3D-Airfoil

[Stian Hjorteland]

Hi guys.

As a part of my master thesis i am simulating flow on an airfoil (90 deg) in openfoam. I also want to compare my results with pyfr (3rd order). Somehow, my setup at the present moment wont start to simulate. I am using Ubuntu on a Windows computer (through Oracle Virtual Box).

My attempt on a setup for the .ini is attached. It is worth mentioning that the mesh is an O-grid from pointwise with 1000000 cells. The .pyfrm file is therefore made by pointwise when exporting and should be fine which is why i think the problem is in the .ini file.

I would be grateful for any tips as i am a new user of PyFR and find it a bit different from openfoam.

Best regards Stian

[Giorgio Giangaspero]

Hi Stian

do you get any error messages?

best

Giorgio Giangaspero

--
You received this message because you are subscribed to the Google Groups "PyFR Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pyfrmailingli...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/pyfrmailinglist/33e1c7ac-ca43-4218-8cd1-506cecb53e38%40googlegroups.com.

[Stian Hjorteland]

Hi, yes:

msk600@virtual:~/AirfoilPyFR$ export OMP_NUM_THREAD=1
msk600@virtual:~/AirfoilPyFR$ pyfr run -b openmp -p Airfoil-DU96-w-180.pyfrm Airfoil-DU96-w-180.ini
Traceback (most recent call last):
  File "/usr/bin/pyfr", line 11, in <module>
    load_entry_point('pyfr==1.5.0', 'console_scripts', 'pyfr')()
  File "/usr/share/pyfr/pyfr/__main__.py", line 110, in main
    args.process(args)
  File "/usr/share/pyfr/pyfr/__main__.py", line 235, in process_run
    args, NativeReader(args.mesh), None, Inifile.load(args.cfg)
  File "/usr/share/pyfr/pyfr/__main__.py", line 216, in _process_common
    solver = get_solver(backend, rallocs, mesh, soln, cfg)
  File "/usr/share/pyfr/pyfr/solvers/__init__.py", line 14, in get_solver
    return get_integrator(backend, systemcls, rallocs, mesh, initsoln, cfg)
  File "/usr/share/pyfr/pyfr/integrators/__init__.py", line 43, in get_integrator
    return integrator(backend, systemcls, rallocs, mesh, initsoln, cfg)
  File "/usr/share/pyfr/pyfr/integrators/std/controllers.py", line 12, in __init__
    super().__init__(*args, **kwargs)
  File "/usr/share/pyfr/pyfr/integrators/std/steppers.py", line 165, in __init__
    super().__init__(*args, **kwargs)
  File "/usr/share/pyfr/pyfr/integrators/std/steppers.py", line 8, in __init__
    super().__init__(*args, **kwargs)
  File "/usr/share/pyfr/pyfr/integrators/std/base.py", line 12, in __init__
    super().__init__(*args, **kwargs)
  File "/usr/share/pyfr/pyfr/integrators/base.py", line 59, in __init__
    self.system = systemcls(backend, rallocs, mesh, initsoln, nreg, cfg)
  File "/usr/share/pyfr/pyfr/solvers/base/system.py", line 34, in __init__
    eles, elemap = self._load_eles(rallocs, mesh, initsoln, nreg, nonce)
  File "/usr/share/pyfr/pyfr/solvers/base/system.py", line 107, in _load_eles
    eles.set_backend(self.backend, nreg, nonce)
  File "/usr/share/pyfr/pyfr/util.py", line 48, in __call__
    return proxylist(x(*args, **kwargs) for x in self)
  File "/usr/share/pyfr/pyfr/util.py", line 48, in <genexpr>
    return proxylist(x(*args, **kwargs) for x in self)
  File "/usr/share/pyfr/pyfr/solvers/navstokes/elements.py", line 12, in set_backend
    super().set_backend(backend, nscalupts, nonce)
  File "/usr/share/pyfr/pyfr/solvers/baseadvecdiff/elements.py", line 26, in set_backend
    super().set_backend(backend, nscal_upts, nonce)
  File "/usr/share/pyfr/pyfr/solvers/baseadvec/elements.py", line 26, in set_backend
    super().set_backend(backend, nscal_upts, nonce)
  File "/usr/share/pyfr/pyfr/solvers/base/elements.py", line 179, in set_backend
    for i in range(nscal_upts)]
  File "/usr/share/pyfr/pyfr/solvers/base/elements.py", line 179, in <listcomp>
    for i in range(nscal_upts)]
  File "/usr/share/pyfr/pyfr/backends/base/backend.py", line 20, in newfn
    m = fn(self, *args, **kwargs)
  File "/usr/share/pyfr/pyfr/backends/base/backend.py", line 133, in matrix
    return self.matrix_cls(self, ioshape, initval, extent, aliases, tags)
  File "/usr/share/pyfr/pyfr/backends/base/types.py", line 110, in __init__
    aliases, tags)
  File "/usr/share/pyfr/pyfr/backends/base/types.py", line 65, in __init__
    backend.malloc(self, extent)
  File "/usr/share/pyfr/pyfr/backends/base/backend.py", line 71, in malloc
    obj.onalloc(data, 0)
  File "/usr/share/pyfr/pyfr/backends/openmp/types.py", line 22, in onalloc
    self._set(self._initval)
  File "/usr/share/pyfr/pyfr/backends/openmp/types.py", line 31, in _set
    self.data[:, :self.ncol] = self._pack(ary)
  File "/usr/share/pyfr/pyfr/backends/base/types.py", line 88, in _pack
    mode='constant')
  File "/usr/lib/python3/dist-packages/numpy/lib/arraypad.py", line 1363, in pad
    newmat = narray.copy()
MemoryError
msk600@virtual:~/AirfoilPyFR$

