US 11,377,932 B2 · Machine learning-based reservoir reserves estimation

Abstract

A system, method, and computer program product for reservoir reserves estimation that includes at least one processing component, at least one memory component, a well data manager, a classification module, and an estimation module. The well data manager consolidates well data for a subject well and at least one neighbor well. The classification module includes a search component that selects at least one closest neighbor well from the neighbor wells. The classification module also includes a similarity matching component that determines a log signature distance between the at least one closest neighbor well and the subject well, as well as an analog selector that selects one or more analog wells from the at least one closest neighbor well. The estimation module includes a decline estimator that generates group decline rates for the one or more analog wells and a subject decline rate based on the group decline rates.

Claims20

Dependent claims

A system for reservoir reserves estimation, comprising:

a memory;
a processor communicatively coupled to the memory, wherein the processor is configured to:
consolidate well data for a group of wells;
select a subject well and at least one neighbor well from the group of wells;
select at least one closest neighbor well from the at least one neighbor well;
determine, for the at least one closest neighbor well, a log signature distance from the subject well;
determine, for the at least one closest neighbor well, a geographical distance from the subject well;
and
generate an analog score based on a weighted value of the log signature distance and a weighted value of the geographical distance;
select, based on the analog score, one or more analog wells from the at least one closest neighbor well;
generate group decline rates for the one or more analog wells;
and
generate a subject decline rate for the subject well based on the group decline rates.

10 more elements

A computer program product for reservoir reserves estimation, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause a device to perform a method, the method comprising:

consolidating well data for a group of wells;
selecting a subject well and at least one neighbor well from the group of wells;
selecting at least one closest neighbor well from the at least one neighbor well;
determining, for the at least one closest neighbor well, a log signature distance from the subject well;
determining, for the at least one closest neighbor well, a geographical distance from the subject well;
generating an analog score based on a weighted value of the log signature distance and a weighted value of the geographical distance;
selecting, based on the analog score, one or more analog wells from the at least one closest neighbor well;
generating group decline rates for the one or more analog wells;
and
generating a subject decline rate for the subject well based on the group decline rates.

7 more elements

Dependencies are inferred from the claim text.

Machine learning-based reservoir reserves estimation

Life of this patent

Machine learning-based reservoir reserves estimation

Life of this patent

Abstract

Claims20