CN115526017A - Method and device for determining remaining oil saturation of oil reservoir, electronic equipment and medium - Google Patents
Method and device for determining remaining oil saturation of oil reservoir, electronic equipment and medium Download PDFInfo
- Publication number
- CN115526017A CN115526017A CN202110702557.XA CN202110702557A CN115526017A CN 115526017 A CN115526017 A CN 115526017A CN 202110702557 A CN202110702557 A CN 202110702557A CN 115526017 A CN115526017 A CN 115526017A
- Authority
- CN
- China
- Prior art keywords
- candidate
- feature
- target
- oil saturation
- well
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The embodiment of the application discloses a method and a device for determining the saturation of residual oil in an oil reservoir, electronic equipment and a medium, wherein the method comprises the following steps: determining a first candidate feature corresponding to the first candidate field data and a feature importance value of a second candidate feature corresponding to the second candidate field data according to the first candidate field data of the target well and the second candidate field data of the adjacent well; determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value; performing model training by adopting target field data corresponding to the target characteristics to obtain a residual oil saturation prediction model; and predicting the residual oil saturation of the distributed well in the oil reservoir based on a residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed well. The scheme solves the problem of low accuracy and high prediction difficulty in the existing residual oil saturation determination process, so that the residual oil saturation in the oil reservoir can be accurately predicted according to partial single-well data.
Description
Technical Field
The embodiment of the application relates to the technical field of oil exploration, in particular to a method and a device for determining the saturation of remaining oil in an oil reservoir, electronic equipment and a medium.
Background
The residual oil saturation distribution is the basis of oil reservoir adjustment scheme compilation, can effectively guide well position deployment, injection and production adjustment and the like, and directly influences the development effect of the oil reservoir. The residual oil saturation has many influencing factors including oil deposit physical properties, crude oil characteristics, injection and production characteristics, formation pressure and the like, and the prediction difficulty is high.
At present, in the scheme for determining the saturation of the residual oil, the schemes for measuring in an indoor experiment, including a normal-pressure dry distillation method, a distillation extraction method, a chromatography method and the like, are small in measured quantity, cannot represent the whole residual oil distribution of an oil reservoir, and have limited significance for field guidance. The single-well saturation monitoring data interpolation scheme has the advantages of high cost, generally fewer monitoring data points and larger error of a prediction result. The numerical simulation scheme relates to equivalent or multi-parameter cooperative operation and is difficult to accurately describe the oil-gas seepage process, so that the accuracy of the prediction result of the scheme is low, the time consumption is long, and the requirements of quick production building and efficient development of an oil reservoir are difficult to meet.
Disclosure of Invention
The embodiment of the application provides a method and a device for determining the residual oil saturation of an oil reservoir, electronic equipment and a medium, so as to improve the accuracy of the residual oil saturation of the oil reservoir.
In one embodiment, the present application provides a reservoir remaining oil saturation determination method, including:
determining a feature importance value of a first candidate feature corresponding to first candidate field data and a feature importance value of a second candidate feature corresponding to second candidate field data according to the first candidate field data of a target well and the second candidate field data of an adjacent well of the target well;
determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature;
performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of a target well;
and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
In another embodiment, the present application further provides a reservoir remaining oil saturation determination apparatus, including:
the characteristic importance value determining module is used for determining a characteristic importance value of a first candidate characteristic corresponding to first candidate field data and a characteristic importance value of a second candidate characteristic corresponding to second candidate field data according to the first candidate field data of a target well and the second candidate field data of an adjacent well of the target well;
a target feature determination module, configured to determine a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature;
the model determining module is used for performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well;
and the prediction module is used for predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
In another embodiment, an embodiment of the present application further provides an electronic device, including: one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a method for reservoir remaining oil saturation determination as described in any of the embodiments herein.
In one embodiment, the present application further provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the reservoir remaining oil saturation determination method according to any one of the embodiments of the present application.
In the embodiment of the application, a feature importance value of a first candidate feature corresponding to first candidate field data and a feature importance value of a second candidate feature corresponding to second candidate field data are determined according to the first candidate field data of a target well and the second candidate field data of a neighboring well of the target well; and determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature, so that the target feature which has a large influence on the residual oil saturation of the target well is accurately selected, and the accuracy of determining the residual oil saturation is improved. Performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well; and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir, so that the residual oil saturation of the distributed wells in the oil reservoir is accurately predicted by combining the model, and the residual oil saturation distribution in the whole range in the oil reservoir is accurately determined.
