3D Euler Equations

The 3D Euler equations model inviscid, compressible flows in two spatial dimensions. It is configured by setting the name in the equation table to euler.

Here is an example for the equation table of this equation system:

  equation = {
    name      = 'euler',
    isen_coef = 1.4,
    r         = 287,
    numflux   = 'hll',
    material = {
      characteristic = 0.0,
      relax_velocity = 0.0,
      relax_temperature = 0.0
    }
  }
  equation.cv = equation.r / (equation.isen_coef - 1.0)

The fluid is described by the ideal gas constant (equation.r), the isentropic expansion coefficient (equation.isen_coef) and the isochoric specific heat (equation.cv). See atl_eqn_euler_module for all options, and the overview example for a configuration file with the usual options that can serve as a template for your own configurations.

Note: you have to use the modg scheme to compute this equation system (scheme.spatial.name = 'modg').

Available setups are:

  • Pulse in density with derived quantity: shows the tracking of derived quantities like kinetic energy and velocity.

  • Pulse in density with FPT: simple setup with FPT projection between modal and nodal representation.

  • Pulse in density with FXT: simple setup with FXT projection between modal and nodal representation. This uses the fast multipole method, implemented in FXTPACK.

  • Pulse in density with L2P: illustrates the use of the L2 projection for transformations between modal and nodal representation.

  • Pulse in density with sponge: shows the use of a sponge to dampen out the state.

  • Modal Estimate: small setup with a pressure pulse that shows the use of modal estimation for the adaptive timestep computation.

  • Multilevel: modelling of a wall by penalization in a multilevel mesh. Employs the IMEX time integration scheme (needed for efficient handling of penalization terms).

  • Overview: this example illustrates the overall options for Euler simulations and can serve as a template for your configuration files.

  • P polynomials: a shear layer in a rectangular tube. This setup makes use of the P-Polynomial construction for the multidimensional polynomials.

  • Q-Criterion: shows tracking of q-criterion and lambda2 values.

  • Shear layer Q4: a simple setup with a velocity layer in a periodic channel discretized with fourth order in space.

  • Shear layer Q8: same as above but with an eighth order discretization in space.

  • Shock locally refined: a shock that travels in a tube along the z-axis through a domain with some local refinement. This setup illustrates covolume with spectral filtering.

  • Shock parallel: a shock travelling through a channel along the z-axis, simulated with the Runge-Kutta Taylor scheme. Utilizes covolume and spectral filtering to stabilize the simulation.

  • Shock periodic: a shock moving through a periodic domain, progressed in time with the strong stability preserving 2-stage Runge-Kutta scheme.

  • Toro1 in X: a Riemann problem in a tube along the x-axis.

  • Toro1 in Y: a Riemann problem in a tube along the y-axis.

  • Toro1 in Z: a Riemann problem in a tube along the z-axis.
  • Track vorticity