Implementation of POD and DMD methods in Apache Spark framework for simulation of unsteady turbulent flow in the model combustor

  • Michael D. Kalugin
  • Sergei V. Strijhak
Abstract.

The paper is devoted to modelling and analysis of unsteady turbulent flow in a model combustor (channel) using LES (Large Eddy Simulation). Simulations were provided for 2D and 3D cases on different grids of a flow in a channel with rearward facing step. The calculation of a flow in a channel was performed on high performance cluster, using the new approach in Apache Spark framework and POD, DMD data processing algorithms. First 4 dynamic modes were defined.

Keywords: Rearward facing step; Turbulent Flow; Finite volume method; Grid; Computational Fluid Dynamics; Vortices; Apache Spark; POD; DMD; snapshots; modes.
The name of the conference, volume, number, page: VII European Congress on Computational Methods in Applied Sciences and Engineering
Download of the paper (pdf)