Drawings
FIG. 1 is a flow chart of a reservoir remaining oil saturation determination method provided in an embodiment of the present application;
FIG. 2 is a schematic diagram of a target well and an adjacent well distribution provided by an embodiment of the present application;
FIG. 3 is a flow chart of a reservoir remaining oil saturation determination method provided in another embodiment of the present application;
FIG. 4 is a schematic diagram of feature importance value ranking provided in another embodiment of the present application;
FIG. 5 is a comparison of predicted results provided by another embodiment of the present application;
FIG. 6 is a flow chart of a reservoir remaining oil saturation determination method provided in yet another embodiment of the present application;
FIG. 7 is a schematic illustration of a reservoir remaining oil saturation profile provided by a further embodiment of the present application;
FIG. 8 is a schematic diagram illustrating a remaining oil saturation determining apparatus according to an embodiment of the present disclosure;
fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be further noted that, for the convenience of description, only some of the structures related to the present application are shown in the drawings, not all of the structures.
Fig. 1 is a flowchart of a method for determining remaining oil saturation of an oil reservoir according to an embodiment of the present disclosure. The method for determining the residual oil saturation of the oil reservoir can be suitable for the situation of determining the residual oil saturation of the oil reservoir. Typically, the method and the device are suitable for predicting the residual oil saturation of the distribution well in the oil reservoir based on the trained residual oil saturation model, and further determining the residual oil saturation distribution condition in the oil reservoir. The method can be specifically executed by a residual oil saturation determining device, the device can be realized by software and/or hardware, and the device can be integrated in electronic equipment capable of realizing the residual oil saturation determining method. Referring to fig. 1, the method of the embodiment of the present application specifically includes:
s110, according to the first candidate field data of the target well and the second candidate field data of the adjacent well of the target well, determining a feature importance value of a first candidate feature corresponding to the first candidate field data and a feature importance value of a second candidate feature corresponding to the second candidate field data.
The target well is a single well arranged in the oil deposit, the type of the single well is not limited, for example, the single well can comprise a geological exploration well, a pre-exploration well, a detailed exploration well, an inspection data well, a production well, an injection well, an adjustment well and the like, and the oil deposit refers to the basic accumulation of oil with the same pressure system in a single trap. If only oil is collected in a trap, it is called a reservoir. When a reservoir contains several oil-bearing sand layers, it is called a multi-layer reservoir. The target well may be a predetermined portion of a single well in the reservoir, as shown in FIG. 2, where X1 is the target well. The number and the position of the target wells are not specifically limited, and the target wells can be selected according to actual conditions. For example, the target well may be selected according to the number of monitoring data obtained by monitoring data of a single well, and the monitoring data may be the remaining oil saturation of the single well. For example, if the number of remaining oil saturations of a single well is monitored to be greater than a preset number threshold, the single well is taken as the target well. The preset number threshold value can be selected according to actual conditions. And the adjacent well is a single well which is closest to the target well at different angles around the target well. As shown in FIG. 2, X3, X4, X5, and X6 are neighbors of the target well X1. The first candidate field data of the target well may be residual oil saturation monitoring data corresponding to the first candidate feature, and may include static data and dynamic data of the target well, and the like, the first candidate feature corresponding to the static data of the target well may include porosity, permeability, formation thickness, perforation horizon, and the like, and the first candidate feature corresponding to the dynamic data of the target well may include cumulative oil production, cumulative water production, cumulative gas production, bottom hole pressure, well opening time rate, and the like. The second candidate field data of the adjacent well may be residual oil saturation monitoring data corresponding to the second candidate feature, and may include static data and dynamic data of the adjacent well, and the like, the second candidate feature corresponding to the static data of the adjacent well may include porosity, permeability, formation thickness, perforation layer, and the like, and the second candidate feature corresponding to the dynamic data of the adjacent well may include cumulative oil production, cumulative water production, cumulative gas production, cumulative water injection, cumulative gas injection, and the like.
For example, the first candidate feature and the second candidate feature may not all have a large influence on the remaining oil saturation of the target well, and therefore, in the embodiment of the present application, the first candidate feature and the second candidate feature are first screened, so as to select the feature having a large influence on the remaining oil saturation of the target well, thereby improving the accuracy of determining the remaining oil saturation. Specifically, the feature importance value of the first candidate feature and the feature importance value of the second candidate feature may be determined according to the first candidate field data and the second candidate field data, so as to screen the first candidate feature and the second candidate feature according to the feature importance values.
And S120, determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature.
For example, the feature importance value may reflect a degree of influence of the feature on the remaining oil saturation of the target well, and a target feature having a greater influence on the remaining oil saturation of the target well may be selected from the first candidate feature and the second candidate feature according to the feature importance value. Specifically, the selection may be performed according to the size of the feature importance value. And if the feature importance value is larger, the influence of the representation feature on the residual oil saturation of the target well is larger, selecting a first candidate feature and a second candidate feature of which the feature importance values are larger than a preset value as the target features, and if the feature importance value is smaller, the influence of the representation feature on the residual oil saturation of the target well is larger, selecting the first candidate feature and the second candidate feature of which the feature importance values are smaller than the preset value as the target features.
