System and method for oil and gas predictive analytics
Embodiments disclosed herein generally relate to a method and system for oil and gas predictive analytics. A computer system receives a set of production information for a well located in a region. The computing system generates a set of general reference groups comprising one or more reference wells for the region. The computing system determines whether the set of production information for the well includes the threshold amount of production information. The computing system selects a subset of reference wells from the general reference groups based on one or more traits of the well. The computing system generates a reference curve based on the set of production information associated with each reference well in the subset of reference wells. The computing system fits a decline curve to the reference curve, to determine an estimated ultimate recovery of the well.
1. A method, comprising:
identifying, by a computing system, a set of production information for a well located in a region, the set of production information directed to past production of the well;
partitioning, by the computing system, a curve representing the past production of the well into a plurality of segments;
determining, by the computing system, that each segment of the plurality of segments comprises a first set of production information that includes a threshold amount of production information;
based on the determining, fitting, by the computing system, a decline curve to the first set of production information using a decline curve model; and
generating, by the computing system, an estimated ultimate recovery of the well based on the decline curve.
2. The method of claim 1 , wherein fitting, by the computing system, the decline curve to the first set of production information using the decline curve model comprises:
identifying a peak producing rate in the first set of production information; and
fitting the decline curve to the first set of production information based on the peak producing rate.
3. The method of claim 1 , wherein fitting, by the computing system, the decline curve to the first set of production information using the decline curve model comprises:
determining that multiple peaks in production rate in the first set of production information exist; and
based on the determining that the multiple peaks in the production rate in the first set of production information exist, utilizing a transition point detection algorithm for identifying a transition point in the production rate in the first set of production information.
4. The method of claim 3 , wherein utilizing the transition point detection algorithm for identifying the transition point in the production rate in the first set of production information comprises:
iteratively comparing each of the multiple peaks to determine which peak of the multiple peaks is the transition point.
5. The method of claim 1 , wherein fitting, by the computing system, the decline curve to the first set of production information using the decline curve model comprises:
determining that multiple peaks in production rate in the first set of production information exist;
identifying a set of candidate transition points based on the multiple peaks;
fitting individual decline curves to the set of candidate transition points; and
identifying an actual transition point based on a candidate transition point from the set of candidate transition points that yielded a lowest root mean square error.
6. The method of claim 1 , wherein partitioning, by the computing system, the curve representing the past production of the well into the plurality of segments comprises:
applying a step detection process to the curve.
7. The method of claim 1 , wherein partitioning, by the computing system, the curve representing the past production of the well into the plurality of segments comprises:
identifying one or more points of change in a time series representation of the first set of production information; and
partitioning the curve based on the one or more points of change.
8. A non-transitory computer readable medium comprising one or more sequences of instructions, which, when executed by a processor, causes a computing system to perform operations comprising:
identifying, by the computing system, a set of production information for a well located in a region, the set of production information directed to past production of the well;
partitioning, by the computing system, a curve representing the past production of the well into a plurality of segments;
determining, by the computing system, that each segment of the plurality of segments comprises a first set of production information that includes a threshold amount of production information;
based on the determining, fitting, by the computing system, a decline curve to the first set of production information using a decline curve model; and
generating, by the computing system, an estimated ultimate recovery of the well based on the decline curve.
9. The non-transitory computer readable medium of claim 8 , wherein fitting, by the computing system, the decline curve to the first set of production information using the decline curve model comprises:
identifying a peak producing rate in the first set of production information; and
fitting the decline curve to the first set of production information based on the peak producing rate.
10. The non-transitory computer readable medium of claim 8 , wherein fitting, by the computing system, the decline curve to the first set of production information using the decline curve model comprises:
determining that multiple peaks in production rate in the first set of production information exist; and
based on the determining that the multiple peaks in the production rate in the first set of production information exist, utilizing a transition point detection algorithm for identifying a transition point in the production rate in the first set of production information.
11. The non-transitory computer readable medium of claim 10 , wherein utilizing the transition point detection algorithm for identifying the transition point in the production rate in the first set of production information comprises:
iteratively comparing each of the multiple peaks to determine which peak of the multiple peaks is the transition point.
12. The non-transitory computer readable medium of claim 8 , wherein fitting, by the computing system, the decline curve to the first set of production information using the decline curve model comprises:
determining that multiple peaks in production rate in the first set of production information exist;
identifying a set of candidate transition points based on the multiple peaks;
fitting individual decline curves to the set of candidate transition points; and
identifying an actual transition point based on a candidate transition point from the set of candidate transition points that yielded a lowest root mean square error.
13. The non-transitory computer readable medium of claim 8 , wherein partitioning, by the computing system, the curve representing the past production of the well into the plurality of segments comprises:
applying a step detection process to the curve.
14. The non-transitory computer readable medium of claim 8 , wherein partitioning, by the computing system, the curve representing the past production of the well into the plurality of segments comprises:
identifying one or more points of change in a time series representation of the first set of production information; and
partitioning the curve based on the one or more points of change.
15. A system, comprising:
a processor; and
a memory having programming instructions stored thereon, which, when executed by the processor, causes the system to perform operations comprising:
identifying a set of production information for a well located in a region, the set of production information directed to past production of the well;
partitioning a curve representing the past production of the well into a plurality of segments;
determining that each segment of the plurality of segments comprises a first set of production information that includes a threshold amount of production information;
based on the determining, fitting a decline curve to the first set of production information using a decline curve model; and
generating an estimated ultimate recovery of the well based on the decline curve.
16. The system of claim 15 , wherein fitting the decline curve to the first set of production information using the decline curve model comprises:
identifying a peak producing rate in the first set of production information; and
fitting the decline curve to the first set of production information based on the peak producing rate.
17. The system of claim 15 , wherein fitting the decline curve to the first set of production information using the decline curve model comprises:
determining that multiple peaks in production rate in the first set of production information exist; and
based on the determining that the multiple peaks in the production rate in the first set of production information exist, utilizing a transition point detection algorithm for identifying a transition point in the production rate in the first set of production information.
18. The system of claim 15 , wherein fitting the decline curve to the first set of production information using the decline curve model comprises:
determining that multiple peaks in production rate in the first set of production information exist;
identifying a set of candidate transition points based on the multiple peaks;
fitting individual decline curves to the set of candidate transition points; and
identifying an actual transition point based on a candidate transition point from the set of candidate transition points that yielded a lowest root mean square error.
19. The system of claim 15 , wherein partitioning the curve representing the past production of the well into the plurality of segments comprises:
applying a step detection process to the curve.
20. The system of claim 15 , wherein partitioning the curve representing the past production of the well into the plurality of segments comprises:
identifying one or more points of change in a time series representation of the first set of production information; and
partitioning the curve based on the one or more points of change.