jank324
commented
1 year ago Example from @ansantam
#!/usr/bin/env python
# ----------------------------------------------------------------------------------------------------------------------------
# Function to generate a particle distribution as input for the collimation routine in SixTrack (round beams at symmetry point)
# Example for SPS:
# generate_distribution.py 64 26e9 'SPS_inj' True 1 1 0.9e-6 3.0e-6 1.50895801 -1.392739114 51.8375213 46.54197726 0 0 0 0 0 0 0.3 10.7e-4
# Example for HL-LHC:
# generate_distribution.py 64 7e12 'HL_coll' True 1 1 2.5e-6 2.5e-6 0.003485 -0.000764 0.150739 0.150235 -7.5e-4 0 0 0.3e-3 0.003652 0.000517 0.0755 1.13e-4
# ----------------------------------------------------------------------------------------------------------------------------
import datetime
import sys
import random
import numpy as np
from util import get_bucket
from util import get_rel_params
from util import get_sigmas
# Pass the Command Line Arguments
# --------------------------------------------------------------------------------------------------------------
particles = int(sys.argv[1])
energy = float(sys.argv[2])
machine = str(sys.argv[3])
fort13 = sys.argv[4]
jobs = int(sys.argv[5])
factor = float(sys.argv[6])
emittance_x = float(sys.argv[7])
emittance_y = float(sys.argv[8])
alpha_x = float(sys.argv[9])
alpha_y = float(sys.argv[10])
beta_x = float(sys.argv[11])
beta_y = float(sys.argv[12])
offset_x = float(sys.argv[13])
offset_xp = float(sys.argv[14])
offset_y = float(sys.argv[15])
offset_yp = float(sys.argv[16])
dispersion_x = float(sys.argv[17])
dispersion_y = float(sys.argv[18])
bunch = float(sys.argv[19])
spread = float(sys.argv[20])
seed = int(sys.argv[21])
def dist_generator(particles, energy, machine, fort13, jobs, factor, emittance_x, emittance_y, alpha_x, alpha_y, beta_x, beta_y, offset_x, offset_xp, offset_y, offset_yp, dispersion_x, dispersion_y, bunch, spread, seed):
job_str = '%s'%jobs
# Getting the Transverse sigmas (amplitudes of phase space ellipse)
# --------------------------------------------------------------------------------------------------------------
gamma_rel, beta_rel, p0, mass = get_rel_params(energy)
tx_max, txp_max = get_sigmas(alpha_x, beta_x, emittance_x, dispersion_x, spread, beta_rel, gamma_rel)
ty_max, typ_max = get_sigmas(alpha_y, beta_y, emittance_y, dispersion_y, spread, beta_rel, gamma_rel)
# Seeding
# --------------------------------------------------------------------------------------------------------------
if seed == 0:
myseed = random.randint(0, 429496729)
else:
myseed = seed
with open('seed.txt', 'a') as g:
print >> g, 'job ', job_str ,'seed ', myseed
np.random.seed(myseed)
random.seed(myseed)
# Generating the Transverse Distribution
# --------------------------------------------------------------------------------------------------------------
x_t = np.asarray(np.random.normal(0, factor * tx_max, round(particles)))
xp_t = np.asarray(np.random.normal(0, factor * txp_max, round(particles)))
y_t = np.asarray(np.random.normal(0, factor * ty_max, round(particles)))
yp_t = np.asarray(np.random.normal(0, factor * typ_max, round(particles)))
# Rotating the Transverse Distribution
# --------------------------------------------------------------------------------------------------------------
angle_x = np.arctan(-alpha_x/beta_x)
x = x_t*np.cos(angle_x) - xp_t*np.sin(angle_x)
xp = x_t*np.sin(angle_x) + xp_t*np.cos(angle_x)
angle_y = np.arctan(-alpha_y/beta_y)
y = y_t*np.cos(angle_y) - yp_t*np.sin(angle_y)
yp = y_t*np.sin(angle_y) + yp_t*np.cos(angle_y)
# Generating the Longitudinal Distribution
# --------------------------------------------------------------------------------------------------------------
z = []
E = []
dp = []
while len(z) < particles:
# Generate for as long time as is needed
particle_z = random.gauss(0,1)
particle_e = random.gauss(0,1)
trial_z = particle_z * bunch
trial_e = energy * (1 + particle_e*spread) #eV
trial_p = np.sqrt((trial_e - mass) * (trial_e + mass))
dPP = (trial_p - p0) / p0
h = get_bucket(machine, plot=False, z=trial_z, DELTA=dPP) # Longitudinal contour
if machine=='HL_coll' or machine=='HL_coll_200' or machine=='HL_coll_tcp' or machine=='HL_coll_tcp_200':
Hmargin = -0.01
elif machine=='SPS_inj':
Hmargin = -1
if h <= Hmargin:
z.append(float(trial_z))
E.append(float(trial_e))
dp.append(dPP)
else:
print 'Outside margin, trying again,', h
zz = np.asarray(z)
EE = np.asarray(E)
ddp = np.asarray(dp)
if fort13=='False':
outfile = 'init_dist_' + job_str + '.txt'
with open(outfile, 'w') as f:
for e1, e2, e3, e4, e5, e6 in zip(x, xp, y, yp, zz*1e3, EE*1e-6):
f.write('%8.6e %8.6e %8.6e %8.6e %8.6e %8.6e\n' % (e1, e2, e3, e4, e5, e6))
elif fort13=='True':
outfile = 'fort.13'
with open(outfile, 'w') as f:
for i in xrange(0, particles, 2):
f.write(str((x[i] + offset_x)*1e3) + "\n") #mm
f.write(str((xp[i] + offset_xp)*1e3) + "\n") #mrad
f.write(str((y[i] + offset_y)*1e3) + "\n") #mm
f.write(str((yp[i] + offset_yp)*1e3) + "\n") #mrad
f.write(str(zz[i]*1e3) + "\n") #mm
f.write(str(ddp[i]) + "\n") #-
f.write(str((x[i+1] + offset_x)*1e3) + "\n") #mm
f.write(str((xp[i+1] + offset_xp)*1e3) + "\n") #mrad
f.write(str((y[i+1] + offset_y)*1e3) + "\n") #mm
f.write(str((yp[i+1] + offset_yp)*1e3) + "\n") #mrad
f.write(str(zz[i+1]*1e3) + "\n") #mm
f.write(str(ddp[i+1]) + "\n") #-
f.write(str(energy*1e-6) + "\n") #MeV
f.write(str(EE[i]*1e-6) + "\n") #MeV
f.write(str(EE[i+1]*1e-6) + "\n") #MeV
else:
print 'Please input True or False in the fourth argument'
# Call the function
# --------------------------------------------------------------------------------------------------------------
job_range = range(1, int(jobs) + 1)
with open('seed.txt', 'a') as g:
print >> g, datetime.datetime.now()
for n in job_range:
j = '%s'%n
dist_generator(particles, energy, machine, fort13, n, factor, emittance_x, emittance_y, alpha_x, alpha_y, beta_x, beta_y, offset_x, offset_xp, offset_y, offset_yp, dispersion_x, dispersion_y, bunch, spread, seed)
🐛 Bug Report
According to @ansantam there is an issue in the particle distribution generation. This should be investigated, understood and fixed.
Motivation
Self-motivating.