Generate the geometry description of a directional non-equiaxed grain structure, e.g., RVE cutted from a cold-rolling sheet. '--reduct' specifies the thickness reduction after rolling, and '-n' specified the number of samples cutted, e.g., n=5, then five samples will be cutted from the sheet along theta = 0 (the rolling direction), theta = 22.5, theta = 45, theta = 67.5, and theta = 90 (the transversal direction).
This commit is contained in:
parent
657ed7ed35
commit
b69a422529
|
@ -0,0 +1,250 @@
|
||||||
|
#!/usr/bin/env python
|
||||||
|
# -*- coding: UTF-8 no BOM -*-
|
||||||
|
|
||||||
|
import threading,time,os,subprocess,shlex,string
|
||||||
|
import os,re,sys,math,string
|
||||||
|
import numpy as np
|
||||||
|
from optparse import OptionParser
|
||||||
|
import damask
|
||||||
|
from collections import defaultdict
|
||||||
|
|
||||||
|
|
||||||
|
scriptID = string.replace('$Id: geom_directionalNonEquiaxedGrain.py 4290 2015-07-24 08:41:08Z hm.zhang $','\n','\\n')
|
||||||
|
scriptName = os.path.splitext(scriptID.split()[1])[0]
|
||||||
|
|
||||||
|
def execute(cmd,streamIn=None,wd='./'):
|
||||||
|
'''
|
||||||
|
executes a command in given directory and returns stdout and stderr for optional stdin
|
||||||
|
'''
|
||||||
|
initialPath=os.getcwd()
|
||||||
|
os.chdir(wd)
|
||||||
|
process = subprocess.Popen(shlex.split(cmd),stdout=subprocess.PIPE,stderr = subprocess.PIPE,stdin=subprocess.PIPE)
|
||||||
|
if streamIn != None:
|
||||||
|
out,error = process.communicate(streamIn.read())
|
||||||
|
else:
|
||||||
|
out,error = process.communicate()
|
||||||
|
os.chdir(initialPath)
|
||||||
|
return out,error
|
||||||
|
|
||||||
|
# --------------------------------------------------------------------
|
||||||
|
# MAIN
|
||||||
|
# --------------------------------------------------------------------
|
||||||
|
identifiers = {
|
||||||
|
'grid': ['a','b','c'],
|
||||||
|
'size': ['x','y','z'],
|
||||||
|
'origin': ['x','y','z'],
|
||||||
|
}
|
||||||
|
mappings = {
|
||||||
|
'grid': lambda x: int(x),
|
||||||
|
'size': lambda x: float(x),
|
||||||
|
'origin': lambda x: float(x),
|
||||||
|
'homogenization': lambda x: int(x),
|
||||||
|
'microstructures': lambda x: int(x),
|
||||||
|
}
|
||||||
|
|
||||||
|
parser = OptionParser(option_class=damask.extendableOption, usage='%prog options [file[s]]', description = """
|
||||||
|
Generate the geometry description of a directional non-equiaxed grain structure, e.g., RVE cutted from a cold-rolling sheet.
|
||||||
|
The initial equiaxed grain structure is generated by standard Voronoi tessellation, '--reduct' specifies the thickness
|
||||||
|
reduction after rolling, and '-n' specified the number of samples cutted, e.g., n=5, then five samples will be cutted from the
|
||||||
|
sheet along \\theta = 0 (the rolling direction), \\theta = 22.5, \\theta = 45, \\theta = 67.5, and \\theta = 90 (the
|
||||||
|
transversal direction ).
