Automated fault uncertainty analysis in hydrocarbon exploration

1. A system comprising:

a processor; and

a memory including instructions that are executable by the processor for causing the processor to:

access a plurality of seismic images of a subterranean formation in a first geological area;

generate a plurality of fault estimates from each of the plurality of seismic images;

generate a processed seismic image of the first geological area by normalizing and merging the plurality of seismic images and the plurality of fault estimates;

generate a statistical fault uncertainty volume of the first geological area using the processed seismic image; and

control a drilling operation in the first geological area using the statistical fault uncertainty volume of the first geological area.

2. The system of claim 1 , wherein the instructions are further executable by the processor for causing the processor to:

generate a prediction model by training a deep neural network using the statistical fault uncertainty volume of the first geological area and a first set of seismic attributes of the first geological area; and

generate a confidence index of a fault location in a second geological area by applying the prediction model to a second set of seismic attributes of the second geological area.

3. The system of claim 2 , wherein the instructions are further executable by the processor for causing the processor to:

control a drilling operation in the second geological area using the confidence index of the fault location in the second geological area.

4. The system of claim 1 , wherein the plurality of seismic images comprise at least three different seismic images.

5. The system of claim 4 , wherein the plurality of seismic images are generated from a plurality of sub-angle ranges used to receive seismic wave reflections from the subterranean formation in the first geological area.

6. The system of claim 1 , wherein controlling the drilling operation in the first geological area comprises controlling the drilling operation to avoid a portion of the first geological area where the statistical fault uncertainty volume indicates a high degree of uncertainty in a fault-presence prediction.

7. The system of claim 1 , wherein controlling the drilling operation in the first geological area is performed prior to drilling into a fault in the first geological area.

8. The system of claim 1 , wherein the statistical fault uncertainty volume comprises an indication of uncertainty that a particular pixel associated with a location in a subterranean formation is along a fault.

9. A method comprising:

accessing, by a processor, a plurality of seismic images of a subterranean formation in a first geological area;

generating, by the processor, a plurality of fault estimates from each of the plurality of seismic images;

generating, by the processor, a processed seismic image of the first geological area by normalizing and merging the plurality of seismic images and the plurality of fault estimates;

generating, by the processor, a statistical fault uncertainty volume of the first geological area using the processed seismic image;

generating, by the processor, a prediction model by training a machine-learning architecture using the statistical fault uncertainty volume of the first geological area and a first set of seismic attributes of the first geological area; and

generating, by the processor, a confidence index of a fault location in a second geological area by applying the prediction model to a second set of seismic attributes of the second geological area.

10. The method of claim 9 , further comprising:

controlling, by the processor, a drilling operation in the second geological area using the confidence index of the fault location in the second geological area.

11. The method of claim 9 , further comprising:

controlling, by the processor, a drilling operation in the first geological area using the statistical fault uncertainty volume of the first geological area.

12. The method of claim 11 , wherein controlling the drilling operation in the first geological area is performed prior to drilling into a fault in the first geological area.

13. The method of claim 9 , wherein the statistical fault uncertainty volume comprises an indication of uncertainty that a particular pixel associated with a location in a subterranean formation is along a fault.

14. The method of claim 9 , wherein generating the plurality of fault estimates from each of the plurality of seismic images comprises:

processing, by the processor, the plurality of seismic images using a plurality of algorithmic hyper-parameter settings on each of the plurality of seismic images to generate the plurality of fault estimates.

15. A non-transitory computer-readable medium comprising program code that is executable by a processor for causing the processor to:

access a plurality of seismic images of a subterranean formation in a first geological area;

generate a plurality of fault estimates from each of the plurality of seismic images;

generate a processed seismic image of the first geological area by normalizing and merging the plurality of seismic images and the plurality of fault estimates;

generate a statistical fault uncertainty volume of the first geological area using the processed seismic image;

generate a prediction model by training a deep neural network using the statistical fault uncertainty volume of the first geological area and a first set of seismic attributes of the first geological area; and

generate a confidence index of a fault location in a second geological area by applying the prediction model to a second set of seismic attributes of the second geological area.

16. The non-transitory computer-readable medium of claim 15 , wherein the program code is further executable by the processor for causing the processor to:

control a drilling operation in the second geological area using the confidence index of the fault location in the second geological area.

17. The non-transitory computer-readable medium of claim 15 , wherein the program code is further executable by the processor for causing the processor to:

control a drilling operation in the first geological area using the statistical fault uncertainty volume of the first geological area.

18. The non-transitory computer-readable medium of claim 15 , wherein the statistical fault uncertainty volume comprises an indication of uncertainty that a particular pixel associated with a location in a subterranean formation is along a fault.