Scripts to help with Nalu-Wind
Contents
- Plot mesh script: plotmesh.py
- Mesh refinement script: buildrefinemesh.sh
- Slice mesh utility: slicemesh.sh
- Plot FAST output script: plotFAST.py
- Backup and write a restart YAML file: restartbackupnalu.py
- Plot sample planes: plotSamplePlaneGUI.py
- Convert sample planes to VTK: convertSamplePlane2VTK.py
[plotmesh.py](plotmesh.py): Plots the mesh refinement levels, turbine locations, and cut-slices
$ module load canopy
$ plotmesh.py YAMLFILEHere YAMLFILE is a yaml file containing the mesh definition, refinement windows, and (optionally) the slice mesh parameters.
Output:
Each different colored rectangle represents a halving of the mesh resolution (8x refinement). The arrow in the middle of the domain points in the wind direction.
If you include the slice_mesh section in the YAML file, then it will also include the areas where the sections are being taken, like this:
The plotmesh.py script will also plot the probes in the data_probes section.
If you include the following in your input file
data_probes:
exodus_name: exoprobe/probes3.exo
output_frequency: 1
search_method: stk_kdtree
search_tolerance: 1.0e-5 #1.0e-3
search_expansion_factor: 2.0
specifications:
- name: probe_surface
from_target_part: Unspecified-2-HEX
line_of_site_specifications:
- name: Probe1 # boostProbes2/probes/Probe1
number_of_points: 3
tip_coordinates: [10, 0, 0]
tail_coordinates: [20, 0, 0]
- name: Probe2 # boostProbes2/probes/Probe2
number_of_points: 11
tip_coordinates: [100, 0, 0]
tail_coordinates: [120, 0, 0]
plane_specifications:
- name: boostProbes2/planes/p1/Plane1
corner_coordinates: [25, -100, 0]
edge1_vector: [450, 0, 0]
edge2_vector: [0, 200, 0]
edge1_numPoints: 51
edge2_numPoints: 21
#offset_vector: [0, 0, 1]
#offset_spacings: [0, 100]
- name: ./Plane2 #boostProbes2/planes/p2/Plane2
edge1_numPoints: 31
edge2_numPoints: 31
corner_coordinates: [100, 0, 0]
edge1_vector: [10, 0, 0]
edge2_vector: [0, 10, 0]
offset_vector: [1, 0, 0]
offset_spacings: [0, 100]
- name: boostProbes2/planes/p3/Plane3
edge1_numPoints: 21
edge2_numPoints: 21
corner_coordinates: [126, -70, -70]
edge1_vector: [0, 140, 0]
edge2_vector: [0, 0, 140]
offset_vector: [1, 0, 0]
offset_spacings: [-20, 20]
output_variables:
- field_name: velocity
field_size: 3Then the output of plotmesh.py will look like:
[buildrefinemesh.sh](buildrefinemesh.sh): Creates a mesh, does local refinement
buildrefinemesh.sh YAMLFILE [OPTIONS]
Arguments
YAMFILE : a yaml file containing the mesh definitions
Options:
-o|--output-mesh OUTFILE : Output filename (default: refinedmesh.exo)
-n|--ncores NPU : Number of cores to use (default: 8)
--no-createmesh : Do not create basic mesh
--no-preproc : Do not use the preprocessor to define mesh refinem ent zones
--no-refine : Do not use mesh_adapt to refine mesh
-h|--help : This help file
The argument YAMLFILE should point to a yaml input file like this:
nalu_abl_mesh:
output_db: mesh_abl.exo
spec_type: bounding_box
fluid_part_name: fluid_part
vertices:
- [0.0, 0.0, 0.0]
- [1000.0, 1000.0, 300.0]
mesh_dimensions: [100, 100, 30]
xmin_boundary_name: west
xmax_boundary_name: east
ymin_boundary_name: south
ymax_boundary_name: north
zmin_boundary_name: lower
zmax_boundary_name: upper
# Mandatory section for Nalu preprocessing
nalu_preprocess:
# Name of the input exodus database
input_db: mesh_abl.exo
# Name of the output exodus database
output_db: mesh_abl.exo
# Nalu preprocessor expects a list of tasks to be performed on the mesh and
# field data structures
tasks:
- mesh_local_refinement
mesh_local_refinement:
fluid_parts: [fluid_part]
write_percept_files: true
percept_file_prefix: adapt
search_tolerance: 11.0
turbine_locations:
- [ 200.0, 200.0, 0.0 ]
- [ 230.0, 300.0, 0.0 ]
turbine_diameters: 15.0 # Provide a list for variable diameters
turbine_heights: 50.0 # Provide a list for variable tower heights
orientation:
type: wind_direction
wind_direction: 225.0
refinement_levels: # Numbers are for upstream, downstream, lateral and vertical length in turbine diameters
- [ 7.0, 12.0, 7.0, 7.0 ]
- [ 5.0, 10.0, 5.0, 5.0 ]
- [ 3.0, 6.0, 3.0, 3.0 ]When you execute the script, for instance
$ ./buildrefinemesh.sh testmesh.yaml -o testmesh.exothe first stage should be the mesh creation part
Nalu ABL Mesh Generation Utility
Input file: testmesh.yaml
HexBlockBase: Registering parts to meta data
Mesh block: fluid_part
Num. nodes = 41616; Num elements = 37500
Generating nodes...done
Generating elements...done
Creating element connectivity... done
Generating X Sideset: west
Generating X Sideset: east
Generating Y Sideset: south
Generating Y Sideset: north
Generating Z Sideset: lower
Generating Z Sideset: upper
Finalizing bulk data modifications ... done
Generating coordinates...