|
||||||
|
|
||||||
|
""", version = scriptID)
|
||||||
|
|
||||||
|
parser.add_option('-N', dest='N', type='int', metavar='int',
|
||||||
|
help='number of seed points to distribute [%default]')
|
||||||
|
parser.add_option('-r', '--rnd', dest='randomSeed', type='int', nargs = 2, metavar=' '.join(['int']*2),
|
||||||
|
help='seed of random number generator [%default]')
|
||||||
|
parser.add_option('-m', '--microstructure', dest='microstructure', type='int', metavar='int',
|
||||||
|
help='first microstructure index [%default]')
|
||||||
|
parser.add_option('-g', '--grid', dest='grid', type='int', nargs = 3, metavar=' '.join(['int']*3),
|
||||||
|
help='a,b,c grid of hexahedral box [from seeds file]')
|
||||||
|
parser.add_option('-s', '--size', dest='size', type='float', nargs = 3, metavar=' '.join(['float']*3),
|
||||||
|
help='x,y,z size of hexahedral box [1.0 along largest grid point number]')
|
||||||
|
parser.add_option('--phase', dest='phase', type='int', metavar = 'int',
|
||||||
|
help='phase index to be used [%default]')
|
||||||
|
parser.add_option('--crystallite', dest='crystallite', type='int', metavar = 'int',
|
||||||
|
help='crystallite index to be used [%default]')
|
||||||
|
parser.add_option('-c', '--configuration', dest='config', action='store_true',
|
||||||
|
help='output material configuration [%default]')
|
||||||
|
parser.add_option('--secondphase', type='float', dest='secondphase', metavar= 'float',
|
||||||
|
help='volume fraction of randomly distribute second phase [%default]')
|
||||||
|
parser.add_option('-l', '--laguerre', dest='laguerre', action='store_true',
|
||||||
|
help='use Laguerre (weighted Voronoi) tessellation [%default]')
|
||||||
|
parser.add_option('-n', dest='number', type='int', metavar='int',
|
||||||
|
help='the angle(degree) between the longitudinal direction of RVE and the rolling direction [%default]')
|
||||||
|
parser.add_option('--reduct', dest='reduction', type='float', metavar='float',
|
||||||
|
help='thickness reduction of rolling [%default]')
|
||||||
|
parser.set_defaults(
|
||||||
|
N = 500,
|
||||||
|
grid = (200,100,50),
|
||||||
|
size = (2.0,1.0,0.5),
|
||||||
|
phase = 1,
|
||||||
|
crystallite = 1,
|
||||||
|
secondphase = 0.0,
|
||||||
|
microstructure = 1,
|
||||||
|
laguerre = False,
|
||||||
|
randomSeed = (None,None),
|
||||||
|
config = False,
|
||||||
|
number = 5,
|
||||||
|
reduction = 0.6
|
||||||
|
)
|
||||||
|
(options,filenames) = parser.parse_args()
|
||||||
|
options.grid = np.array(options.grid)
|
||||||
|
|
||||||
|
sizeX = sizeY = max(options.size[0], options.size[1])
|
||||||
|
gridX, gridY = int(np.ceil(sizeX/options.size[0]*options.grid[0]))+1,int(np.ceil(sizeX/options.size[1]*options.grid[1]))+1
|
||||||
|
gridx, gridy, gridz = options.grid; sizex, sizey, sizez = options.size
|
||||||
|
nGrids = gridx*gridy*gridz; avgGrids = nGrids/options.N
|
||||||
|
dx, dy = options.size[0]/options.grid[0], options.size[1]/options.grid[1]
|
||||||
|
|
||||||
|
Ngrains = int(np.ceil(sizeX*sizeY/options.size[0]/options.size[1]*options.N))
|
||||||
|
filename = 'grains'+str(Ngrains)+'_'+str(gridX)+str(gridY)+str(options.grid[2])
|
||||||
|
thickness = 1.0-options.reduction
|
||||||
|
|
||||||
|
print 'run seeds_fromRandom'
|
||||||
|
execute('seeds_fromRandom -N %i -g %i %i %i %s.seeds'%(Ngrains,int(gridX*thickness)+1, gridY, int(gridz/thickness)+1, filename))
|
||||||
|
print 'run geom_fromVoronoiTessellation'
|
||||||
|
execute('geom_fromVoronoiTessellation -s %s %s %s < %s.seeds'%(sizeX*thickness, sizeY, sizez/thickness,filename))
|
||||||
|
print 'run geom_rescale'
|
||||||
|
execute('geom_rescale -g %i %i %i -s %s %s %s < %s.geom > %s_scale.geom'%(gridX, gridY, gridz,
|
||||||
|
sizeX, sizeY, sizez,filename,filename) )
|
||||||
|
print ('the size of the cutted RVE is %sX%sX%s'%(sizex, sizey, sizez) )
|
||||||
|
print ('the thickness reduction is %s'%('{:.1%}'.format(options.reduction)))
|
||||||
|
|
||||||