And S130, performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well.
Illustratively, the target field data is data corresponding to a target feature selected from the first candidate field data and the second candidate field data. Wherein, the adopted model can be a deep neural network model. Part of data in the target field data can be selected as training input data, the corresponding residual oil saturation is selected as training output data, model training is carried out, and part of data is used as prediction input data and is compared with the prediction output data, so that the accuracy of the trained model is evaluated. The ratio of the number of training data and prediction data may be determined according to actual conditions. For example, the ratio of the training data to the prediction data is 8:2.
S140, predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
The distributed wells can be all single wells distributed in the oil reservoir, and can also be partial single wells distributed in the oil reservoir. And predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation model, so as to determine the distribution condition of the residual oil saturation of the distributed wells in the oil reservoir. And determining the residual oil saturation in the whole oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir, and determining the distribution condition of the residual oil saturation in the whole oil reservoir range. For example, the remaining oil saturation of the distributed wells in the oil reservoir may be interpolated to obtain the remaining oil saturation at each point in the entire oil reservoir, and determine the remaining oil saturation distribution in the entire oil reservoir.
In the embodiment of the application, a feature importance value of a first candidate feature corresponding to first candidate field data and a feature importance value of a second candidate feature corresponding to second candidate field data are determined according to the first candidate field data of a target well and the second candidate field data of an adjacent well of the target well; and determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature, so that the target feature which has a large influence on the residual oil saturation of the target well is accurately selected, and the accuracy of determining the residual oil saturation is improved. Performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well; and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir, so that the residual oil saturation of the distributed wells in the oil reservoir is accurately predicted by combining the model, and the residual oil saturation distribution in the whole range in the oil reservoir is accurately determined.
Fig. 3 is a flowchart of a reservoir remaining oil saturation determination method according to another embodiment of the present disclosure. In the embodiment of the application, for further optimization of the above embodiment, S110 is refined to S210, S120 is refined to S220, and S130 is refined to S230-S240. Details which are not described in detail in the examples of the present application are described in the above examples. Referring to fig. 3, a method for determining remaining oil saturation of a reservoir according to an embodiment of the present disclosure may include:
s210, determining a feature importance value of a first candidate feature and a feature importance value of a second candidate feature according to first candidate field data of a target well and second candidate field data of a neighboring well of the target well based on a machine learning algorithm.
Illustratively, each first candidate field data of the target well is respectively used as input data of a machine learning algorithm to obtain a feature importance value of a first candidate feature, and each second candidate field data of the adjacent well is respectively used as input data of the machine learning algorithm to obtain a feature importance value of a second candidate feature.
In the embodiment of the application, the machine learning algorithm is a random forest algorithm, the first candidate characteristics comprise target well accumulated oil, target well accumulated gas, target well accumulated water, target well bottom flowing pressure, target well porosity, target well permeability and target well opening time rate, and the second candidate characteristics comprise adjacent well accumulated gas injection, adjacent well accumulated water injection, adjacent well accumulated oil, adjacent well accumulated water, adjacent well bottom flowing pressure, adjacent well porosity, adjacent well permeability and adjacent well opening time rate; correspondingly, determining a feature importance value of the first candidate feature and a feature importance value of the second candidate feature based on a random forest algorithm, and selecting the first candidate feature and the second candidate feature with the feature importance values larger than a preset value to obtain target features including target well accumulated oil, adjacent well accumulated water, target well accumulated gas, adjacent well accumulated gas injection, adjacent well accumulated oil and adjacent well accumulated gas.
Illustratively, first candidate field data corresponding to a first candidate characteristic target well accumulated produced oil, a target well accumulated produced gas, a target well accumulated produced water, a target well bottom flow pressure, a target well porosity, a target well permeability and a target well opening time rate are respectively used as input data of a random forest algorithm to obtain characteristic importance values of the first candidate characteristic target well accumulated produced oil, the target well accumulated produced gas, the target well accumulated produced water, the target well bottom flow pressure, the target well porosity, the target well permeability and the target well opening time rate, and second candidate characteristic adjacent well accumulated gas injection, adjacent well accumulated water injection, adjacent well accumulated oil, adjacent well accumulated water, adjacent well flow pressure, adjacent well porosity, adjacent well permeability and adjacent well opening time are used as input data of the random forest algorithm to obtain second candidate characteristic adjacent well accumulated gas injection, adjacent well accumulated water injection, adjacent well accumulated oil, adjacent well accumulated water production water, adjacent well flow pressure, adjacent well permeability and adjacent well opening time rate, and obtain second candidate characteristic importance values of the adjacent well opening time and the second candidate characteristic adjacent well accumulated gas injection, the adjacent well accumulated water production, the adjacent well permeability and the adjacent well opening time importance values are used as input data of the second candidate field data of the second candidate characteristic importance values are shown in a second candidate characteristic feature map, wherein the second candidate characteristic feature is shown in a second candidate.
