Diffusion modeling based subsurface formation evaluation

1 . A method for controlling a computer to geologically characterize a space relative to a borehole, the method comprising:

configuring a diffusion process applied to information and data about samples of reservoir parameters; and

determining, via the diffusion process, a probability distribution of the reservoir parameters in the space relative to the borehole.

2 . The method of claim 1 , wherein the information and data include at least one of geophysical data, borehole data, or geological information.

3 . The method of claim 1 , wherein the reservoir parameters include at least one of porosity, salt, saturation, permeability, or fault.

4 . The method of claim 1 , wherein the reservoir parameters relate to at least one of a hydrocarbon reservoir or a salt reservoir.

5 . The method of claim 1 , wherein configuring the diffusion process includes:

inputting, into a learning machine, sample images indicating the samples of the reservoir parameters, wherein the samples of the reservoir parameters have the probability distribution.

6 . The method of claim 5 , further comprising:

generating a generated image depicting the reservoir parameters in the space relative to the borehole.

7 . The method of claim 1 , wherein the configuring the diffusion process includes:

generating, by a neural network of the computer, a first noisy image by adding a first amount of noise to a sample seismic image, wherein the sample seismic image represents at least one of the reservoir parameters of the space relative to the borehole, and wherein the at least one of the reservoir parameters in the sample seismic image has the probability distribution;

determining, by the neural network, a first estimated amount of noise in the first noisy image;

determining, by the neural network, a difference between the first estimated amount of noise and the first amount of noise; and

updating the neural network based on the difference.

8 . The method of claim 7 , further comprising:

receiving a second seismic image in the neural network; and

generating, based on the second seismic image and on the probability distribution of the reservoir parameters, an output image indicating one or more of the reservoir parameters in the space relative to the borehole.

9 . One or more non-transitory machine-readable storage mediums including instructions, which when executed by one or more processors, control a computer to geologically characterize a subsurface space, the instructions comprising:

instructions to configure a diffusion process applied to information and data about samples of reservoir parameters; and

instructions to determine, via the diffusion process, a probability distribution of the reservoir parameters in the subsurface space.

10 . The one or more non-transitory machine-readable storage mediums of claim 9 , wherein the information and data include at least one or more geophysical data, borehole data, or geological information.

11 . The one or more non-transitory machine-readable storage mediums of claim 9 , wherein the reservoir parameters include at least one of porosity, salt, saturation, permeability, or fault.

12 . The one or more non-transitory machine-readable storage mediums of claim 9 , wherein the reservoir parameters relate to at least one of a hydrocarbon reservoir or a salt reservoir.

13 . The one or more non-transitory machine-readable storage mediums of claim 9 , wherein the instructions to configure the diffusion process include:

instructions to input, into the diffusion process, sample images indicating the samples of the reservoir parameters, wherein the samples of the reservoir parameters have the probability distribution.

14 . The one or more non-transitory machine-readable storage mediums of claim 13 , further comprising:

generating a generated image depicting the reservoir parameters in the subsurface space.

15 . The one or more non-transitory machine-readable storage mediums of claim 9 , wherein the instructions to configure the diffusion process include:

instructions to generate, by a neural network, a first noisy image by adding a first amount of noise to a sample seismic image, wherein the sample seismic image represents at least one of the reservoir parameters of the subsurface space, and wherein the at least one of the reservoir parameters in the sample seismic image has the probability distribution;

instructions to determine, by the neural network, a first estimated amount of noise in the first noisy image;

instructions to determine, by the neural network, a difference between the first estimated amount of noise and the first amount of noise; and

instructions to update the neural network based on the difference.

16 . The one or more non-transitory machine-readable storage mediums of claim 9 , further comprising:

receiving a second seismic image in the neural network; and

generating, based on the second seismic image and on the probability distribution of the reservoir parameters, a generative image indicating one or more of the reservoir parameters in the subsurface space.

17 . A system comprising:

one or more processors;

one or more machine-readable mediums including instructions, which when executed by the one or more of the processors, to control the one or more processors to geologically characterize a subsurface space, the instructions comprising:

instructions to configure a diffusion process applied to information and data about samples of reservoir parameters; and

instructions to determine, via the diffusion process, a probability distribution of the reservoir parameters in the subsurface space.

18 . The system of claim 17 , wherein the information and data include at least one or more of geophysical data, borehole data, or geological information.

19 . The system of claim 17 , wherein the reservoir parameters include at least one of porosity, salt, saturation, permeability, or fault.

20 . The system of claim 17 , wherein the reservoir parameters relate to at least one of a hydrocarbon reservoir or a salt reservoir.