamics. A comparison with other generation techniques, that are closely related by their features and their conceptual origins, is presented with particular emphasis on the correct representation of the coherence of the velocity field. Particularly, an expression for the coherency function for the MDSRFG is derived. After an analysis and revision of these generation methodologies, the turbulent air flow around a rectangular prismatic model is computationally simulated. A comparison of the results obtained from different methods is performed. The resulting wind loads on the model, along with the statistical characteristics of the flow, show that the MDSRFG technique allows to represent a field of spatially correlated velocities correctly. Fil: Castro, Hugo Guillermo. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Tecnológica Nacional. Facultad Regional Resistencia. Grupo de Investigación en Mecanica de Fluidos; Argentina Fil: Paz, Rodrigo Rafael. Livermore Software Technology Corporation; Estados Unidos. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Santa Fe. Centro de Investigaciones en Métodos Computacionales. Universidad Nacional del Litoral. Centro de Investigaciones en Métodos Computacionales; Argentina Fil: Mroginski, Javier Luis. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Nacional del Nordeste. Facultad de Ingeniería; Argentina Fil: Storti, Mario Alberto. Universidad Nacional del Litoral; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Santa Fe. Centro de Investigaciones en Métodos Computacionales. Universidad Nacional del Litoral. Centro de Investigaciones en Métodos Computacionales; Argentina