S220, selecting the first candidate feature and the second candidate feature with the feature importance value larger than a preset value as target features.
Wherein, the preset value can be set according to the actual situation. The first candidate feature and the second candidate feature may be ranked with respect to the feature importance value, and a target feature having a large influence on the remaining oil saturation may be selected according to a ranking result. For example, when the preset value is set to 0.05, and sorted according to the feature importance value in fig. 4, the target well accumulated oil, the adjacent well accumulated water, the target well accumulated gas, the adjacent well accumulated oil, and the adjacent well accumulated gas with the feature importance value greater than 0.05 may be selected as the target feature.
And S230, training the model by adopting training input data and residual oil saturation monitoring data corresponding to different times in the target field data to obtain training output data.
For example, the target field data may be data acquired at different times, and training input data in the target field data acquired at different times is used as training input data, so that model training is performed from different time dimensions, and the model has temporal universality. Training input data of different time in the target field data are used as training input data, residual oil saturation monitoring data of a target well or an adjacent well corresponding to the training input data are used as label data, and the model is trained. And inputting the training input data into the model to obtain training output data. For example, the target field data includes C n ＝[C X,1,n,t ,C X,2,n,t ,C X,3,n,t ,C X,4,n,t ,C X,5,n,t ,C X,6,n,t ]Wherein, the lower corner marks 1 to 6 respectively represent six target characteristic target wells for accumulating oil, accumulating water in the adjacent wells, accumulating gas in the target wells, accumulating gas in the adjacent wells, accumulating oil in the adjacent wells and accumulating gas in the adjacent wells, n is the sample number, t is the time dimension, and X is the well number. The output data set is Y n ＝S o,X,n In which S is o And the saturation degree of the remaining oil is shown, X is the target well number, and n is the number of samples. The training input data may include: c n1 ＝[C X,1,n1,t ,C X,2,n1,t ,C X,3,n1,t ,C X,4,n1,t ,C X,5,n1,t ,C X,6,n1,t ]The training output data includes: y is n1 ＝S o,X,n1 。
And S240, iteratively updating the model based on a back propagation algorithm according to an error value between training output data corresponding to the training input data and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well.
Illustratively, according to training output data corresponding to training input data and an error value between residual oil saturation monitoring data of a target well or an adjacent well corresponding to the training input data, the model is iteratively updated based on a back propagation algorithm to optimize the model, so that the accuracy of the model is improved until the error value between the training output data corresponding to the training input data and the residual oil saturation monitoring data meets an error value requirement, for example, is smaller than a preset error value threshold, the accuracy of the model is determined to meet the requirement, and the model obtained through current training is used as a residual oil saturation prediction model of the target well.
And S250, predicting the residual oil saturation prediction model by adopting prediction input data corresponding to different time in target field data to obtain prediction output data.
For example, after the residual oil saturation prediction model is obtained, the residual oil saturation prediction model may be predicted, for example, target field data except training input data is selected as prediction input data and input into the residual oil saturation prediction model to obtain prediction output data. The prediction input data may include: c n2 ＝[C X,1,n2,t ,C X,2,n2,t ,C X,3,n2,t ,C X,4,n2,t ,C X,5,n2,t ,C X,6,n2,t ]The prediction output data may comprise Y n2 ＝S o,X,n2 。
And S260, evaluating the accuracy of the residual oil saturation prediction model according to an error value between prediction output data corresponding to the prediction input data and residual oil saturation monitoring data.
Illustratively, the accuracy of the residual oil saturation prediction model is evaluated according to the prediction output data corresponding to the preset input data and the error value between the residual oil saturation monitoring data of the target well or the adjacent well corresponding to the prediction input data. For example, it is determined whether the error value satisfies the error value requirement, such as being less than a predetermined error value threshold, if the error value satisfies the error value requirement. And if the error value does not meet the requirement of the error value, determining that the residual oil saturation prediction model has low accuracy and is not suitable for residual oil saturation prediction. Illustratively, the residual oil saturation prediction model and the mathematical model calculation method in the embodiment of the present application are respectively used for predicting ten single wells P1 to P10, and a comparison graph of prediction results is shown in fig. 5, where a first result corresponding to each single well is a residual oil saturation prediction result obtained by the method provided in the embodiment of the present application, a second result is an actual value of the residual oil saturation of the single well, and a third result is a residual oil saturation prediction result obtained by the mathematical model calculation. As can be seen from the graph, the accuracy of the result obtained by the residual oil saturation prediction model obtained by the method in the embodiment of the application is high.