Generating x spacing: constant_spacing
Generating y spacing: constant_spacing
Generating z spacing: constant_spacing
Writing mesh to file: mesh_abl.exo
Followed by the preprocessing stage which marks out areas for local refinement:
Nalu Preprocessing Utility
Input file: testmesh.yaml
Found 1 tasks
- mesh_local_refinement
Performing metadata updates...
Metadata update completed
Reading mesh bulk data... done.
--------------------------------------------------
Begin task: mesh_local_refinement
Then it will run mesh_adapt through the multiple stages of refinement:
Running mesh refinement step
------------------
STAGE 1 REFINEMENT: tempmesh0.e --> tempmesh1.e
------------------
mpirun -n 8 mesh_adapt --refine=DEFAULT --input_mesh=tempmesh0.e --output_mesh=tempmesh1.e --RAR_info=adapt1.yaml --ioss_read_options="auto-decomp:yes"
INFO: ioss_read_options=auto-decomp:yes ioss_write_options=
PerceptMesh:: opening tempmesh0.e
Using decomposition method 'RIB' on 8 processors.
At the very end, it will copy over the final mesh and report the new mesh blocks which are included:
‘tempmesh2.e’ -> ‘testmesh.exo’
New mesh blocks:
eb_names =
"fluid_part",
"fluid_part.pyramid_5._urpconv",
"fluid_part.tetrahedron_4._urpconv",
"fluid_part.pyramid_5._urpconv.Tetrahedron_4._urpconv" ;
}
[slicemesh.sh](slicemesh.sh): Creates the slice geometry for a mesh
./slicemesh.sh MESHYAMLFILE [OPTIONS]where MESHYAMLFILE is the yaml file containing the slice mesh parameters.
Optional Arguments:
-y|--yaml-output YAMLFILE Print out the corresponding yaml inputs needed to include the slice mesh
output during the simulation. YAMLFILE is the main simulation yamlfile
-v|--extra-vars VARLIST An extra list of variables to include when writing out the sliced meshes.
VARLIST is of the form "var1:N var2:N ..." where var1, var2, are the
variable names, and N is the number of components for that variableThe MESHYAMLFILE input file defining the slices to take:
slice_mesh:
output_db: temp.exo # sliceplanes.exo
slices:
# X-Y plane
- axis1: [1.0, 0.0, 0.0]
axis2: [0.0, 1.0, 0.0]
axis3: [0.0, 0.0, 1.0]
origin: [0.0, 0.0, 0.0]
grid_lengths: [2500.0, 2500.0]
grid_dx: [4.0, 4.0]
num_planes: 1
plane_offsets: [0.0]
part_name_prefix: turbineHH
# Y-Z plane
- axis1: [1.0, 1.0, 0.0]
axis2: [0.0, 0.0, 1.0]
axis3: [-1, 1.0, 0.0]
origin: [0.0, 00, -100.0]
grid_lengths: [3000.0, 200.0]
grid_dx: [4.0, 4.0]
num_planes: 1
plane_offsets: [0.0]
part_name_prefix: turbineSlice2
If you execute it with the input files:
$ slicemesh.sh mesh1.yaml
yamlfile = mesh1.yaml
simyamlfile =
extravars =
Loading modules
Output mesh name = temp.exo
Slice Mesh Generation Utility
Input file: mesh1.yaml
Loading slice inputs...
Initializing slices...