|
# --- loop over input files -------------------------------------------------------------------------
|
||||||
|
filenames = [filename+'.geom']
|
||||||
|
|
||||||
|
for theta in np.linspace(0, np.pi/2, options.number):
|
||||||
|
postfix = str(int(np.round(theta*180.0/np.pi)))
|
||||||
|
c_t, s_t = np.cos(theta), np.sin(theta)
|
||||||
|
offsetX, offsetY = 0.5*( sizeX - (sizex*c_t-sizey*s_t) ), 0.5*( sizeY - (sizex*s_t+sizey*c_t) )
|
||||||
|
for name in filenames:
|
||||||
|
if name == 'STDIN':
|
||||||
|
file = {'name':'STDIN', 'input':sys.stdin, 'output':sys.stdout, 'croak':sys.stderr}
|
||||||
|
file['croak'].write('\033[1m'+scriptName+'\033[0m\n')
|
||||||
|
else:
|
||||||
|
if not os.path.exists(name): continue
|
||||||
|
file = {'name':name, 'input':open(name), 'output':open(name+postfix+'_tmp','w'), 'croak':sys.stderr}
|
||||||
|
file['croak'].write('\033[1m'+scriptName+'\033[0m: '+file['name']+'\n')
|
||||||
|
|
||||||
|
table = damask.ASCIItable(file['input'],file['output'],buffered=False) # make unbuffered ASCII_table
|
||||||
|
table.head_read() # read ASCII header info
|
||||||
|
#--- interpret header ----------------------------------------------------------------------------
|
||||||
|
info = {
|
||||||
|
'grid': np.zeros(3,'i'),
|
||||||
|
'size': np.array((0.0,0.0,0.0)),
|
||||||
|
'origin': np.zeros(3,'d'),
|
||||||
|
'microstructures': 0,
|
||||||
|
'homogenization': 0,
|
||||||
|
}
|
||||||
|
newInfo = {
|
||||||
|
'microstructures': 0,
|
||||||
|
}
|
||||||
|
extra_header = []
|
||||||
|
|
||||||
|
for header in table.info:
|
||||||
|
headitems = map(str.lower,header.split())
|
||||||
|
if len(headitems) == 0: continue
|
||||||
|
if headitems[0] in mappings.keys():
|
||||||
|
if headitems[0] in identifiers.keys():
|
||||||
|
for i in xrange(len(identifiers[headitems[0]])):
|
||||||
|
info[headitems[0]][i] = \
|
||||||
|
mappings[headitems[0]](headitems[headitems.index(identifiers[headitems[0]][i])+1])
|
||||||
|
else:
|
||||||
|
info[headitems[0]] = mappings[headitems[0]](headitems[1])
|
||||||
|
else:
|
||||||
|
extra_header.append(header)
|
||||||
|
newInfo['microstructures'] = info['microstructures']
|
||||||
|
if 0 not in options.grid: # user-specified grid
|
||||||
|
info['grid'] = np.array(options.grid)
|
||||||
|
|
||||||
|
for i in xrange(3):
|
||||||
|
if info['size'][i] <= 0.0: # any invalid size?
|
||||||
|
info['size'][i] = float(info['grid'][i])/max(info['grid'])
|
||||||
|
file['croak'].write('rescaling size %s...\n'%{0:'x',1:'y',2:'z'}[i])
|
||||||
|
if np.any(info['grid'] < 1):
|
||||||
|
file['croak'].write('invalid grid a b c.\n')
|
||||||
|
continue
|
||||||
|
if np.any(info['size'] <= 0.0):
|
||||||
|
file['croak'].write('invalid size x y z.\n')
|
||||||
|
continue
|
||||||
|
|
||||||
|
# read the topological data from maternal RVE file
|
||||||
|
print ( 'cut the RVE along the direction of theta = %i'%(np.round(theta*180/np.pi)) )
|
||||||
|
GrainNo = np.chararray((gridX, gridY, gridz), itemsize=6)
|
||||||
|
for i in xrange(gridY*gridz):
|
||||||
|
content = file['input'].readline().split()
|
||||||
|
for j in xrange(gridX): GrainNo[j, np.mod(i,gridY), i/gridY] = content[j]
|
||||||
|
|
||||||
|
# cut a sub-RVE cooresponding to the specified direction from the maternal RVE
|
||||||
|
subGrainNo = np.chararray((gridx, gridz, gridy), itemsize=6)
|
||||||
|
for i in xrange(gridx):
|
||||||
|
for j in xrange(gridy):
|
||||||
|
I = int(np.floor( (dx*(i+0.5)*c_t - dy*(j+0.5)*s_t + offsetX)/dx))
|
||||||
|
J = int(np.floor( (dx*(i+0.5)*s_t + dy*(j+0.5)*c_t + offsetY)/dy))
|
||||||
|
I = min(I, gridX-1); J = min(J, gridY-1)
|
||||||
|
for k in xrange(gridz): subGrainNo[i,k,j] = GrainNo[I,J,k]
|
||||||
|
subGrainNoVec = subGrainNo.reshape(nGrids)
|
||||||
|
|
||||||
|
# count the number of grains in the sub-RVE
|
||||||
|
index = defaultdict(list)
|
||||||
|
for i in xrange(nGrids): index[subGrainNoVec[i]].append(i)
|
||||||
|
ngrains = len(index)