And S270, predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
According to the scheme in the embodiment of the application, the characteristic importance value of the first candidate feature and the characteristic importance value of the second candidate feature are determined based on a machine learning algorithm according to the first candidate field data of the target well and the second candidate field data of the adjacent wells of the target well, the first candidate feature and the second candidate feature with the characteristic importance values larger than the preset value are selected to serve as the target features, so that the target features with large influence on the residual oil saturation of the target well are selected, model training is performed by adopting the target field data corresponding to the target features, and therefore the accuracy of the model for predicting the residual oil saturation is improved. And the accuracy of the residual oil saturation model is improved through the training and prediction of the model.
Fig. 6 is a flowchart of a specific implementation of a method for determining remaining oil saturation of an oil reservoir according to another embodiment of the present disclosure. Details which are not described in detail in the examples of the present application are described in the above examples. Referring to fig. 6, a specific implementation procedure of the method for determining the remaining oil saturation of the oil reservoir provided by the embodiment of the present application may include:
s310, according to the first candidate field data of the target well and the second candidate field data of the adjacent well of the target well, determining a feature importance value of a first candidate feature corresponding to the first candidate field data and a feature importance value of a second candidate feature corresponding to the second candidate field data.
S320, determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature.
And S330, performing model training by adopting the target field data corresponding to the target characteristics and the residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well.
And S340, taking the field data of the distributed wells corresponding to the target characteristics of the distributed wells in the oil reservoir as input data, and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model to obtain the residual oil saturation of the distributed wells in the oil reservoir.
Illustratively, the field data of the distributed wells corresponding to the target characteristics of the distributed wells in the oil reservoir are input into the residual oil saturation prediction model, so that the residual oil saturation of the distributed wells in the oil reservoir in the whole range is predicted, and the residual oil saturation of the distributed wells in the oil reservoir in the whole range is determined. The field data of the distributed wells can also be data acquired at different times so as to represent data of different time dimensions.
And S350, interpolating the residual oil saturation of the distributed wells in the oil reservoir based on an interpolation algorithm, and determining the residual oil saturation in the oil reservoir.
Illustratively, interpolation is performed based on the determined residual oil saturation of the distributed wells in the oil reservoir, so that residual oil saturation corresponding to more points in the oil reservoir is obtained, and further the distribution condition of the residual oil saturation in the oil reservoir is obtained. The interpolation algorithm may be a kriging interpolation algorithm. Illustratively, a final residual oil saturation profile in the reservoir may be obtained as shown in FIG. 7.
According to the scheme in the embodiment of the application, the field data of the distributed wells corresponding to the target characteristics of the distributed wells in the oil reservoir are used as input data, the residual oil saturation of the distributed wells in the oil reservoir is predicted based on the residual oil saturation prediction model, the residual oil saturation of the distributed wells in the oil reservoir is obtained, interpolation is carried out on the residual oil saturation of the distributed wells in the oil reservoir based on an interpolation algorithm, the residual oil saturation in the oil reservoir is determined, the residual oil saturation distribution in the whole oil reservoir is predicted accurately, and the accuracy of prediction of the residual oil saturation at each point in the oil reservoir is improved.
Fig. 8 is a schematic structural diagram of a reservoir remaining oil saturation determining apparatus according to an embodiment of the present disclosure. The device is applicable to the situation of determining the remaining oil saturation in an oil reservoir. Typically, the method and the device are suitable for predicting the residual oil saturation of the distribution well in the oil reservoir based on the trained residual oil saturation model, and further determining the residual oil saturation distribution condition in the oil reservoir. The apparatus may be implemented by software and/or hardware, and the apparatus may be integrated in an electronic device. Referring to fig. 8, the apparatus specifically includes:
a feature importance value determining module 410, configured to determine a feature importance value of a first candidate feature corresponding to first candidate field data and a feature importance value of a second candidate feature corresponding to second candidate field data according to the first candidate field data of a target well and the second candidate field data of a neighboring well of the target well;
a target feature determining module 420, configured to determine a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature;
the model determining module 430 is configured to perform model training by using target field data corresponding to the target features and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well;
and the prediction module 440 is configured to predict the remaining oil saturation of the distributed wells in the oil reservoir based on the remaining oil saturation prediction model, and determine the remaining oil saturation in the oil reservoir according to the remaining oil saturation of the distributed wells in the oil reservoir.
