Back to results
US 9,043,188 B2Granted
System and method for forecasting production from a hydrocarbon reservoir
AssigneeChevron U.S.A. Inc.
InventorsBurak Yeten, Joseph Martin Hovadik, Pierre Muron
Life of this patent
Estimating the term…
Abstract
A system and method is taught to substantially automate forecasting for a hydrocarbon producing reservoir through integration of modeling module workflows. A control management module automatically generates static and dynamic offspring models, with static and dynamic modeling software, until a performance objective associated with the forecasting of the reservoir is satisfied. The performance objective can include an experimental design table to determine a sensitivity of a particular parameter or can be directed towards reservoir optimization, i.e., ultimate hydrocarbon recovery, net present value, reservoir percentage yield, reservoir fluid flow rate, or history matching error.
Claims19
A computer-implemented method to automate forecasting of a hydrocarbon producing reservoir, the method comprising:
- (a) providing a performance objective associated with forecasting of a hydrocarbon producing reservoir via a user interface in communication with a reservoir modeling program, wherein the performance objective is an uncertainty assessment, and the uncertainty assessment comprises an experimental design table including a parameter selected from the group consisting of permeability distributions, porosity distributions, fluid contacts, fluid saturations, geobody connectivity, pore volume, fault transmissibility, sedimentological properties, reservoir fluid flow rates, reservoir pressure characteristics, reservoir temperature characteristics, upscaling properties, and training images;
- (b) automatically generating a gridded static model representative of at least a portion of the hydrocarbon producing reservoir via a static modeling module of the reservoir modeling program using static reservoir data and a discrete set of static parameters;
- (c) automatically generating a dynamic model representative of at least a portion of the hydrocarbon producing reservoir via a dynamic modeling module of the reservoir modeling program using the static model and a discrete set of dynamic parameters;
- (d) evaluating the dynamic model against the performance objective via a control management module of the reservoir modeling program;
- (e) until the performance objective is satisfied, iteratively combining the static and dynamic models, via the control management module of the reservoir modeling program, to automatically generate an offspring model per iteration, wherein combining the static and dynamic models includes regenerating at least one of the dynamic model or the static model during each iteration;
- and
- (f) outputting a result according to the performance objective associated with the forecasting of the hydrocarbon producing reservoir subsequent to completion of step (e).
4 more elements
A computer-implemented method to automate forecasting of a hydrocarbon producing reservoir, the method comprising:
- (a) modeling a gridded static model representative of at least a portion of a hydrocarbon producing reservoir with a static modeling module, wherein the gridded static model is automatically generated using static reservoir data and a discrete set of static parameters;
- (b) modeling a dynamic model representative of at least a portion of the hydrocarbon producing reservoir with a dynamic modeling module, wherein the dynamic model is automatically generated using the static model and a discrete set of dynamic parameters;
- (c) generating a discrete set of parameters responsive to the static model and the dynamic model with a control management module, wherein the discrete set of parameters is selected from the group consisting of permeability distributions porosity distributions, fluid contacts, fluid saturations, geobody connectivity, pore volume, fault transmissibility, sedimentological properties, reservoir fluid flow rates, reservoir pressure characteristics, reservoir temperature characteristics, upscaling properties, and training images;
- (d) modeling an offspring model responsive to at least one parameter from the discrete set of parameters, the modeling of the offspring model is performed with at least one of the static and dynamic modeling modules, and wherein the modeling of the offspring model includes combining the static and dynamic models to automatically generate the offspring model;
- (e) updating the discrete set of parameters responsive to the offspring model with the control management module;
- (f) repeating steps (d) and (e), including regenerating at least one of the dynamic model or the static model during each iteration, until one of the following occurs selected from the group consisting of the offspring model has been generated a predetermined number of iterations, the offspring model converges to a predetermined variance, and a sensitivity of the at least one parameter has been determined;
- and
- (g) outputting a visual display of a solution associated with forecasting of the hydrocarbon producing reservoir subsequent to completion of step (f), wherein the visual display of the solution comprises a result to an uncertainty assessment.
