US 12,455,392 B2Granted

Method to correct VSP data

AssigneeSaudi Arabian Oil Company

InventorsMohammed Ibrahim Albabli, Hussain Alnasser, Hamad Alswaidan

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods are disclosed. The method includes obtaining vertical seismic profiling (VSP) data and surface seismic (SS) data for a subterranean region of interest. The VSP data includes a corrupt section and a valid section. The method further includes determining a VSP attribute and a VSP spectrum using the VSP data, determining an SS attribute using the SS data, and determining a corrected VSP attribute for the corrupt section. The method still further includes training a neural network using the VSP attribute, the SS attribute, and the VSP spectrum for the valid section, predicting a corrected VSP spectrum for the corrupt section by inputting the corrected VSP attribute and the SS attribute for the corrupt section into the trained neural network, and determining corrected VSP data for the corrupt section using the corrected VSP attribute and the corrected VSP spectrum.

Claims18

Dependent claims

A method comprising:

obtaining vertical seismic profiling (VSP) data for a subterranean region of interest,
wherein the VSP data comprises a plurality of VSP traces, and
wherein each of the plurality of VSP traces comprises a plurality of VSP time samples;
separating the VSP data into a corrupt VSP section and a valid VSP section;
obtaining surface seismic (SS) data for the subterranean region of interest,
wherein the SS data comprises a plurality of SS traces, and
wherein each of the plurality of SS traces comprises a plurality of SS time samples;
separating the SS data into a first valid SS section corresponding to the corrupt VSP section and a second valid SS section corresponding to the valid VSP section;
for each of the plurality of VSP time samples among each of the plurality of VSP traces within the valid VSP section, determining a VSP attribute and a VSP spectrum using the VSP data;
for each of the plurality of SS time samples among each of the plurality of SS traces, determining an SS attribute using the SS data;
for each of the plurality of VSP time samples among each of the plurality of VSP traces within the corrupt VSP section, determining a corrected VSP attribute using the VSP attribute and the SS attribute;
training a neural network using the VSP attribute, the SS attribute associated with the second valid SS section, and the VSP spectrum,
wherein the neural network is trained to predict a corrected VSP spectrum associated with the corrupt VSP section based on the corrected VSP attribute and the SS attribute associated with the first valid SS section;
predicting the corrected VSP spectrum associated with the corrupt VSP section by inputting the corrected VSP attribute and the SS attribute associated with the first valid SS section into the trained neural network;
determining corrected VSP data within the corrupt VSP section using the corrected VSP attribute and the corrected VSP spectrum;
determining a velocity model using the corrected VSP data;
transforming the SS data from a time domain to a depth domain using the velocity model;
and
locating a hydrocarbon reservoir using the SS data in the depth domain.

16 more elements

A system comprising:

a vertical seismic profiling (VSP) acquisition system configured to obtain VSP data for a subterranean region of interest,
wherein the VSP data comprises a plurality of VSP traces, and
wherein each of the plurality of VSP traces comprises a plurality of VSP time samples;
a surface seismic (SS) acquisition system configured to obtain SS data for the subterranean region of interest,
wherein the SS data comprises a plurality of SS traces, and
wherein each of the plurality of SS traces comprises a plurality of SS time samples;
a seismic processing system configured to:
receive, from the VSP acquisition system, the VSP data;
separating the VSP data into a corrupt VSP section and a valid VSP section,
receive, from the SS acquisition system, the SS data;
separating the SS data into a first valid SS section corresponding to the corrupt VSP section and a second valid SS section corresponding to the valid VSP section;
for each of the plurality of VSP time samples among each of the plurality of VSP traces within the valid VSP section, determine a VSP attribute and a VSP spectrum using the VSP data,
for each of the plurality of SS time samples among each of the plurality of SS traces, determine an SS attribute using the SS data, and
for each of the plurality of VSP time samples among each of the plurality of VSP traces within the corrupt VSP section, determine a corrected VSP attribute using the VSP attribute and the SS attribute;
and
a computer processor configured to:
train a neural network using the VSP attribute, the SS attribute associated with the second valid SS section, and the VSP spectrum,
wherein the neural network is trained to predict a corrected VSP spectrum associated with the corrupt VSP section based on the corrected VSP attribute and the SS attribute associated with the first valid SS section,
predict the corrected VSP spectrum associated with the corrupt VSP section by inputting the corrected VSP attribute and the SS attribute associated with the first valid SS section into the trained neural network,
determine corrected VSP data within the corrupt VSP section using the corrected VSP attribute and the corrected VSP spectrum,
determine a velocity model using the corrected VSP data, and
transform the SS data from a time domain to a depth domain using the velocity model;
and
a seismic interpretation workstation configured to locate a hydrocarbon reservoir using the SS data in the depth domain.

21 more elements

Dependencies are inferred from the claim text.