|
||||||
|
|
||||||
|
# count the broken (scattered) grains due to the cutting, and merge them.
|
||||||
|
if ngrains > options.N*1.1:
|
||||||
|
N1, N2 = 0, nGrids
|
||||||
|
for key in index:
|
||||||
|
if len(index[key])>=0.4*avgGrids:
|
||||||
|
N1+=1; N2-=len(index[key]) # N1: valid grains; N2: number of grids needed to be re-assigned orientation
|
||||||
|
ngrid2 = min(int(0.8*avgGrids), N2/(options.N-N1)+1) # grid in each grain
|
||||||
|
|
||||||
|
newGrains = N2/ngrid2+1
|
||||||
|
a = [ngrid2]*(newGrains)
|
||||||
|
for key in index:
|
||||||
|
if len(index[key])<0.4*avgGrids:
|
||||||
|
for i in xrange(newGrains):
|
||||||
|
if a[i] >= len(index[key]):
|
||||||
|
for j in index[key]: subGrainNoVec[j] = -i-1
|
||||||
|
a[i] -= len(index[key])
|
||||||
|
break
|
||||||
|
|
||||||
|
index = defaultdict(list)
|
||||||
|
for i in xrange(nGrids): index[subGrainNoVec[i]].append(i)
|
||||||
|
ngrains = len(index)
|
||||||
|
|
||||||
|
# assign orientations
|
||||||
|
grainsNo = np.arange(ngrains); np.random.shuffle(grainsNo)
|
||||||
|
for i,key in enumerate(index):
|
||||||
|
no = str(grainsNo[i]+1)
|
||||||
|
for j in index[key]: subGrainNoVec[int(j)] = no
|
||||||
|
|
||||||
|
newInfo['microstructures'] = ngrains
|
||||||
|
if newInfo['microstructures'] == 0:
|
||||||
|
file['croak'].write('no grain info found.\n')
|
||||||
|
continue
|
||||||
|
|
||||||
|
#--- write header ---------------------------------------------------------------------------------
|
||||||
|
table.labels_clear()
|
||||||
|
table.info_clear()
|
||||||
|
table.info_append(extra_header+[
|
||||||
|
scriptID + ' ' + ' '.join(sys.argv[1:]),
|
||||||
|
"grid\ta %i\tb %i\tc %i"%(info['grid'][0],info['grid'][1],info['grid'][2],),
|
||||||
|
"size\tx %f\ty %f\tz %f"%(options.size[0],options.size[1],options.size[2],),
|
||||||
|
"origin\tx %f\ty %f\tz %f"%(info['origin'][0],info['origin'][1],info['origin'][2],),
|
||||||
|
"homogenization\t%i"%info['homogenization'],
|
||||||
|
"microstructures\t%i"%(newInfo['microstructures']),
|
||||||
|
])
|
||||||
|
table.head_write()
|
||||||
|
# --- write microstructure information ------------------------------------------------------------
|
||||||
|
formatwidth = 1+int(math.log10(newInfo['microstructures']))
|
||||||
|
table.data = np.array(map(int, subGrainNoVec)).reshape(gridx, gridy*gridz).T
|
||||||
|
table.data_writeArray('%%%ii'%(formatwidth),delimiter=' ')
|
||||||
|
|
||||||
|
#--- output finalization --------------------------------------------------------------------------
|
||||||
|
if file['name'] != 'STDIN':
|
||||||
|
prefix = os.path.splitext(file['name'])[0].replace('grains'+str(Ngrains),'grains'+str(ngrains))
|
||||||
|
os.rename(name+postfix+'_tmp',
|
||||||
|
prefix+'%s'%('_material.config' if options.config else '_'+postfix+'.geom'))
|
||||||
|
table.close()
|
Loading…
Reference in New Issue