Markov decision process-based support tool for reservoir development planning
First Claim
1. A method for reservoir development planning, comprising the steps of:
- receiving data relevant to reservoir development;
processing the received data by using a computer-implemented Markov decision process-based model,wherein the using the computer-implemented Markov decision process-based model includesdefining a plurality of stages, each stage representing a discrete step in time;
defining a plurality of states in each stage, each state representing a potential state of the reservoir development plan, anddetermining a future state by using a probability that the reservoir development transitions from a current state in a current stage to the future state in a subsequent stage, the probability being determined by the current state of the reservoir development and an action to be taken in the current stage that causes the transition to the subsequent stage; and
generating a reservoir development plan in response to processing the received data via the computer-implemented Markov decision process-based model.
0 Assignments
0 Petitions
Accused Products
Abstract
A Markov decision process-based support tool for reservoir development planning can comprise a source of input data, an optimization model, a high fidelity model for simulating the reservoir, and one or more solution routines interfacing with the optimization model. The optimization model can consider unknown parameters having uncertainties directly within the optimization model. The model incorporates the flexibility that a decision-maker has in the real world and allows the decision-maker to adjust the decisions based on new information. The model can systematically address uncertain data, for example comprehensively or even taking all uncertain data into account. Accordingly, the optimization model can provide flexible or robust solutions that remain feasible over an uncertainty space. Once the reservoir model is optimized, final development plans may be generated.
-
Citations
24 Claims
-
1. A method for reservoir development planning, comprising the steps of:
-
receiving data relevant to reservoir development; processing the received data by using a computer-implemented Markov decision process-based model, wherein the using the computer-implemented Markov decision process-based model includes defining a plurality of stages, each stage representing a discrete step in time; defining a plurality of states in each stage, each state representing a potential state of the reservoir development plan, and determining a future state by using a probability that the reservoir development transitions from a current state in a current stage to the future state in a subsequent stage, the probability being determined by the current state of the reservoir development and an action to be taken in the current stage that causes the transition to the subsequent stage; and generating a reservoir development plan in response to processing the received data via the computer-implemented Markov decision process-based model. - View Dependent Claims (2, 3, 4, 5, 6, 24)
-
-
7. A method for reservoir development planning, comprising the steps of:
-
receiving data relevant to reservoir development planning, wherein uncertainty is associated with the data; processing the received data by using a computer-implemented Markov decision process-based model, wherein the using the computer-implemented Markov decision process-based model includes defining a plurality of stages, each stage representing a discrete step in time; defining a plurality of states in each stage, each state representing a potential state of the reservoir development plan, and determining a future state by using a probability that the reservoir development transitions from a current state in a current stage to the future state in a subsequent stage, the probability being determined by the current state of the reservoir development and an action to be taken in the current stage that causes the transition to the subsequent stage; producing at least some portion of a reservoir development plan in response to processing the received data with the computer-based optimization model that incorporates the uncertainty; and undertaking one or more corrective decisions as the uncertainty unfolds over time. - View Dependent Claims (8, 9, 10)
-
-
11. A method for decision support regarding development of petroleum resources, comprising the steps of:
-
receiving a plurality of data elements regarding a decision about developing a petroleum reservoir, wherein a respective characterization of uncertainty is associated with each of the data elements; and outputting a recommendation for the decision in response to processing each of the data elements and each of the respective characterizations of uncertainty by using a computer-implemented Markov decision process-based model, wherein the using the computer-implemented Markov decision process-based model includes; defining a plurality of stages, each stage representing a discrete step in time; defining a plurality of states in each stage, each state representing a potential state of the reservoir development plan, and determining a future state by using a probability that the reservoir development transitions from a current state in a current stage to the future state in a subsequent stage, the probability being determined by the current state of the reservoir development and an action to be taken in the current stage that causes the transition to the subsequent stage. - View Dependent Claims (12, 13)
-
-
14. A computer-based method of optimizing development planning for a reservoir, comprising:
-
providing input data that comprises uncertainty; generating a first simulation of the reservoir using a high fidelity model for reservoir or surface facility behavior utilizing the input data, wherein the first simulation generates first high fidelity output data; generating a low fidelity model for reservoir or surface facility behavior using the input data and the first high fidelity output data, wherein the low fidelity model generates a prediction; optimizing a reservoir development planning model utilizing the input data and the low fidelity model for reservoir or surface facility behavior, wherein the reservoir development planning model generates reservoir development planning output data, wherein the optimizing the reservoir development planning model comprises using a computer-implemented Markov decision process-based model, and the using the computer-implemented Markov decision process-based model includes; defining a plurality of stages, each stage representing a discrete step in time; defining a plurality of states in each stage, each state representing a potential state of the reservoir development plan, and determining a future state by using a probability that the reservoir development transitions from a current state in a current stage to the future state in a subsequent stage, the probability being determined by the current state of the reservoir development and an action to be taken in the current stage that causes the transition to the subsequent stage; performing a second simulation of the reservoir using the high fidelity model for reservoir or surface facility behavior utilizing the input data and the reservoir development planning output data, wherein the second simulation generates second high fidelity output data; comparing the second high fidelity output data with the prediction of the low fidelity model; repeating the steps comprising generating a low fidelity model for reservoir or surface facility behavior, optimizing a reservoir development planning model, performing a second simulation of the reservoir using the high fidelity model for reservoir or surface facility behavior, and comparing the second high fidelity output data with the prediction of the low fidelity model until the second high fidelity output data is substantially consistent with the prediction of the low fidelity model. - View Dependent Claims (15, 16, 17)
-
-
18. A method of producing hydrocarbons from a subterranean reservoir, comprising:
-
generating a reservoir development planning system based on input data; optimizing the reservoir development planning system according to an uncertainty space, wherein the reservoir development planning system is optimized using a computer-implemented Markov decision process-based model, wherein the using the computer-implemented Markov decision process-based model includes; defining a plurality of stages, each stage representing a discrete step in time; defining a plurality of states in each stage, each state representing a potential state of the reservoir development plan, and determining a future state by using a probability that the reservoir development transitions from a current state in a current stage to the future state in a subsequent stage, the probability being determined by the current state of the reservoir development and an action to be taken in the current stage that causes the transition to the subsequent stage; and producing hydrocarbons from the reservoir according to output from the optimized reservoir development planning system. - View Dependent Claims (19, 20, 21, 22, 23)
-
Specification