Slice: Registering parts to meta data:
- turbineHH_1
Slice: Registering parts to meta data:
- turbineSlice2_1
Generating slices for: turbineHH
Creating nodes... 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Creating elements... 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Generating coordinate field
- turbineHH_1
Generating slices for: turbineSlice2
Creating nodes... 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Creating elements... 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Generating coordinate field
- turbineSlice2_1
Writing mesh to file: temp.exo
Memory usage: Avg: 156.043 MB; Min: 156.043 MB; Max: 156.043 MBIf you execute it with the -y option, like slicemesh.sh mesh1test.yaml -y alm_simulation.yaml, then some additional output will be generated:
# === Auto-generated YAML below ======
# Goes under [realms:]
- name: ioRealm
mesh: temp.exo
type: input_output
automatic_decomposition_type: rcb
field_registration:
specifications:
- field_name: velocity_slice
target_name: &fieldreg [ turbineHH_1, turbineSlice2_1 ]
field_size: 3
field_type: node_rank
output:
output_data_base_name: ./sliceDataInstantaneous/temp.exo
output_frequency: 1
output_node_set: no
output_variables:
- velocity_slice
transfers:
- name: turbineHH_1
type: geometric
realm_pair: [realm_1, ioRealm]
to_target_name: turbineHH_1
from_target_name: ['fluid_part', 'fluid_part.Pyramid_5._urpconv', 'fluid_part.Tetrahedron_4._urpconv']
objective: input_output
transfer_variables:
- [velocity, velocity_slice]
- name: turbineSlice2_1
type: geometric
realm_pair: [realm_1, ioRealm]
to_target_name: turbineSlice2_1
from_target_name: ['fluid_part', 'fluid_part.Pyramid_5._urpconv', 'fluid_part.Tetrahedron_4._urpconv']
objective: input_output
transfer_variables:
- [velocity, velocity_slice]
# === End auto-generated YAML ======These yaml parameters can be added to alm_simulation.yaml to extract
the slice during simulations.
[plotFAST.py](plotFAST.py): Plots FAST output
$ module load canopy
$ plotFAST.py FAST.T1.out [FAST.T2.out ... ]Output:
It's pretty self-explanatory. Check the variables on the left you would like to plot, and hit Plot.
If the output files get updated, hit Reload data to reread the files from disk.
[restartbackupnalu.py](restartbackupnalu.py): Retarts and backups up a simulation from a YAML file
This script takes in the current YAML input file, then automatically sets up the right restart parameters and spits out a new YAML file. Works on both ABL and FAST turbine runs. Optionally backups all the restart/output files so they don't get overwritten.
usage: restartbackupnalu.py [-h] [--dobackup] [--Nsteps NSTEPS]
[--suffix SUFFIX]
yamlfile [yamlfile ...]
Create a restart YAML file for Nalu.
positional arguments:
yamlfile
optional arguments:
-h, --help show this help message and exit
--dobackup Backup files [default=False]
--addNsteps ADDNSTEPS
Add another ADDNSTEPS to the run [default 100]
--runToNsteps RUNTONSTEPS
Run until RUNTONSTEPS are reached [default is
ADDNSTEPS, not RUNTONSTEPS]
--suffix SUFFIX Suffix to attach to backup files [default is date/time
based suffix]
[plotSamplePlaneGUI.py](plotSamplePlaneGUI.py): GUI to plot the sample planes
usage: plotSamplePlaneGUI.py [-h] [--nogui] [--planenum PLANENUM]
[--varnum VARNUM]
[PLANEFILE [PLANEFILE ...]]
Plot sample mesh
positional arguments:
PLANEFILE Plot this sample plane
optional arguments:
-h, --help show this help message and exit
--nogui Use command line only [default=False]
--planenum PLANENUM Plot this plane number
--varnum VARNUM Plot this variable number
For example, let's say you created a set of planes with the following specifications:
specifications:
- name: probe_surface
from_target_part: Unspecified-2-HEX
plane_specifications:
- name: Probes/Plane3
edge1_numPoints: 21
edge2_numPoints: 21
corner_coordinates: [126, -70, -70]
edge1_vector: [0, 140, 0]
edge2_vector: [0, 0, 140]
offset_vector: [1, 0, 0]
offset_spacings: [-20, 20]
output_variables:
- field_name: velocity
field_size: 3These planes can be plotted by loading them in plotSamplePlaneGUI.py:
$ plotSamplePlaneGUI.py Plane3*_1.datOn the left hand side there are the different plane files, plot
variables, and plane numbers in each file. Select the parameters to
display, and hit the Plot button to see something like:
[convertSamplePlane2VTK.py](convertSamplePlane2VTK.py): This script converts the text output sample planes to ASCII VTK format.
usage: convertSamplePlane2VTK.py [-h] [--planenum PLANENUM]
PLANEFILE [PLANEFILE ...]
Convert sample planes to ASCII VTK format
positional arguments:
PLANEFILE Sample plane file(s) to convert
optional arguments:
-h, --help show this help message and exit
--planenum PLANENUM Convert only this offset plane number [default: convert
all planes]$ convertSamplePlane2VTK.py --planenum 0 HHplane_0009[0-3]*.dat
Converting HHplane_0009000_0.dat
-> writing HHplane_0009000_0_plane0.vtk
Converting HHplane_0009100_0.dat
-> writing HHplane_0009100_0_plane0.vtk
Converting HHplane_0009200_0.dat
-> writing HHplane_0009200_0_plane0.vtk
Converting HHplane_0009300_0.dat
-> writing HHplane_0009300_0_plane0.vtk