Note
Go to the end to download the full example code.
Maps: ComboMaps#
Invert synthetic magnetic data with variable background values and a single block anomaly buried at depth. We will use the Sum Map to invert for both the background values and an heterogeneous susceptibiilty model.
1
/home/vsts/work/1/s/simpeg/utils/model_builder.py:37: BreakingChangeWarning:
Since SimPEG v0.25.0, the 'get_indices_block' function returns a single array with the cell indices, instead of a tuple with a single element. This means that we don't need to unpack the tuple anymore to access to the cell indices.
If you were using this function as in:
ind = get_indices_block(p0, p1, mesh.cell_centers)[0]
Make sure you update it to:
ind = get_indices_block(p0, p1, mesh.cell_centers)
To hide this warning, add this to your script or notebook:
import warnings
from simpeg.utils import BreakingChangeWarning
warnings.filterwarnings(action='ignore', category=BreakingChangeWarning)
Running inversion with SimPEG v0.25.1.dev1+g9a8c46e88
================================================= Projected GNCG =================================================
# beta phi_d phi_m f |proj(x-g)-x| LS iter_CG CG |Ax-b|/|b| CG |Ax-b| Comment
-----------------------------------------------------------------------------------------------------------------
0 5.50e+05 9.12e+06 4.27e-04 9.12e+06 0 inf inf
1 5.50e+05 1.31e+05 7.67e-02 1.73e+05 6.29e+01 0 3 7.80e-04 1.07e+06
2 2.75e+05 4.38e+04 2.38e-01 1.09e+05 3.53e+01 0 9 6.04e-04 2.67e+04
3 1.37e+05 1.87e+04 3.66e-01 6.90e+04 4.91e+01 0 10 1.93e-02 2.77e+04
4 6.87e+04 5.97e+03 4.77e-01 3.87e+04 5.53e+01 0 9 4.71e-04 4.97e+03
5 3.44e+04 4.39e+03 5.10e-01 2.19e+04 4.73e+01 1 10 9.60e-03 4.81e+03
6 1.72e+04 8.94e+02 5.98e-01 1.12e+04 3.57e+01 0 10 4.91e-04 5.25e+03
7 8.59e+03 8.86e+02 5.99e-01 6.03e+03 4.85e+01 4 10 1.11e-01 1.61e+04
8 4.29e+03 4.40e+02 6.42e-01 3.20e+03 5.30e+01 0 10 5.63e-02 6.31e+04
9 2.15e+03 4.40e+02 6.42e-01 1.82e+03 3.35e+01 7 10 8.56e-02 2.32e+04
10 1.07e+03 3.92e+02 6.59e-01 1.10e+03 3.51e+01 0 10 1.65e-01 5.39e+04
Reached starting chifact with l2-norm regularization: Start IRLS steps...
irls_threshold 0.010233860704126293
irls_threshold 0.012188763900495128
11 1.07e+03 4.02e+02 8.86e-01 1.35e+03 5.40e+01 1 10 1.09e+00 1.63e+05
12 1.07e+03 3.98e+02 9.46e-01 1.41e+03 6.22e+01 3 10 5.31e-02 7.13e+04
13 1.07e+03 3.98e+02 9.92e-01 1.46e+03 3.53e+01 11 10 5.58e-01 3.55e+05
14 1.07e+03 4.02e+02 1.01e+00 1.49e+03 3.54e+01 4 10 2.06e-01 1.32e+05
15 1.07e+03 4.09e+02 9.91e-01 1.47e+03 5.79e+01 0 10 1.30e-01 7.11e+04
16 1.07e+03 4.10e+02 9.98e-01 1.48e+03 5.24e+01 5 10 1.76e+00 2.44e+05
17 1.07e+03 4.16e+02 9.55e-01 1.44e+03 3.27e+01 0 10 3.44e-01 1.19e+05
18 1.07e+03 4.22e+02 9.39e-01 1.43e+03 3.61e+01 2 10 6.51e-02 6.08e+04
19 1.07e+03 4.25e+02 8.94e-01 1.38e+03 6.06e+01 4 10 2.76e-02 3.31e+04
20 1.07e+03 4.37e+02 8.22e-01 1.32e+03 6.10e+01 2 10 3.57e-02 4.85e+04
21 1.07e+03 4.33e+02 7.39e-01 1.23e+03 3.56e+01 0 10 7.56e-03 9.26e+03