In this embodiment of the application, the feature importance value determining module 410 is specifically configured to:
determining a feature importance value of a first candidate feature and a feature importance value of a second candidate feature according to first candidate field data of a target well and second candidate field data of an adjacent well of the target well on the basis of a machine learning algorithm;
accordingly, the target feature determination module 420 is specifically configured to:
and selecting the first candidate feature and the second candidate feature with the feature importance value larger than the preset value as the target feature.
In the embodiment of the application, the machine learning algorithm is a random forest algorithm, the first candidate characteristics comprise target well accumulated oil, target well accumulated gas, target well accumulated water, target well bottom flowing pressure, target well porosity, target well permeability and target well opening time rate, and the second candidate characteristics comprise adjacent well accumulated gas injection, adjacent well accumulated water injection, adjacent well accumulated oil, adjacent well accumulated water, adjacent well bottom flowing pressure, adjacent well porosity, adjacent well permeability and adjacent well opening time rate;
correspondingly, the characteristic importance value of the first candidate characteristic and the characteristic importance value of the second candidate characteristic are determined based on a random forest algorithm, and the target characteristics obtained by selecting the first candidate characteristic and the second candidate characteristic with the characteristic importance values larger than the preset values comprise target well accumulated oil, adjacent well accumulated water, target well accumulated gas, adjacent well accumulated gas injection, adjacent well accumulated oil and adjacent well accumulated gas.
In this embodiment, the model determining module 430 includes:
the training output data determining unit is used for training the model by adopting training input data and residual oil saturation monitoring data corresponding to different times in target field data to obtain training output data;
and the iteration updating unit is used for performing iteration updating on the model based on a back propagation algorithm according to an error value between training output data corresponding to the training input data and the residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well.
In an embodiment of the present application, the apparatus further includes:
the prediction output data determining module is used for predicting the residual oil saturation prediction model by adopting prediction input data corresponding to different times in target field data to obtain prediction output data;
and the evaluation module is used for evaluating the accuracy of the residual oil saturation prediction model according to an error value between prediction output data corresponding to the prediction input data and residual oil saturation monitoring data.
In an embodiment of the present application, the prediction module 440 includes:
and the distribution well prediction unit is used for predicting the residual oil saturation of the distribution wells in the oil reservoir based on the residual oil saturation prediction model by taking the field data of the distribution wells corresponding to the target characteristics of the distribution wells in the oil reservoir as input data to obtain the residual oil saturation of the distribution wells in the oil reservoir.
In an embodiment of the present application, the prediction module 440 includes:
and the interpolation unit is used for interpolating the residual oil saturation of the distributed wells in the oil reservoir based on an interpolation algorithm to determine the residual oil saturation in the oil reservoir.
The device for determining the residual oil saturation of the oil reservoir, provided by the embodiment of the application, can execute the method for determining the residual oil saturation of the oil reservoir, provided by any embodiment of the application, and has corresponding functional modules and beneficial effects of the execution method.
Fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application. FIG. 9 illustrates a block diagram of an exemplary electronic device 512 suitable for use in implementing embodiments of the present application. The electronic device 512 shown in fig. 9 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 9, the electronic device 512 may include: one or more processors 516; the memory 528 is used for storing one or more programs, when the one or more programs are executed by the one or more processors 516, the one or more processors 516 implement the method for determining remaining oil saturation provided by the embodiment of the present application, including:
determining a feature importance value of a first candidate feature corresponding to first candidate field data and a feature importance value of a second candidate feature corresponding to second candidate field data according to the first candidate field data of a target well and the second candidate field data of an adjacent well of the target well;
determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature;
performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well;
and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
Components of electronic device 512 may include, but are not limited to: one or more processors 516, a memory 528, and a bus 518 that connects the various device components, including the memory 528 and the processors 516.
The electronic device 512 typically includes a variety of computer device-readable storage media. These storage media may be any available storage media that can be accessed by electronic device 512 and includes both volatile and nonvolatile storage media, removable and non-removable storage media.
The memory 528 may include computer device readable storage media in the form of volatile memory, such as Random Access Memory (RAM) 530 and/or cache memory 532. The electronic device 512 may further include other removable/non-removable, volatile/nonvolatile computer device storage media. By way of example only, storage system 534 may be used to read from and write to non-removable, nonvolatile magnetic storage media (not shown in FIG. 9 and commonly referred to as a "hard drive"). Although not shown in FIG. 9, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical storage medium) may be provided. In such cases, each drive may be connected to bus 518 through one or more data storage media interfaces. Memory 528 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the application.
A program/utility 540 having a set (at least one) of program modules 542 may be stored, for example, in memory 528, such program modules 542 including, but not limited to, an operating device, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may include an implementation of a network environment. Program modules 542 generally perform the functions and/or methods of the embodiments described herein.