5 more elements
A system to automate forecasting of a hydrocarbon producing reservoir, the system comprising:
- a computer processor;
- a computer program executable on the computer processor, the computer program comprising:
- a first modeling module to automatically generate a first model representative of at least a portion of the hydrocarbon producing reservoir using first reservoir data and a first discrete set of parameters, wherein the first modeling module is a static modeling model and the first model is a static model;
- a second modeling module to generate a second model representative of at least a portion of the hydrocarbon producing reservoir using the first model and a second discrete set of parameters, wherein the second modeling module is a dynamic modeling model and the second model is a dynamic model;
- and
- a control management module to evaluate the second model against a performance objective associated with the forecasting of the system, and until the performance objective is satisfied, to automatically generate offspring models responsive to the first and second models with at least one of the first modeling module and the second modeling module, wherein generating the offspring models includes regenerating at least one of the second model or the first model, and wherein the control management module outputs a result according to the performance objective subsequent to the performance objective being satisfied,
- wherein the performance objective is an uncertainty assessment, and the uncertainty assessment comprises an experimental design table including a parameter selected from the group consisting of permeability distributions, porosity distributions, fluid contacts, fluid saturations, geobody connectivity, pore volume, fault transmissibility, sedimentological properties, reservoir fluid flow rates, reservoir pressure characteristics, reservoir temperature characteristics, upscaling properties, and training images.
4 more elements
A system to automate forecasting of a hydrocarbon producing reservoir, the system comprising:
- a computer processor;
- a software program executable on the computer processor, the software program comprising:
- (a) a first modeling module to automatically generate a first model representative of at least a portion of the hydrocarbon producing reservoir using first reservoir data and a first discrete set of parameters, wherein the first modeling module is a static modeling module and the first model is a static model;
- (b) a second modeling module to automatically generate a second model representative of at least a portion of the hydrocarbon producing reservoir using the first model and a second discrete set of parameters, wherein the second modeling module is a dynamic modeling module and the second model is a dynamic model;
- and
- (c) a control management module that initiates a plurality of operations for the software program responsive to a performance objective, the control management module including at least one of a model retriever to retrieve the first and second models over a communications network, a model generator to automatically generate offspring models responsive to the first and second models with the first and second modeling modules, wherein generating the offspring models includes regenerating at least one of the second model or the first model, a proxy generator to generate a proxy function responsive to the first and second models, a model filter that prohibits the model generator from generating the offspring models responsive to the proxy function, and a model evaluator to evaluate the offspring models and generate a forecast for the hydrocarbon producing reservoir,
- (d) wherein the performance objective is an uncertainty assessment, and the uncertainty assessment comprises an experimental design table including a parameter selected from the group consisting of permeability distributions, porosity distributions, fluid contacts, fluid saturations, geobody connectivity, pore volume, fault transmissibility, sedimentological properties, reservoir fluid flow rates, reservoir pressure characteristics, reservoir temperature characteristics, upscaling properties, and training images;
- a user control interface for inputting information into the system, the information including the performance objective;
- a database to store the performance objective, first and second models, and offspring models;
- and
- a user reporting means in communication with the control management module over the communications network to output the forecast for the hydrocarbon producing reservoir.
8 more elements
A non-transitory processor readable medium containing computer readable software instructions used to automate forecasting of a hydrocarbon producing reservoir, the software comprising:
- (a) a first modeling module to automatically generate a first model representative of at least a portion of the hydrocarbon producing reservoir using first reservoir data and a first discrete set of parameters, wherein the first modeling module is a static modeling module and the first model is a static model;
- (b) a second modeling module to automatically generate a second model representative of at least a portion of the hydrocarbon producing reservoir using the first model and a second discrete set of parameters, wherein the second modeling module is a dynamic modeling module and the second model is a dynamic model;
- and
- (c) a control management module that initiates a plurality of operations for the software program responsive to a performance objective, the control management module including at least one of a model retriever to retrieve the first and second models over a communications network, a model generator to automatically generate offspring models responsive to the first and second models with the first and second modeling modules, wherein generating the offspring models includes regenerating at least one of the second model or the first model, a proxy generator to generate a proxy function responsive to the first and second models, a model filter that prohibits the model generator from generating the offspring models responsive to the proxy function, and a model evaluator to evaluate the offspring models and generate a forecast for the hydrocarbon producing reservoir,
- (d) wherein the performance objective is an uncertainty assessment, and the uncertainty assessment comprises an experimental design table including a parameter selected from the group consisting of permeability distributions, porosity distributions, fluid contacts, fluid saturations, geobody connectivity, pore volume, fault transmissibility, sedimentological properties, reservoir fluid flow rates, reservoir pressure characteristics, reservoir temperature characteristics, upscaling properties, and training images.
2 more elements
Dependencies are inferred from the claim text.