244
9 High-Dimension Model Representation via Sparse GridTechniques
IR = 9.34mm
OR = 18.4mm
HT = 9mm
Turns = 408
Transmit Coil:
Cores:
R = 7.34mm
2mm
Receive Coils:
IR = 7.34mm
OR = 8.34mm
HT = 9mm
μ=2000
HT=9mm
Turns = 100
Fig. 9.31 The split-D coil configuration
Fig. 9.32 Illustrating the two
Euler angles that define the
orientation of the split-D
probe in the test problem. The
probe axis is orthogonal to the
plane of Fig. 9.31. The input
model data for the inversion
are Ψ = 22 ◦ and Θ = 0 ◦
X
Z
Y
Θ
Ψ
Probe Axis
9.10 Noisy Data and Uncertainty Propagation
In developing the stochastic inverse model of Chap. 6, we assumed that the input
data were given, and that the only stochastic feature of the problem was the random
vector of unknown parameters that were to be determined. Each component of the
vector was uniformly distributed over a certain range.
9 High-Dimension Model Representation via Sparse GridTechniques
IR = 9.34mm
OR = 18.4mm
HT = 9mm
Turns = 408
Transmit Coil:
Cores:
R = 7.34mm
2mm
Receive Coils:
IR = 7.34mm
OR = 8.34mm
HT = 9mm
μ=2000
HT=9mm
Turns = 100
Fig. 9.31 The split-D coil configuration
Fig. 9.32 Illustrating the two
Euler angles that define the
orientation of the split-D
probe in the test problem. The
probe axis is orthogonal to the
plane of Fig. 9.31. The input
model data for the inversion
are Ψ = 22 ◦ and Θ = 0 ◦
X
Z
Y
Θ
Ψ
Probe Axis
9.10 Noisy Data and Uncertainty Propagation
In developing the stochastic inverse model of Chap. 6, we assumed that the input
data were given, and that the only stochastic feature of the problem was the random
vector of unknown parameters that were to be determined. Each component of the
vector was uniformly distributed over a certain range.