File: Case1_SiloFlow_customFilling.py

package info (click to toggle)
yade 2025.2.0-2
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 33,308 kB
  • sloc: cpp: 93,298; python: 50,409; sh: 577; makefile: 162
file content (260 lines) | stat: -rw-r--r-- 9,228 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
# -*- encoding=utf-8 -*-
# 2020 © Vasileios Angelidakis <v.angelidakis2@ncl.ac.uk>
# 2020 © Bruno Chareyre <bruno.chareyre@grenoble-inp.fr>

# Benchmark of basic performance of open-source DEM simulation systems,
# Case 1: Silo flow
# this version uses custom positions for spheres, and custom contact properties, change

particleRadius = 8  # 2mm in the benchmark, bigger for less particles and quicker simulations
numThreads = 4
reportTiming = False

# Configure MPI module if needed
mpi = 'MPI' in yade.config.features
if mpi:
	from yade import mpy as mp
else:
	print("yade is compiled without MPI support, numThreads>1 ignored")
	numThreads = 4

# The stl file is in mm -> We can keep the units below or scale the stl file to meters (ymport.stl does not have a scale parameter, but I can edit the .stl file manually)

# -------------------------------------------------------------------- #
# Units: N/mm/MPa/ton
# 	force:   1 N
# 	length:  1 mm = 1e-3 m
# 	stress:  1 MPa = 1 N/mm^2 = 1e6 Pa = 1e-3 GPa
#	mass:    1 ton = 1e3 kg
#	density: 1 ton/mm^3 = 1e-12 kg/m^3

# -------------------------------------------------------------------- #
# Input Data -> Define Material and Orifice size. Uncomment the prefered choice

#FIXME: most of what follows don't have to be guarded by rank==0
fileName = 'SiloLargeOrifice'
#fileName='SiloSmallOrifice'
if fileName == 'SiloLargeOrifice':
	z = 70  # This is the height of the lowest point of the funnel (at the orifice), measuring from the lowest cylindrical cross section of the silo
elif fileName == 'SiloSmallOrifice':
	z = 80

granularMaterial = 'M1'
#granularMaterial='M2'

# -------------------------------------------------------------------- #
# Materials
Steel = O.materials.append(FrictMat(young=210e-3, poisson=0.2, density=7200e-12, label='Steel'))

# -------------------------------------------------------------------- #
# Asign values based on the Material and Orifice size
# Coeff of restitution (e) / Coeff of friction (f)
e_M1_M2 = 0.45
f_M1_M2 = 0.2
e_M1_M1 = 0.5
f_M1_M1 = 0.3
e_M1_St = 0.4
f_M1_St = 0.2
e_M2_M2 = 0.4
f_M2_M2 = 0.4
e_M2_St = 0.4
f_M2_St = 0.2

if granularMaterial == 'M1':
	M1 = O.materials.append(FrictMat(young=1.0e-3, poisson=0.2, density=2500e-12, label='M1'))
	e_gg = e_M1_M1  # Coefficient of restitution (e) between granular material (g) and granular material (g)
	f_gg = f_M1_M1  # Coefficient of friction (f)...
	e_gs = e_M1_St  # Coefficient of restitution (e) between granular material (g) and steel (s)
	f_gs = f_M1_St  # Coefficient of friction (f)...
elif granularMaterial == 'M2':
	M2 = O.materials.append(FrictMat(young=0.5e-3, poisson=0.2, density=2000e-12, label='M2'))
	e_gg = e_M2_M2
	f_gg = f_M2_M2
	e_gs = e_M2_St
	f_gs = f_M2_St

F_gg = atan(f_gg)  # Friction Angle between granular material (g) and granular material (g)
F_gs = atan(f_gs)  # Friction Angle between granular material (g) and steel (s)

# -------------------------------------------------------------------- #
## Engines
O.engines = [
        ForceResetter(),
        InsertionSortCollider([Bo1_Sphere_Aabb(), Bo1_Facet_Aabb()], label="collider"),
        InteractionLoop(
                [Ig2_Sphere_Sphere_ScGeom(), Ig2_Facet_Sphere_ScGeom()],
                [
                        Ip2_FrictMat_FrictMat_MindlinPhys(
                                frictAngle=MatchMaker(matches=((1, 1, F_gg), (0, 1, F_gs))),  # 0 being the id of Steel and
                                en=MatchMaker(matches=((1, 1, e_gg), (0, 1, e_gs)))  # 1 being the id of granularMaterial
                        )
                ],
                [Law2_ScGeom_MindlinPhys_Mindlin()],
        ),
        NewtonIntegrator(damping=0, gravity=[0, 0, -9.810], label="newton"),
        #GlobalStiffnessTimeStepper(active=1,timestepSafetyCoefficient=0.8, timeStepUpdateInterval=100, parallelMode=False, label = "ts",defaultDt=PWaveTimeStep()), #FIXME Remember to reinstate parallelMode=True when we use MPI
        #VTKRecorder(virtPeriod=0.04,fileName='/tmp/Silo-',recorders=['spheres','facets']),
]

# This condition is not abolutely necessary but it would be inelegant to
# download *.stl and generate densePack N times when we need it done only on master (centralized scene method)
if not mpi or mp.rank == 0:

	# -------------------------------------------------------------------- #
	# Generate initial packing. Choose among regularOrtho, regularHexa or randomDensePack (which I think is best).

