summaryrefslogtreecommitdiff
path: root/openEMS/python/openEMS/nf2ff.py
diff options
context:
space:
mode:
Diffstat (limited to 'openEMS/python/openEMS/nf2ff.py')
-rw-r--r--openEMS/python/openEMS/nf2ff.py210
1 files changed, 210 insertions, 0 deletions
diff --git a/openEMS/python/openEMS/nf2ff.py b/openEMS/python/openEMS/nf2ff.py
new file mode 100644
index 0000000..b6d38d0
--- /dev/null
+++ b/openEMS/python/openEMS/nf2ff.py
@@ -0,0 +1,210 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2015,20016 Thorsten Liebig (Thorsten.Liebig@gmx.de)
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published
+# by the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+import os
+import numpy as np
+import h5py
+from openEMS import _nf2ff
+from openEMS import utilities
+
+class nf2ff:
+ """
+ Create an nf2ff recording box. The nf2ff can either record in time-domain
+ or frequency-domain. Further more certain directions and boundary condition
+ mirroring can be enabled/disabled.
+
+ :param name: str -- Name for this recording box.
+ :param start/stop: (3,) array -- Box start/stop coordinates.
+ :param directions: (6,) bool array -- Enable/Disables directions.
+ :param mirror: (6,) int array -- 0 (Off), 1 (PEC) or 2 (PMC) boundary mirroring
+ :param frequency: array like -- List of frequencies (FD-domain recording)
+ """
+ def __init__(self, CSX, name, start, stop, **kw):
+ self.CSX = CSX
+ self.name = name
+ self.start = start
+ self.stop = stop
+
+ self.freq = None
+ self.theta = None
+ self.phi = None
+ self.center = None
+
+ self.directions = [True]*6 # all directions by default
+ if 'directions' in kw:
+ self.directions = kw['directions']
+ del kw['directions']
+ assert len(self.directions)==6
+
+ self.mirror = [0]*6
+ if 'mirror' in kw:
+ self.mirror = kw['mirror']
+ del kw['mirror']
+ assert len(self.mirror)==6
+
+ self.dump_type = 0 # default Et/Ht
+ self.dump_mode = 1 # default cell interpolated
+
+ self.freq = None # broadband recording by defualt
+ if 'frequency' in kw:
+ self.freq = kw['frequency']
+ del kw['frequency']
+ self.dump_type = 10 # Ef/Hf
+
+ if np.isscalar(self.freq):
+ self.freq = [self.freq]
+
+ self.e_file = '{}_E'.format(self.name)
+ self.h_file = '{}_H'.format(self.name)
+
+ self.e_dump = CSX.AddDump(self.e_file, dump_type=self.dump_type , dump_mode=self.dump_mode, file_type=1, **kw)
+ self.h_dump = CSX.AddDump(self.h_file, dump_type=self.dump_type+1, dump_mode=self.dump_mode, file_type=1, **kw)
+ if self.freq is not None:
+ self.e_dump.SetFrequency(self.freq)
+ self.h_dump.SetFrequency(self.freq)
+
+# print(self.directions)
+ for ny in range(3):
+ pos = 2*ny
+ if self.directions[pos]:
+ l_start = np.array(start)
+ l_stop = np.array(stop)
+ l_stop[ny] = l_start[ny]
+ self.e_dump.AddBox(l_start, l_stop)
+ self.h_dump.AddBox(l_start, l_stop)
+ if self.directions[pos+1]:
+ l_start = np.array(start)
+ l_stop = np.array(stop)
+ l_start[ny] = l_stop[ny]
+ self.e_dump.AddBox(l_start, l_stop)
+ self.h_dump.AddBox(l_start, l_stop)
+
+ def CalcNF2FF(self, sim_path, freq, theta, phi, radius=1, center=[0,0,0], outfile=None, read_cached=False, verbose=0):
+ """ CalcNF2FF(sim_path, freq, theta, phi, center=[0,0,0], outfile=None, read_cached=True, verbose=0):
+
+ Calculate the far-field after the simulation is done.
