WELLBORE TRAJECTORY SYSTEM
Embodiments disclosed herein generally relate to a method and a system to generate well trajectories. A computing device receives one or more parameters associated with a target well in a target location. The computing device receives two or more data points for the target well in the target location. The computing device generates a modified wellbore path based on the one or more parameters associated with a target well and the two or more data points via a trained wellbore prediction model. The computing device compares the modified wellbore path for the target well to one or more wellbore paths of one or more wells co-located with the target well in the target location. The computing device updates the modified wellbore path for the target well by adjusting one or more coordinates of the modified wellbore path based on the comparison. The computing device generates a three-dimensional model of the target location.
1 . A method, comprising:
receiving, by a computing system, two or more locational data points for a proposed wellbore path to be generated for a proposed well in a target location;
generating, by the computing system, the proposed wellbore path for an entirety of the proposed well based on the two or more locational data points via a trained wellbore prediction model, wherein the proposed wellbore path extends through a first locational data point of the two or more locational data points and a second locational data point of the two or more locational data points, and wherein the proposed wellbore path comprises a plurality of locational data points in addition to the first locational data point and the second locational data point;
comparing, by the computing system, the plurality of locational data points of the proposed wellbore path for the proposed well to a further plurality of locational data points of one or more wellbore paths of one or more wells co-located with the proposed well in the target location;
determining, by the computing system, that at least one locational data point of the plurality of locational data points is within a threshold distance of at least one locational data point of the further plurality of locational data points of a wellbore path of the one or more wellbore paths; and
based on the determining, generating, by the computing system, a proposed modified wellbore path for the proposed well by adjusting the at least one locational data point of the plurality of locational data points of the proposed wellbore path.
2 . The method of claim 1 , further comprising:
generating, by the computing system, a three-dimensional model of the target location, wherein the three-dimensional model of the target location comprises the one or more wellbore paths of the one or more wells co-located with the proposed well and the proposed modified wellbore path for the proposed well.
3 . The method of claim 1 , further comprising:
adding, by the computing system, the proposed modified wellbore path into the target location; and
generating, by the computing system, a production prediction for the target location based on adding the proposed modified wellbore path.
4 . The method of claim 3 , further comprising:
associating, by the computing system, the production prediction with a reward amount;
determining, by the computing system, that the reward amount is greater than zero; and
generating, by the computing system, a second proposed wellbore path using the target location with the proposed modified wellbore path.
5 . The method of claim 1 , further comprising:
receiving, by the computing system, one or more desired parameters associated the proposed well.
6 . The method of claim 5 , wherein the proposed wellbore path is generated in accordance with the one or more desired parameters.
7 . The method of claim 1 , wherein the trained wellbore prediction model is generated by:
gathering, from one or more remote computing devices, a plurality of sets of historical wellbore information for a plurality of test wellbores;
standardizing the plurality of sets of historical wellbore information from a first state to a second state, wherein the second state is uniform across each of the plurality of sets of historical wellbore information; and
training a machine learning model based on the standardized historical wellbore information.
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:
receiving, by the computing system, two or more locational data points for a proposed wellbore path to be generated for a proposed well in a target location;
generating, by the computing system, the proposed wellbore path for an entirety of the proposed well based on the two or more locational data points via a trained wellbore prediction model, wherein the proposed wellbore path extends through a first locational data point of the two or more locational data points and a second locational data point of the two or more locational data points, and wherein the proposed wellbore path comprises a plurality of locational data points in addition to the first locational data point and the second locational data point;
comparing, by the computing system, the plurality of locational data points of the proposed wellbore path for the proposed well to a further plurality of locational data points of one or more wellbore paths of one or more wells co-located with the proposed well in the target location;
determining, by the computing system, that at least one locational data point of the plurality of locational data points is within a threshold distance of at least one locational data point of the further plurality of locational data points of a wellbore path of the one or more wellbore paths; and
based on the determining, generating, by the computing system, a proposed modified wellbore path for the proposed well by adjusting the at least one locational data point of the plurality of locational data points of the proposed wellbore path.
