Finite-volume Eulerian solver for simulation of particle-laden flows for icing applications

2021 
Abstract Particle-laden flows can be encountered in a plethora of engineering applications, e.g., sediment transport, fluidized beds, vehicle soiling, and aircraft icing. The latter is caused by supercooled water droplets or ice crystals impacting and adhering to an aerodynamic surface during flight. Simulating such flows requires proper application of physical models through the use of particle flow solvers, which can rely on Lagrangian or Eulerian formulations for the solution of particle dynamics. Although Lagrangian formulations present certain advantages in terms of time resources and simplicity of implementation, Eulerian solvers can easily provide complete details of the particle flow field, such as volume fraction distribution. Several Eulerian solvers exist for the particle flow, although they rely on systems of coupled transport equations, which can be further simplified in order to reduce modeling and computational effort. Therefore, the current study presents an elaborated 2D finite-volume Eulerian solver for the resolution of the dynamics of particle-laden flows through the use of a system of decoupled transport equations. Additionally, we introduced a modification into the traditional decoupled formulation in order to allow the implementation of 2nd order upwind schemes for the convective terms. Two cases are used as test benchmarks to show that the solver effectively achieves mesh independence, convergence stability, and 2nd order accuracy. Also, the solver physical modeling capabilities are evaluated by comparing results against the ones obtained through the Lagrangian formulation, the Ansys FLUENT Eulerian solver and a set of experimental results. Therefore, the elaborated Eulerian solver can be used to facilitate the modeling of in-flight icing phenomena while still retaining the accuracy of traditional formulations.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    26
    References
    0
    Citations
    NaN
    KQI
    []
    Baidu
    map