The electronic device 512 may also communicate with one or more external devices 514 and/or a display 524, etc., and may also communicate with one or more devices that enable a user to interact with the electronic device 512, and/or with any devices (e.g., network cards, modems, etc.) that enable the electronic device 512 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 522. Also, the electronic device 512 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) via the network adapter 520. As shown in FIG. 9, the network adapter 520 communicates with the other modules of the electronic device 512 via the bus 518. It should be appreciated that although not shown in FIG. 9, other hardware and/or software modules may be used in conjunction with the electronic device 512, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID devices, tape drives, and data backup storage devices, among others.
The processor 516 executes various functional applications and data processing by executing at least one of other programs of the programs stored in the memory 528, for example, to implement a method for determining remaining oil saturation of a reservoir provided in the embodiments of the present application.
One embodiment of the present application provides a storage medium containing computer-executable instructions which, when executed by a computer processor, perform a reservoir remaining oil saturation determination method, comprising:
determining a feature importance value of a first candidate feature corresponding to first candidate field data and a feature importance value of a second candidate feature corresponding to second candidate field data according to the first candidate field data of a target well and the second candidate field data of an adjacent well of the target well;
determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature;
performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well;
and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
The computer storage media of embodiments of the present application may take any combination of one or more computer-readable storage media. The computer readable storage medium may be a computer readable signal storage medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor device, apparatus, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In embodiments of the present application, a computer readable storage medium may be any tangible storage medium that can contain, or store a program for use by or in connection with an instruction execution apparatus, device, or apparatus.
A computer readable signal storage medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal storage medium may be any computer readable storage medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution apparatus, device, or apparatus.
Program code embodied on a computer readable storage medium may be transmitted using any appropriate storage medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or device. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present application and the technical principles employed. Those skilled in the art will appreciate that the present application is not limited to the particular embodiments described herein, but is capable of many obvious modifications, rearrangements and substitutions without departing from the scope of the application. Therefore, although the present application has been described in more detail with reference to the above embodiments, the present application is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present application, and the scope of the present application is determined by the scope of the appended claims.
Claims (10)
1. A method for determining remaining oil saturation of a reservoir, the method comprising:
determining a feature importance value of a first candidate feature corresponding to first candidate field data and a feature importance value of a second candidate feature corresponding to second candidate field data according to the first candidate field data of a target well and the second candidate field data of an adjacent well of the target well;
determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature;
performing model training by adopting target field data corresponding to the target characteristics and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well;
and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model, and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
2. The method of claim 1, wherein determining a feature importance value for a first candidate feature corresponding to a first candidate field data of a target well and a feature importance value for a second candidate feature corresponding to a second candidate field data based on the first candidate field data of the target well and the second candidate field data of a neighboring well to the target well comprises:
determining a feature importance value of a first candidate feature and a feature importance value of a second candidate feature according to first candidate field data of a target well and second candidate field data of a neighboring well of the target well on the basis of a machine learning algorithm;
accordingly, determining a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature comprises:
and selecting the first candidate feature and the second candidate feature with the feature importance value larger than a preset value as target features.
3. The method of claim 2, wherein the machine learning algorithm is a random forest algorithm, the first candidate features include target well accumulated oil, target well accumulated gas, target well accumulated water, target well bottom stream pressure, target well porosity, target well permeability, and target well kick-in time, and the second candidate features include adjacent well accumulated gas injection, adjacent well accumulated water, adjacent well accumulated oil, adjacent well accumulated water, adjacent well bottom stream pressure, adjacent well porosity, adjacent well permeability, and adjacent well kick-in time;
correspondingly, determining a feature importance value of the first candidate feature and a feature importance value of the second candidate feature based on a random forest algorithm, and selecting the first candidate feature and the second candidate feature with the feature importance values larger than a preset value to obtain target features including target well accumulated oil, adjacent well accumulated water, target well accumulated gas, adjacent well accumulated gas injection, adjacent well accumulated oil and adjacent well accumulated gas.
4. The method of any one of claims 1-3, wherein performing model training using target field data corresponding to the target characteristic and residual oil saturation monitoring data to obtain a residual oil saturation prediction model of a target well comprises:
training the model by adopting training input data corresponding to different times in target field data and residual oil saturation monitoring data to obtain training output data;
and iteratively updating the model based on a back propagation algorithm according to an error value between training output data corresponding to the training input data and the residual oil saturation monitoring data to obtain a residual oil saturation prediction model of the target well.
5. The method of claim 4, wherein after obtaining the prediction model of remaining oil saturation of the target well, the method further comprises:
predicting the residual oil saturation prediction model by adopting prediction input data corresponding to different times in target field data to obtain prediction output data;
and evaluating the accuracy of the residual oil saturation prediction model according to an error value between prediction output data corresponding to the prediction input data and residual oil saturation monitoring data.