Any ideas?

Best regards Stian

fredag 21. februar 2020 10.33.36 UTC+1 skrev Giorgio Giangaspero følgende:

Hi Stian

do you get any error messages?

best

Giorgio Giangaspero

On Thu, Feb 20, 2020 at 3:14 PM Stian Hjorteland <stian.h...@gmail.com> wrote:

Hi guys.

As a part of my master thesis i am simulating flow on an airfoil (90 deg) in openfoam. I also want to compare my results with pyfr (3rd order). Somehow, my setup at the present moment wont start to simulate. I am using Ubuntu on a Windows computer (through Oracle Virtual Box).

My attempt on a setup for the .ini is attached. It is worth mentioning that the mesh is an O-grid from pointwise with 1000000 cells. The .pyfrm file is therefore made by pointwise when exporting and should be fine which is why i think the problem is in the .ini file.

I would be grateful for any tips as i am a new user of PyFR and find it a bit different from openfoam.

Best regards Stian

--
You received this message because you are subscribed to the Google Groups "PyFR Mailing List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pyfrmai...@googlegroups.com.

[Vishal Saini]

Hi Stian,

A 1 million elements mesh with polynomial order 3 is not a small problem, of course, depending on your machine.

You might be exceeding the memory limit. Have you had a look at that?

BR,

Vishal

---
Vishal SAINI

Master of Research,
University of Cambridge.
Master in Turbulence
EC Lille, ENSIP and ENSMA France.

--

You received this message because you are subscribed to the Google Groups "PyFR Mailing List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pyfrmailingli...@googlegroups.com.

[Stian Hjorteland]

Hi guys!

Sorry for the late reply, i was on a small vacation.

The memory limit is not a problem in this case. I just checked. 

fredag 21. februar 2020 13.14.05 UTC+1 skrev Vishal Saini følgende:

Hi Stian,

A 1 million elements mesh with polynomial order 3 is not a small problem, of course, depending on your machine.

You might be exceeding the memory limit. Have you had a look at that?

BR,

Vishal

---
Vishal SAINI

Master of Research,
University of Cambridge.
Master in Turbulence
EC Lille, ENSIP and ENSMA France.

On Thu, Feb 20, 2020 at 3:14 PM Stian Hjorteland <stian.h...@gmail.com> wrote:

Hi guys.

As a part of my master thesis i am simulating flow on an airfoil (90 deg) in openfoam. I also want to compare my results with pyfr (3rd order). Somehow, my setup at the present moment wont start to simulate. I am using Ubuntu on a Windows computer (through Oracle Virtual Box).

My attempt on a setup for the .ini is attached. It is worth mentioning that the mesh is an O-grid from pointwise with 1000000 cells. The .pyfrm file is therefore made by pointwise when exporting and should be fine which is why i think the problem is in the .ini file.

I would be grateful for any tips as i am a new user of PyFR and find it a bit different from openfoam.

Best regards Stian

--
You received this message because you are subscribed to the Google Groups "PyFR Mailing List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pyfrmai...@googlegroups.com.