	## Using regularOrtho
	#sp=pack.regularOrtho(pack.inCylinder(Vector3(0,0,0),Vector3(0,0,305),radius),radius=2,gap=r*1/10.,material=granularMaterial)

	## Using regularHexa
	#sp=pack.regularHexa(pack.inCylinder(Vector3(0,0,0),Vector3(0,0,215),radius),radius=2,gap=r*1/10.,material=granularMaterial)

	# Using randomDensePack
	sp = pack.randomDensePack(
	        pack.inCylinder((0, 0, 0), (0, 0, 254), 100),
	        radius=particleRadius,
	        spheresInCell=500,
	        returnSpherePack=False,
	        material=granularMaterial,
	        seed=1
	)

	# -------------------------------------------------------------------- #
	# Sort packing in ascending Z coordinates and delete excess particles to achieve sample size of 122k
	zValues = []
	for s in sp:
		zValues.append(s.state.pos[2])

	from operator import itemgetter
	indices, zValues_sorted = zip(*sorted(enumerate(zValues), key=itemgetter(1)))
	list(zValues)
	list(indices)

	sp_new = []
	for i in range(0, len(sp)):
		sp_new.append(sp[indices[i]])

	Nspheres = 122000
	sp_new = sp_new[0:Nspheres]

	from yade import ymport
	if not os.path.exists(fileName + '.stl'):
		print("Downloading mesh file")
		try:
			os.system('wget http://yade-dem.org/publi/data/DEM8/' + fileName + '.stl')
		except:
			print("** probably no internet connection, grab the *.stl files by yourself **")
	facets = ymport.stl(fileName + '.stl', color=(0, 1, 0), material=Steel)
	fctIds = range(len(facets))

	O.bodies.append(facets)
	O.bodies.append(sp_new)

	# -------------------------------------------------------------------- #
	# Count the number of spherical particles to verify sample size. We can comment this out later on.
	numSpheres = 0
	for b in O.bodies:
		if isinstance(b.shape, Sphere):
			numSpheres = numSpheres + 1
	print('The total number of spheres is: ', numSpheres)

	collider.verletDist = 0.6 * particleRadius
	O.dt = 0.6 * PWaveTimeStep()
	O.dynDt = False

# -------------------------------------------------------------------- #
# Erase particles flowing out of the silo


def eraseEscapedParticles():
	global numErased
	count = 0
	ts = time.time()
	ers = []
	for b in O.bodies:
		if isinstance(b.shape, Sphere) and b.state.pos[
		        2] < -z - 20:  # I do not delete the particles right after they pass the orifice, to disturb the simulation as little as possible
			ers.append(b.id)
			count += 1
	if mpi:
		mp.bodyErase(ers)
	else:
		for b in ers:
			O.bodies.erase(b.id)
	numErased += count


#-------------------------------------------------------------------- #
#Record time-dependent number of retained particles and vtk export

from yade import plot

plot.plots = {'time': (('retained', 'bo--'), None, ('Cu', "kx--"))}

numErased = 0


def addPlotData(Cu):
	plot.addData(retained=numSpheres - numErased, time=O.time, Cu=Cu)


from yade import export

vtk = export.VTKExporter("spheresFinal")

# -------------------------------------------------------------------- #
# Run iterations

if mpi:  # import and tune MPI module
	mp.DOMAIN_DECOMPOSITION = True
	mp.ACCUMULATE_FORCES = False
	mp.MERGE_W_INTERACTIONS = False
	mp.REALLOCATE_MINIMAL = False  # if true, intersections are minimized before reallocations, hence minimizing the number of reallocated bodies
	mp.REALLOCATE_FREQUENCY = 20
	mp.USE_CPP_REALLOC = True
	mp.MINIMAL_INTERSECTIONS = True
	mp.MASTER_UPDATE_STATES = False
	mp.ERASE_REMOTE_MASTER = True
	mp.YADE_TIMING = reportTiming
else:
	O.timingEnabled = reportTiming

substeps = 500
#while len(O.bodies)-numErased>0 or len(O.bodies)==0:
for k in range(100):
	t1 = time.time()
	if mpi:
		mp.mpirun(substeps, numThreads, withMerge=True)  # if numThreads=1 this will fall-back to normal O.run() and mp.rank=0
	else:
		O.run(substeps, True)
	tbf = time.time()
	eraseEscapedParticles()
	if mpi and mp.rank > 0:
		continue  # mpi workers do not record
	t2 = time.time()
	addPlotData((numSpheres - numErased) * substeps / (t2 - t1))
	vtk.exportSpheres(what=dict(particleVelocity='b.state.vel', domain='b.subdomain'))
	plot.plot(noShow=True).savefig(fileName + '_' + granularMaterial + '_np' + str(numThreads) + '.png')
	plot.saveDataTxt(fileName + '_' + granularMaterial + '.txt')
	print("iter=", O.iter, ", last substep erased", numErased, "in", t2 - t1, "s")

# -------------------------------------------------------------------- #
# GUI
if opts.nogui == False:
	from yade import qt
	v = qt.View()

	v.eyePosition = Vector3(0, -600, 100)
	v.upVector = Vector3(0, 0, 1)
	v.viewDir = Vector3(0, 1, 0)
	#	v.grid=(False,True,False)

	rndr = yade.qt.Renderer()
	#rndr.shape=False
	#rndr.bound=True

## To play interactively with mpi execution:
## mp.mpirun(100,numThreads,True) #'True' so we see merged scene after the run
## eraseEscapedParticles()
## mp.mpirun(100,numThreads,True)
## etc.