6. The method of claim 1, wherein predicting remaining oil saturation of a well pattern in a reservoir based on the remaining oil saturation prediction model comprises:
and taking the field data of the distributed wells corresponding to the target characteristics of the distributed wells in the oil reservoir as input data, and predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model to obtain the residual oil saturation of the distributed wells in the oil reservoir.
7. The method of claim 1, wherein determining remaining oil saturation in the reservoir from remaining oil saturation of a well pattern in the reservoir comprises:
and interpolating the residual oil saturation of the distributed wells in the oil reservoir based on an interpolation algorithm to determine the residual oil saturation in the oil reservoir.
8. An apparatus for determining a remaining oil saturation of a reservoir, the apparatus comprising:
the characteristic importance value determining module is used for determining a characteristic importance value of a first candidate characteristic corresponding to first candidate field data and a characteristic importance value of a second candidate characteristic corresponding to second candidate field data according to the first candidate field data of a target well and the second candidate field data of an adjacent well of the target well;
a target feature determination module, configured to determine a target feature from the first candidate feature and the second candidate feature according to the feature importance value of the first candidate feature and the feature importance value of the second candidate feature;
the model determining module is used for performing model training by adopting target field data and residual oil saturation monitoring data corresponding to the target characteristics to obtain a residual oil saturation prediction model of a target well;
and the prediction module is used for predicting the residual oil saturation of the distributed wells in the oil reservoir based on the residual oil saturation prediction model and determining the residual oil saturation in the oil reservoir according to the residual oil saturation of the distributed wells in the oil reservoir.
9. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the reservoir remaining oil saturation determination method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the method for reservoir remaining oil saturation determination according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110702557.XA CN115526017A (en) | 2021-06-24 | 2021-06-24 | Method and device for determining remaining oil saturation of oil reservoir, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110702557.XA CN115526017A (en) | 2021-06-24 | 2021-06-24 | Method and device for determining remaining oil saturation of oil reservoir, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115526017A true CN115526017A (en) | 2022-12-27 |
Family
ID=84695001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110702557.XA Pending CN115526017A (en) | 2021-06-24 | 2021-06-24 | Method and device for determining remaining oil saturation of oil reservoir, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115526017A (en) |
-
2021
- 2021-06-24 CN CN202110702557.XA patent/CN115526017A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103177180B (en) | Method for screening modeling samples of prediction model | |
| CN109115987B (en) | Rock physical model-based fluid factor evaluation method and device | |
| NO20181025A1 (en) | Machine learning for production prediction | |
| US8234073B2 (en) | System and method for estimating geological architecture of a geologic volume | |
| AU2007211291A1 (en) | Methods, systems, and computer-readable media for fast updating of oil and gas field production models with physical and proxy simulators | |
| CA2806532A1 (en) | Systems and methods for predicting well performance | |
| US11137514B2 (en) | Method for determining a drilling plan for a plurality of new wells in a reservoir | |
| CN113792936A (en) | Intelligent lithology while drilling identification method, system, equipment and storage medium | |
| CN113065279A (en) | Method, device, equipment and storage medium for predicting total organic carbon content | |
| JP2022123078A (en) | Method for predicting oil accumulation depth lower limit of deep and ultra-depth marine carbonate reservoir | |
| US10908308B1 (en) | System and method for building reservoir property models | |
| CN112578441A (en) | Reservoir thickness prediction analysis method, computer equipment and storage medium | |
| CN114723155A (en) | Transverse wave curve prediction method, device, computing equipment and storage medium | |
| CN114609667A (en) | Geophysical oil gas prediction method, equipment and storage medium | |
| Bisseil et al. | Reservoir history matching using the method of gradients on a workstation | |
| CN111580179B (en) | Method, device and system for determining organic carbon content | |
| CN112112618A (en) | Fracturing well selection and layer selection method and system based on flow process | |
| CN115526017A (en) | Method and device for determining remaining oil saturation of oil reservoir, electronic equipment and medium | |
| CN115239001A (en) | Method and device for determining bottom hole flow pressure, electronic equipment and medium | |
| US11578595B2 (en) | Systems and methods for selecting and performing gas deliverability tests | |
| CN112561171B (en) | Landslide prediction method, device, equipment and storage medium | |
| US11320553B2 (en) | System and method for subsurface structural interpretation | |
| CN114563823A (en) | Reservoir lithology and physical property parameter combined Bayes inversion method and system | |
| CN114021700A (en) | Permeability calculation method and device based on petrophysical constraint neural network | |
| RU2530324C2 (en) | Method for determining position of marker depth coordinates when constructing geological model of deposit |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |