Back to results
US 2016/0238729 A1Published application
Method of, and apparatus for, full waveform inversion
AssigneeImperial Innovations Limited
InventorsMike WARNER, Lluis GUASCH BATALLA
Life of this patent
Estimating the term…
Loading patent…
Loading patent…
AssigneeImperial Innovations Limited
InventorsMike WARNER, Lluis GUASCH BATALLA
Estimating the term…
A method of subsurface exploration includes generating a representation of a portional volume of the Earth from a seismic measurement of a physical parameter. The method includes providing observed seismic dataset having three distinct nonzero data values derived from three distinct nonzero seismic measured values of said portional volume of the Earth, generating a predicted seismic dataset having three distinct nonzero data values, generating a nontrivial convolutional filter including three nonzero filter coefficients, generating a convolved observed dataset by convolving the convolutional filter with said observed seismic dataset, generating primary objective functions to measure the similarity between said convolved observed dataset and said predicted dataset, maximizing and/or minimizing said primary objective functions by modifying at least one filter coefficient of the convolutional filter, generating predetermined reference filters having at least three reference coefficients generating secondary objective functions to measure the similarity between filter coefficients for the nontrivial filter and reference coefficients for the predetermined reference filters, and minimizing and/or maximizing said secondary objective functions by modifying a model coefficient of a subsurface model of a portion of the Earth to produce an updated subsurface model of a portion of the Earth.
A method of subsurface exploration, the method comprising generating a geophysical representation of a portion of the volume of the Earth from a seismic measurement of at least one physical parameter, and comprising the steps of:
A method of subsurface exploration, the method comprising generating a geophysical representation of a portion of the volume of the Earth from a seismic measurement of at least one physical parameter, and comprising the steps of:
A method according to any one of the preceding claims, wherein, subsequent to step h), the method further comprises:
A method according to any one of the preceding claims, wherein a plurality of convolutional filters is provided.
A method according to any one of the preceding claims, wherein a plurality of further convolutional filters is provided.
A method according to any one of the preceding claims, wherein a plurality of observed seismic data sets and/or a plurality of predicted seismic data sets is provided.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters involve convolution in time.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters involve convolution in space.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters involve convolution in more than one dimension.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters comprise Wiener filters.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters are calculated in the time domain.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters are calculated in the temporal-frequency domain.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters are calculated in the space domain.
A method according to any one of the preceding claims, wherein one or more of the convolutional filters and/or further convolutional filters are calculated in the wave-number domain.
A method according to any one of the preceding claims, wherein the filter coefficients are regularised and/or interpolated between said data sets and/or subsets of said data sets.
A method according to any one of the preceding claims, wherein the filter coefficients are regularised and/or interpolated between filters.
A method according to any one of the preceding claims, wherein the secondary objective function involves the application of a weighting to the filter coefficients of said first and/or said second convolutional filter, and wherein the weighting is dependant upon the temporal lag of said convolutional filter.
A method according to any one of the preceding claims, wherein the secondary objective function involves the application of a weighting to the filter coefficients of said first and/or said second convolutional filter, and wherein the weighting is dependant upon the spatial lag of said convolutional filter.
A method according to any one of the preceding claims, wherein at least one of said secondary objective functions comprises a norm misfit objective function.
A method according to any one of the preceding claims, wherein at least one of said secondary objective functions comprises a least-squares misfit objective function.
A method according to any one of the preceding claims, wherein step i) comprises minimising the gradient of one or more of the secondary objective functions with respect to said model parameters.
A method according any one of the preceding claims, wherein step i) is solved using adjoint-state methods.
A method according to any one of the preceding claims, wherein at least one of the pre-determined reference filters is a convolutional filter which leaves the dataset to which it is applied unaltered, or approximately unaltered, other than by multiplication by a constant.
A method according to any one of the preceding claims, wherein at least one of said pre-determined reference filters is a convolutional filter which leaves the dataset to which it is applied unaltered, or approximately unaltered, other than by a shift in time or space.
A method according to any one of the preceding claims, wherein at least one of said pre-determined reference filters is a convolutional filter which leaves the dataset to which it is applied unaltered, or approximately unaltered, other than by a rotation of phase.
A method according to any one of the preceding claims, wherein at least a portion of the observed seismic dataset is numerically propagated to a subsurface region of the model that is spatially removed from the location at which they were originally recorded.
A method according to any one of the preceding claims, further comprising, subsequent to step i), repeating steps b) to h) for the updated subsurface model of a portion of the Earth until a convergence criterion is met.
A method according to any one of the preceding claims, wherein step j) further comprises:
A method according to any one of the preceding claims, wherein said at least one physical parameter comprises pressure, particle velocity or displacement.
A method according to any one of the preceding claims, wherein the observed data set and the predicted data set comprise values of a plurality of physical parameters.
A method substantially as shown in and/or described with reference to any one or more of FIGS. 1 to 17 of the accompanying drawings.
A computer program product substantially as described with reference to any one or more of FIGS. 1 to 17 of the accompanying drawings.
Dependencies are inferred from the claim text.