fracspy.location.imaging.xcorri

fracspy.location.imaging.xcorri(data, n_xyz, Op, niter=100, l1eps=0.8, lr=1e-05, nforhc=10)

Cross-correlation-based imaging for microseismic source location

This routine performs imaging of microseismic data by inversion using an objective function based on the Pearson correlation coefficient. This idea is borrowed from the field of seismic migration, and more specifically from [1], and is intended to create a location algorithm that is less sensitive to inaccuracies in the knowledge of the source signature.

[1]

Zhang, Y., and Duan, L., and Xie, Y. “A stable and practical implementation of least-squares reverse time migration”, Geophysics, Vol. 80, 1, pp. 1JF-Z39, 2015.

Parameters:

datanumpy.ndarray

Data of shape :math`n_r imes n_t`

n_xyztuple

Number of grid points in X-, Y-, and Z-axes for the imaging area

Op:obj:` pyfrac.modelling.kirchhoff.Kirchhoff`

Kirchhoff operator

niterint, optional

Number of iterations

l1epsfloat, optional

Weight of the L1 regularization term

lrint, optional

Learning rate used by the optimizer

nforhcint, optional

Number of points for hypocenter

Returns:

mls_torchnumpy.ndarray

Migrated volume

hcnumpy.ndarray

Estimated hypocentral location

dls_torchnumpy.ndarray

Predicted data volume

losshistlist

Loss history