22 1.07e+03 4.39e+02 6.96e-01 1.19e+03 3.44e+01 2 10 2.54e-02 1.42e+04
23 1.07e+03 4.37e+02 6.15e-01 1.10e+03 6.08e+01 0 10 1.07e-02 1.11e+04
24 1.07e+03 4.45e+02 5.61e-01 1.05e+03 3.15e+01 2 10 1.83e-01 2.07e+04
25 8.85e+02 4.27e+02 4.95e-01 8.66e+02 3.62e+01 0 10 1.05e-02 1.07e+04
26 8.85e+02 4.46e+02 4.07e-01 8.06e+02 5.96e+01 0 10 5.14e-01 2.10e+05
27 7.29e+02 4.35e+02 3.60e-01 6.97e+02 6.23e+01 0 10 4.45e-03 8.67e+03
28 7.29e+02 4.19e+02 2.95e-01 6.34e+02 3.78e+01 0 10 2.65e-03 5.23e+03
29 7.29e+02 4.19e+02 2.50e-01 6.02e+02 3.57e+01 1 10 1.10e-02 6.25e+03
30 7.29e+02 4.13e+02 2.19e-01 5.73e+02 6.08e+01 0 10 7.30e-03 4.47e+03
Reach maximum number of IRLS cycles: 20
------------------------- STOP! -------------------------
1 : |fc-fOld| = 9.0362e+00 <= tolF*(1+|f0|) = 9.1155e+05
1 : |xc-x_last| = 3.1119e-03 <= tolX*(1+|x0|) = 1.0075e-01
0 : |proj(x-g)-x| = 6.0797e+01 <= tolG = 1.0000e-03
0 : |proj(x-g)-x| = 6.0797e+01 <= 1e3*eps = 1.0000e-03
0 : maxIter = 100 <= iter = 30
------------------------- DONE! -------------------------
from discretize import TensorMesh
from discretize.utils import active_from_xyz
from simpeg import (
utils,
maps,
regularization,
data_misfit,
optimization,
inverse_problem,
directives,
inversion,
)
from simpeg.potential_fields import magnetics
import numpy as np
import matplotlib.pyplot as plt
def run(plotIt=True):
h0_amplitude, h0_inclination, h0_declination = (50000.0, 90.0, 0.0)
# Create a mesh
dx = 5.0
hxind = [(dx, 5, -1.3), (dx, 10), (dx, 5, 1.3)]
hyind = [(dx, 5, -1.3), (dx, 10), (dx, 5, 1.3)]
hzind = [(dx, 5, -1.3), (dx, 10)]
mesh = TensorMesh([hxind, hyind, hzind], "CCC")
# Lets create a simple Gaussian topo and set the active cells
[xx, yy] = np.meshgrid(mesh.nodes_x, mesh.nodes_y)
zz = -np.exp((xx**2 + yy**2) / 75**2) + mesh.nodes_z[-1]
# We would usually load a topofile
topo = np.c_[utils.mkvc(xx), utils.mkvc(yy), utils.mkvc(zz)]
# Go from topo to array of indices of active cells
actv = active_from_xyz(mesh, topo, "N")
nC = int(actv.sum())
# Create and array of observation points
xr = np.linspace(-20.0, 20.0, 20)
yr = np.linspace(-20.0, 20.0, 20)
X, Y = np.meshgrid(xr, yr)
# Move the observation points 5m above the topo
Z = -np.exp((X**2 + Y**2) / 75**2) + mesh.nodes_z[-1] + 5.0
# Create a MAGsurvey
rxLoc = np.c_[utils.mkvc(X.T), utils.mkvc(Y.T), utils.mkvc(Z.T)]
rxLoc = magnetics.Point(rxLoc)
srcField = magnetics.UniformBackgroundField(
receiver_list=[rxLoc],
amplitude=h0_amplitude,
inclination=h0_inclination,
declination=h0_declination,
)
survey = magnetics.Survey(srcField)
# We can now create a susceptibility model and generate data
model = np.zeros(mesh.nC)
# Change values in half the domain
model[mesh.gridCC[:, 0] < 0] = 0.01
# Add a block in half-space
model = utils.model_builder.add_block(
mesh.gridCC, model, np.r_[-10, -10, 20], np.r_[10, 10, 40], 0.05
)
model = utils.mkvc(model)
model = model[actv]
# Create active map to go from reduce set to full
actvMap = maps.InjectActiveCells(mesh, actv, np.nan)
# Create reduced identity map
idenMap = maps.IdentityMap(nP=nC)
