Homec4science
Diffusion RationalROMPy (feature_greedy_coarsening) Browse

RROMPy/examples/scipy/PadeLagrange.py
cc0997c6ca03feature_greedy_coarseni…

Last Change
Raw File

PadeLagrange.py
View Options

import numpy as np
from rrompy.hfengines.scipy import HelmholtzSquareBubbleProblemEngine as HSBPE
from rrompy.hfengines.scipy import HelmholtzSquareTransmissionProblemEngine as HSTPE
from rrompy.hfengines.scipy import HelmholtzBoxScatteringProblemEngine as HBSPE
from rrompy.reduction_methods.lagrange import ApproximantLagrangePade as Pade
from rrompy.utilities.parameter_sampling import QuadratureSampler as QS
testNo = 2
verb = 0
homog = True
homog = False
if testNo == 1:
k0s = np.power([10 + 0.j, 14 + 0.j], .5)
k0 = np.mean(k0s)
ktar = (11 + .5j) ** .5
rescaling = lambda x: np.power(x, 2.)
rescalingInv = lambda x: np.power(x, .5)
params = {'N':4, 'M':3, 'S':5, 'POD':True,
'sampler':QS(k0s, "CHEBYSHEV", rescaling, rescalingInv)}
solver = HSBPE(kappa = 12 ** .5, theta = np.pi / 3, n = 40,
verbosity = verb)
solver.omega = np.real(k0)
approx = Pade(solver, mu0 = k0, approxParameters = params,
verbosity = verb)
approx.setupApprox()
# approx.plotSamples()
approx.plotApp(ktar, name = 'u_Pade''')
approx.plotHF(ktar, name = 'u_HF')
approx.plotErr(ktar, name = 'err')
approx.plotRes(ktar, name = 'res')
appErr, solNorm = approx.normErr(ktar), approx.normHF(ktar)
resNorm, RHSNorm = approx.normRes(ktar), approx.normRHS(ktar)
print(('SolNorm:\t{}\nErr:\t{}\nErrRel:\t{}').format(solNorm, appErr,
np.divide(appErr, solNorm)))
print(('RHSNorm:\t{}\nRes:\t{}\nResRel:\t{}').format(RHSNorm, resNorm,
np.divide(resNorm, RHSNorm)))
print('\nPoles Pade'':')
print(approx.getPoles())
############
elif testNo == 2:
k0s = [3.85 + 0.j, 4.15 + 0.j]
k0 = np.mean(k0s)
ktar = 4 + .15j
rescaling = lambda x: np.power(x, 2.)
rescalingInv = lambda x: np.power(x, .5)
params = {'N':8, 'M':9, 'S':10, 'POD':True,
'sampler':QS(k0s, "CHEBYSHEV", rescaling, rescalingInv)}
solver = HSTPE(nT = 2, nB = 1, theta = np.pi * 45/180, kappa = 4., n = 50,
verbosity = verb)
solver.omega = np.real(k0)
approx = Pade(solver, mu0 = k0, approxParameters = params,
verbosity = verb, homogeneize = homog)
approx.setupApprox()
approx.plotSamples()
approx.plotApp(ktar, name = 'u_Pade''')
approx.plotHF(ktar, name = 'u_HF')
approx.plotErr(ktar, name = 'err')
approx.plotRes(ktar, name = 'res')
appErr, solNorm = approx.normErr(ktar), approx.normHF(ktar)
resNorm, RHSNorm = approx.normRes(ktar), approx.normRHS(ktar)
print(('SolNorm:\t{}\nErr:\t{}\nErrRel:\t{}').format(solNorm, appErr,
np.divide(appErr, solNorm)))
print(('RHSNorm:\t{}\nRes:\t{}\nResRel:\t{}').format(RHSNorm, resNorm,
np.divide(resNorm, RHSNorm)))
print('\nPoles Pade'':')
print(approx.getPoles())
############
elif testNo == 3:
k0s = [2, 5]
k0 = np.mean(k0s)
ktar = 4.5 - 0.j
params = {'N':10, 'M':11, 'S':15, 'POD':True,
'sampler':QS(k0s, "CHEBYSHEV")}
solver = HBSPE(R = 7, kappa = 3, theta = - np.pi * 75 / 180, n = 40,
verbosity = verb)
solver.omega = np.real(k0)
approx = Pade(solver, mu0 = k0, approxParameters = params,
verbosity = verb, homogeneize = homog)
approx.setupApprox()
approx.plotSamples()
approx.plotApp(ktar, name = 'u_Pade''')
approx.plotHF(ktar, name = 'u_HF')
approx.plotErr(ktar, name = 'err')
approx.plotRes(ktar, name = 'res')
appErr, solNorm = approx.normErr(ktar), approx.normHF(ktar)
resNorm, RHSNorm = approx.normRes(ktar), approx.normRHS(ktar)
print(('SolNorm:\t{}\nErr:\t{}\nErrRel:\t{}').format(solNorm, appErr,
np.divide(appErr, solNorm)))
print(('RHSNorm:\t{}\nRes:\t{}\nResRel:\t{}').format(RHSNorm, resNorm,
np.divide(resNorm, RHSNorm)))
print('\nPoles Pade'':')
print(approx.getPoles())
c4science · Help