+
+ :param sim_path: str -- Simulation path
+ :param freq: array like -- list of frequency for transformation
+ :param theta/phi: array like -- Theta/Phi angles to calculate the far-field
+ :param radius: float -- Radius to calculate the far-field (default is 1m)
+ :param center: (3,) array -- phase center, must be inside the recording box
+ :param outfile: str -- File to save results in. (defaults to recording name)
+ :param read_cached: bool -- enable/disable read already existing results (default off)
+ :param verbose: int -- set verbose level (default 0)
+
+ :returns: nf2ff_results class instance
+ """
+ if np.isscalar(freq):
+ freq = [freq]
+ self.freq = freq
+ if np.isscalar(theta):
+ theta = [theta]
+ self.theta = theta
+ if np.isscalar(phi):
+ phi = [phi]
+ self.phi = phi
+ self.center = center
+
+ if outfile is None:
+ fn = os.path.join(sim_path, self.name + '.h5')
+ else:
+ fn = os.path.join(sim_path, outfile)
+ if not read_cached or not os.path.exists(fn):
+ nfc = _nf2ff._nf2ff(self.freq, np.deg2rad(theta), np.deg2rad(phi), center, verbose=verbose)
+
+ for ny in range(3):
+ nfc.SetMirror(self.mirror[2*ny] , ny, self.start[ny])
+ nfc.SetMirror(self.mirror[2*ny+1], ny, self.stop[ny])
+
+ nfc.SetRadius(radius)
+
+ for n in range(6):
+ fn_e = os.path.join(sim_path, self.e_file + '_{}.h5'.format(n))
+ fn_h = os.path.join(sim_path, self.h_file + '_{}.h5'.format(n))
+ if os.path.exists(fn_e) and os.path.exists(fn_h):
+ assert nfc.AnalyseFile(fn_e, fn_h)
+
+ nfc.Write2HDF5(fn)
+
+ result = nf2ff_results(fn)
+ if result.phi is not None:
+ assert np.abs((result.r-radius)/radius)<1e-6, 'Radius does not match. Did you read an invalid chached result? Try "read_cached=False"'
+ assert utilities.Check_Array_Equal(np.rad2deg(result.theta), self.theta, 1e-4), 'Theta array does not match. Did you read an invalid chached result? Try "read_cached=False"'
+ assert utilities.Check_Array_Equal(np.rad2deg(result.phi), self.phi, 1e-4), 'Phi array does not match. Did you read an invalid chached result? Try "read_cached=False"'
+ assert utilities.Check_Array_Equal(result.freq, self.freq, 1e-6, relative=True), 'Frequency array does not match. Did you read an invalid chached result? Try "read_cached=False"'
+ return result
+
+class nf2ff_results:
+ """
+ nf2ff result class containing all results obtained by the nf2ff calculation.
+ Usueally returned from nf2ff.CalcNF2FF
+
+ Available attributes:
+
+ * `fn` : file name
+ * `theta`: theta angles
+ * `phi` : phi angles
+ * `r` : radius
+ * `freq` : frequencies
+ * `Dmax` : directivity over frequency
+ * `Prad` : total radiated power over frequency
+
+ * `E_theta` : theta component of electric field over frequency/theta/phi
+ * `E_phi` : phi component of electric field over frequency/theta/phi
+ * `E_norm` : abs component of electric field over frequency/theta/phi
+ * `E_cprh` : theta component of electric field over frequency/theta/phi
+ * `E_cplh` : theta component of electric field over frequency/theta/phi
+ * `P_rad` : radiated power (S) over frequency/theta/phi
+ """
+ def __init__(self, fn):
+ self.fn = fn
+ h5_file = h5py.File(fn, 'r')
+ mesh_grp = h5_file['Mesh']
+ self.phi = np.array(mesh_grp['phi'])
+ self.theta = np.array(mesh_grp['theta'])
+ self.r = np.array(mesh_grp['r'])
+
+ data = h5_file['nf2ff']
+ self.freq = np.array(data.attrs['Frequency'])
+
+ self.Dmax = np.array(data.attrs['Dmax'])
+ self.Prad = np.array(data.attrs['Prad'])
+
+ THETA, PHI = np.meshgrid(self.theta, self.phi, indexing='ij')
+ cos_phi = np.cos(PHI)
+ sin_phi = np.sin(PHI)
+
+ self.E_theta = []
+ self.E_phi = []
+ self.P_rad = []
+ self.E_norm = []
+ self.E_cprh = []
+ self.E_cplh = []
+ for n in range(len(self.freq)):
+ E_theta = np.array(h5_file['/nf2ff/E_theta/FD/f{}_real'.format(n)]) + 1j*np.array(h5_file['/nf2ff/E_theta/FD/f{}_imag'.format(n)])
+ E_theta = np.swapaxes(E_theta, 0, 1)
+ E_phi = np.array(h5_file['/nf2ff/E_phi/FD/f{}_real'.format(n)]) + 1j*np.array(h5_file['/nf2ff/E_phi/FD/f{}_imag'.format(n)])
+ E_phi = np.swapaxes(E_phi, 0, 1)
+ self.P_rad .append(np.swapaxes(np.array(h5_file['/nf2ff/P_rad/FD/f{}'.format(n)]), 0, 1))
+
+ self.E_theta.append(E_theta)
+ self.E_phi .append(E_phi)
+ self.E_norm .append(np.sqrt(np.abs(E_theta)**2 + np.abs(E_phi)**2))
+ self.E_cprh .append((cos_phi+1j*sin_phi) * (E_theta+1j*E_phi)/np.sqrt(2.0))
+ self.E_cplh .append((cos_phi-1j*sin_phi) * (E_theta-1j*E_phi)/np.sqrt(2.0))