# Create the forward model operator
prob = magnetics.Simulation3DIntegral(
mesh,
survey=survey,
chiMap=idenMap,
active_cells=actv,
store_sensitivities="forward_only",
)
# Compute linear forward operator and compute some data
data = prob.make_synthetic_data(
model, relative_error=0.0, noise_floor=1, add_noise=True
)
# Create a homogenous maps for the two domains
domains = [mesh.gridCC[actv, 0] < 0, mesh.gridCC[actv, 0] >= 0]
homogMap = maps.SurjectUnits(domains)
# Create a wire map for a second model space, voxel based
wires = maps.Wires(("homo", len(domains)), ("hetero", nC))
# Create Sum map
sumMap = maps.SumMap([homogMap * wires.homo, wires.hetero])
# Create the forward model operator
prob = magnetics.Simulation3DIntegral(
mesh, survey=survey, chiMap=sumMap, active_cells=actv, store_sensitivities="ram"
)
# Make sensitivity weighting
# Take the cell number out of the scaling.
# Want to keep high sens for large volumes
wr = (
prob.getJtJdiag(np.ones(sumMap.shape[1]))
/ np.r_[homogMap.P.T * mesh.cell_volumes[actv], mesh.cell_volumes[actv]] ** 2.0
)
# Scale the model spaces independently
wr[wires.homo.index] /= np.max((wires.homo * wr)) * utils.mkvc(
homogMap.P.sum(axis=0).flatten()
)
wr[wires.hetero.index] /= np.max(wires.hetero * wr)
wr = wr**0.5
## Create a regularization
# For the homogeneous model
regMesh = TensorMesh([len(domains)])
reg_m1 = regularization.Sparse(regMesh, mapping=wires.homo)
reg_m1.set_weights(weights=wires.homo * wr)
reg_m1.norms = [0, 2]
reg_m1.reference_model = np.zeros(sumMap.shape[1])
# Regularization for the voxel model
reg_m2 = regularization.Sparse(
mesh, active_cells=actv, mapping=wires.hetero, gradient_type="components"
)
reg_m2.set_weights(weights=wires.hetero * wr)
reg_m2.norms = [0, 0, 0, 0]
reg_m2.reference_model = np.zeros(sumMap.shape[1])
reg = reg_m1 + reg_m2
# Data misfit function
dmis = data_misfit.L2DataMisfit(simulation=prob, data=data)
# Add directives to the inversion
opt = optimization.ProjectedGNCG(
maxIter=100,
lower=0.0,
upper=1.0,
maxIterLS=20,
cg_maxiter=10,
cg_rtol=1e-3,
tolG=1e-3,
eps=1e-6,
)
invProb = inverse_problem.BaseInvProblem(dmis, reg, opt)
betaest = directives.BetaEstimate_ByEig(beta0_ratio=1e-2)
# Here is where the norms are applied
# Use pick a threshold parameter empirically based on the distribution of
# model parameters
IRLS = directives.UpdateIRLS(f_min_change=1e-3)
update_Jacobi = directives.UpdatePreconditioner()
inv = inversion.BaseInversion(invProb, directiveList=[IRLS, betaest, update_Jacobi])
# Run the inversion
m0 = np.ones(sumMap.shape[1]) * 1e-4 # Starting model
prob.model = m0
mrecSum = inv.run(m0)
if plotIt:
mesh.plot_3d_slicer(
actvMap * model,
aspect="equal",
zslice=30,
pcolor_opts={"cmap": "inferno_r"},
transparent="slider",
)
mesh.plot_3d_slicer(
actvMap * sumMap * mrecSum,
aspect="equal",
zslice=30,
pcolor_opts={"cmap": "inferno_r"},
transparent="slider",
)
if __name__ == "__main__":
run()
plt.show()
Total running time of the script: (0 minutes 25.905 seconds)
Estimated memory usage: 344 MB