Pay zone prediction
AssigneeSchlumberger Technology Corporation
InventorsHerwig Fruehbauer, Georg Zangl, Michael Stundner
Life of this patent
Estimating the term…
Abstract
Implementations of pay zone prediction are described. More particularly, apparatus and techniques described herein allow a user to predict pay zones in wells. By accurately predicting pay zones, the user can perforate an existing well at predefined well depths to access hydrocarbon bearing strata while avoiding other undesirable strata (such as water bearing strata). For example, in one possible implementation, well data and syntactic data from a first set of one or more existing wells can be used to create one or more syntactic models. These syntactic models can then be used with water cut and well data from the one or more existing wells to create a pay zone prediction model which can be used with wells outside of the first set of existing wells.
Claims20
A method comprising, using at least one processor:
- accessing well data, including well test data, from a first set of one or more existing wells;
- accessing syntactic data associated with the well data, wherein the syntactic data includes volume of clay material data, total volume data, shale volume data, porosity data or permeability data;
- training one or more syntactic models from the well data and syntactic data, wherein the one or more syntactic models includes a volume of clay material relative to total volume (VCL) model, an effective porosity model or a permeability model;
- training a pay zone predictor model using the well test data, the syntactic data, and at least some of the well data, wherein the pay zone predictor model is configured to predict at least one pay zone in at least one well, and wherein the well test data used to create the pay zone predictor model includes water cut data.
The method of claim wherein the creating a pay zone predictor model further includes training the pay zone predictor model on the syntactic models, the well test data and the at least some of the well data using one or more of:
- correlation methods;
- clustering methods.
A non-transitory computer-readable medium having a set of computer-readable instructions residing thereon that, when executed, perform acts comprising:
- training one or more syntactic models from well data and syntactic data associated with a set of one or more existing wells, wherein the one or more syntactic models includes a volume of clay material relative to total volume (VCL) model, an effective porosity model or a permeability model, and wherein the syntactic data includes volume of clay material data, total volume data, shale volume data, porosity data or permeability data;
- training a pay zone predictor model, configured to predict at least one pay zone in at least one well, using information including:
- at least some of the well data from the set of one or more existing wells;
- well test result data from the set of one or more existing wells;
- and
- outputs of the one or more syntactic models.
A non-transitory computer-readable medium having a set of computer-readable instructions residing thereon that, when executed, perform acts comprising:
- accepting well data from a well of interest into one or more syntactic models trained from well data and syntactic data associated with a group of known wells excluding the well of interest, wherein the one or more syntactic models includes a volume of clay material relative to total volume (VCL) model, an effective porosity model or a permeability model, and wherein the syntactic data includes volume of clay material data, total volume data, shale volume data, porosity data or permeability data;
- outputting syntactic data estimated for the well of interest from the one or more syntactic models;
- accepting, into a pay zone predictor model configured to predict at least one pay zone in at least one well:
- at least some of the well data from the well of interest;
- and
- the syntactic data estimated for the well of interest and output from the one or more syntactic models, and
- outputting from the pay zone predictor model one or more of:
- estimated fluid content values for the well of interest;
- estimated pay zone values for the well of interest.
Dependencies are inferred from the claim text.