9 . The non-transitory computer readable medium of claim 8 , further comprising:
generating, by the computing system, a three-dimensional model of the target location, wherein the three-dimensional model of the target location comprises the one or more wellbore paths of the one or more wells co-located with the proposed well and the proposed modified wellbore path for the proposed well.
10 . The non-transitory computer readable medium of claim 8 , further comprising:
adding, by the computing system, the proposed modified wellbore path into the target location; and
generating, by the computing system, a production prediction for the target location based on adding the proposed modified wellbore path.
11 . The non-transitory computer readable medium of claim 10 , further comprising:
associating, by the computing system, the production prediction with a reward amount;
determining, by the computing system, that the reward amount is greater than zero; and
generating, by the computing system, a second proposed wellbore path using the target location with the proposed modified wellbore path.
12 . The non-transitory computer readable medium of claim 8 , further comprising:
receiving, by the computing system, one or more desired parameters associated the proposed well.
13 . The non-transitory computer readable medium of claim 12 , wherein the proposed wellbore path is generated in accordance with the one or more desired parameters.
14 . The non-transitory computer readable medium of claim 8 , wherein the trained wellbore prediction model is generated by:
gathering, from one or more remote computing devices, a plurality of sets of historical wellbore information for a plurality of test wellbores;
standardizing the plurality of sets of historical wellbore information from a first state to a second state, wherein the second state is uniform across each of the plurality of sets of historical wellbore information; and
training a machine learning model based on the standardized historical wellbore information.
15 . A system, comprising:
one or more processors; and
a memory comprising one or more sequences of instructions, which, when executed by the one or more processors, causes the system to perform operations comprising:
receiving two or more locational data points for a proposed wellbore path to be generated for a proposed well in a target location;
generating the proposed wellbore path for an entirety of the proposed well based on the two or more locational data points via a trained wellbore prediction model, wherein the proposed wellbore path extends through a first locational data point of the two or more locational data points and a second locational data point of the two or more locational data points, and wherein the proposed wellbore path comprises a plurality of locational data points in addition to the first locational data point and the second locational data point;
comparing the plurality of locational data points of the proposed wellbore path for the proposed well to a further plurality of locational data points of one or more wellbore paths of one or more wells co-located with the proposed well in the target location;
determining that at least one locational data point of the plurality of locational data points is within a threshold distance of at least one locational data point of the further plurality of locational data points of a wellbore path of the one or more wellbore paths; and
based on the determining, generating a proposed modified wellbore path for the proposed well by adjusting the at least one locational data point of the plurality of locational data points of the proposed wellbore path.
16 . The system of claim 15 , wherein the operations further comprise:
generating a three-dimensional model of the target location, wherein the three-dimensional model of the target location comprises the one or more wellbore paths of the one or more wells co-located with the proposed well and the proposed modified wellbore path for the proposed well.
17 . The system of claim 15 , wherein the operations further comprise:
adding the proposed modified wellbore path into the target location; and
generating a production prediction for the target location based on adding the proposed modified wellbore path.
18 . The system of claim 17 , wherein the operations further comprise:
associating the production prediction with a reward amount;
determining that the reward amount is greater than zero; and
generating a second proposed wellbore path using the target location with the proposed modified wellbore path.
19 . The system of claim 15 , wherein the operations further comprise:
receiving one or more desired parameters associated the proposed well, and
wherein the proposed wellbore path is generated in accordance with the one or more desired parameters.
20 . The system of claim 15 , wherein the trained wellbore prediction model is generated by:
gathering, from one or more remote computing devices, a plurality of sets of historical wellbore information for a plurality of test wellbores;
standardizing the plurality of sets of historical wellbore information from a first state to a second state, wherein the second state is uniform across each of the plurality of sets of historical wellbore information; and
training a machine learning model based on the standardized historical wellbore information.