Time Varying Inlet BC for 3D Simulations

[ErcanUmut60]

Hello everybody,

I am using a custom inlet BC that is given in the pywraapper Test Cases. I want to increase the pressure at the inlet with time. This code works fine for 2D simulations but it diverges after some initial time steps in 3D. I couldn't figure out what I should change in the code for 3D nozzle simulations. Should I change anything for 3D?

#!/usr/bin/env python

## \file run.py
#  \brief Unsteady inlet boundary conditions.
#  \version 8.1.0 "Harrier"

import sys
import pysu2
import math
# from mpi4py import MPI

def main():
  """
  Run the flow solver with a custom inlet (function of time and space).
  """
  # comm = MPI.COMM_WORLD
  comm = 0

  # Initialize the primal driver of SU2, this includes solver preprocessing.
  try:
    driver = pysu2.CSinglezoneDriver('lam_flatplate.cfg', 1, comm)
  except TypeError as exception:
    print('A TypeError occured in pysu2.CSinglezoneDriver : ', exception)
    raise

  # Get the ID of the inlet marker.
  all_marker_ids = driver.GetMarkerIndices()
  marker_name = 'x_minus'
  marker_id = all_marker_ids[marker_name] if marker_name in all_marker_ids else -1

  # Run the time loop in python to vary the inlet conditions.
  dt = driver.GetUnsteadyTimeStep()

  print("\n------------------------------ Begin Solver -----------------------------")
  sys.stdout.flush()

  for time_iter in range(driver.GetNumberTimeIter()):
    # Change the total pressure.
    if marker_id >= 0:
      for i_vertex in range(driver.GetNumberMarkerNodes(marker_id)):
        y = driver.MarkerCoordinates(marker_id)(i_vertex, 1)
        t = time_iter * dt
        pt = 1e5 + 2e4 * y / 0.01 * (1 - math.cos(2 * math.pi * t / 0.1))
        driver.SetMarkerCustomInletFlowVar1(marker_id, i_vertex, pt)
    driver.BoundaryConditionsUpdate()

    driver.Preprocess(time_iter)

    # Run one time iteration.
    driver.Run()
    driver.Postprocess()
    driver.Update()

    # Monitor the solver and output solution to file if required.
    driver.Monitor(time_iter)
    driver.Output(time_iter)

  # Finalize the solver and exit cleanly.
  driver.Finalize()

if __name__ == '__main__':
  main()

[pcarruscag]

Nothing else should be necessary if you are only changing pressure.

[ErcanUmut60]

Initial conditions should be okay because a colleague run this on another code in the same manner and it worked. We have an outlet boundary which is 1atm, other than that I don't know if there is any other instance that can cause a reverse flow.

[pcarruscag]

care to post your su2 config?

[ErcanUmut60]

Yeah, sure.

nozzle.txt

[pcarruscag]

Use FGMRES instead of BCGSTAB.
Make sure you have enough inner iterations to converge each time step reasonably well.
You could try running a steady state problem with the initial pressure and initialize from that instead of freestream conditions, it may not be what you want to do but it could help rule out issues with the initial conditions.

[pcarruscag]

You could also try ramping up from the initial pressure you have (atmospheric) instead of starting from 0.6MPa