Oilfield development early-stage development index prediction method based on random modeling

Abstract

The invention discloses an oil field development early-stage development index prediction method based on random modeling. Indoor relative permeability experiment research is carried out according to oil field coring data; carrying out quantitative characterization on the permeability distribution rule according to an oil field well logging interpretation result; establishing a relationship between porosity and permeability according to a core analysis result; carrying out relative permeability curve normalization; dividing flow pipe nodes according to an actual well position and geological knowledge; setting a basic model, and carrying out index prediction; setting different flow pipe numbers to carry out index prediction, and determining the optimal flow pipe number; setting different node numbers to carry out index prediction, and determining an optimal node number; and performing random assignment, calculating oilfield development indexes under different parameters, and determining and outputting the development indexes through an index error function. According to the method, the heterogeneous property of the reservoir is described by adopting the multi-node multi-flow-pipe network model, the development index is predicted by calculating the flow of each node at different moments, the optimal flow pipe number and the optimal node number are determined through the error function, and the operation speed is high.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas field exploration and development, and in particular relates to an oil field development early stage development index prediction method based on random modeling. Background Art

[0002] The prediction of development indicators is the basis for allocating production and injection, diagnosing the development status of oil fields, and making adjustments to oil field development. Due to the limitations of prediction methods and the heterogeneity of oil fields, actual production dynamics often deviate from early prediction indicators, affecting the normal operation of engineering facilities and restricting the formulation of reasonable technical policies for oil field development. Therefore, it is of great and practical significance to reasonably and accurately predict oil field development indicators in the early stages of oil field development.

[0003] At present, the prediction of development indicators mainly relies on the empirical formula of water drive curve and numerical simulation of oil reservoir. The empirical formula of water drive curve predicts oil field indicators by using the linear relationship between parameters such as liquid volume, oil volume and water content. Since the linear relationship appears in the middle and late stages of oil field, when the water content is high, this method cannot be used in the early stages of oil field, when the water content is low. The numerical simulation method of oil reservoir requires a large amount of data to be prepared and requires the combination of multiple disciplines. In particular, when the knowledge of heterogeneity in the early stage of oil field is insufficient and a large number of models need to be used for probability prediction, the running time is too long and the prediction efficiency is low.

[0004] Therefore, at this stage, it is urgent to establish a method for predicting early development indicators of oil field development, which can easily, quickly, accurately and reasonably predict oil field development indicators and guide the preparation of oil field development planning and development adjustment plans. Summary of the invention

[0005] The problem to be solved by the present invention is to provide a method for predicting early development indicators of oil field development based on random modeling. The method adopts a multi-node multi-flow pipe network model to describe the heterogeneous properties of the reservoir, and predicts the development indicators by calculating the flow rate of each node at different times, so as to solve the technical problems of long running time and low prediction accuracy of existing indicator prediction methods.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for predicting early development indicators of oil field development based on stochastic modeling, comprising the following steps:

[0007] S1: Conduct indoor relative permeability experimental research based on oilfield coring data;

[0008] S2: Quantitative characterization of permeability distribution law is carried out based on the well logging interpretation results of the oil field;

[0009] S3: Establish the relationship between porosity and permeability based on core analysis results;

[0010] S4: normalize the phase permeability curve;

[0011] S5: Divide the flow tube nodes according to the actual well location and geological knowledge;

[0012] S6: Set up the basic model and carry out oilfield indicator prediction;

[0013] S7: setting different numbers of flow tubes to carry out index prediction, and determining the optimal number of flow tubes through the error function of the number of flow tubes;

[0014] S8: Set different numbers of nodes to carry out indicator prediction and determine the optimal number of nodes through the node number error function;

[0015] S9: Calculate the oilfield development index under different parameters by random assignment, and determine the oilfield development index by the index error function;

[0016] S10: Output development indicators.

[0017] Further, the S1 comprises the following steps:

[0018] S11: Select typical coring wells in different layers and well areas of the oil field;

[0019] S12: Classify the cores into different levels of high permeability, medium permeability and low permeability;

[0020] S13: Carry out conventional core tests and special core analysis tests respectively to obtain the porosity, permeability and relative permeability curve parameters of different cores;

[0021] S14: Ensure that the core test covers different reservoir conditions in the oil field and reflects the heterogeneity of the oil field.

[0022] Further, the S2 comprises the following steps:

[0023] S21: Utilize the logging interpretation results of oilfield development wells to calculate the permeability data of the oilfield;

[0024] S22: arranging the permeability data from small to large, and determining a permeability distribution function according to the permeability distribution characteristics;

[0025] S23: Obtain key parameters of the permeability distribution function through curve fitting.

[0026] Further, the S3 comprises the following steps:

[0027] S31: Based on the porosity and permeability parameters obtained from rock tests, establish an intersection chart of porosity and permeability;

[0028] S32: Establish the relationship between porosity and permeability through least squares fitting;

[0029] S33: Determine the distribution law of porosity based on the permeability distribution function determined in S2.

[0030] Furthermore, in S4, normalization is performed based on the phase permeability curves under different conditions obtained in S1 to form a representative curve for the entire oil field.

[0031] Further, the S5 comprises the following steps:

[0032] S51: dividing the number of flow tubes, determining the lowest limit of the number of flow tubes according to the heterogeneity of the reservoir, and setting a unit change number of the number of flow tubes;

[0033] S52: dividing the number of flow tube nodes according to the distance between injection and production wells, determining the lowest limit of the number of nodes, and setting the unit change number of the number of nodes, wherein the number of flow tubes and the division of flow tube nodes are affected by the prediction accuracy and the prediction complexity.

[0034] Further, the S6 comprises the following steps:

[0035] S61: Establishing a flow pipe network model with the minimum number of flow pipes and the minimum number of nodes determined in S5;

[0036] S62: setting physical property parameters according to the permeability distribution function and the relationship between porosity and permeability;

[0037] S63: setting injection and production parameters according to actual oil field production conditions;

[0038] S64: Predict development indicators to 98% water content and calculate the relationship between water content and recovery degree.

[0039] Further, the S7 comprises the following steps:

[0040] S71: Based on the basic model, the number of flow tubes is gradually increased, and the number of flow tubes increased each time is the unit change number of the number of flow tubes. Each time the number is increased, the development index is predicted once, and the relationship between the recovery degree and the water cut increase rate is calculated, and the error compared with the previous time is calculated;

[0041] S72: Set the flow tube quantity error function to:

[0042]

[0043] S73: Set the minimum error E of the number of flow tubes LGmin , gradually increase the number of flow tubes until the calculated E is met LG Less than E LGmin So far, the number of flow tubes calculated at this time is the optimal number of flow tubes.

[0044] Further, the S8 comprises the following steps:

[0045] S81: Based on the optimal number of flow tubes calculated in S7, the number of nodes is used as the lowest limit to determine the basic model again, and the number of nodes is gradually increased. The number of nodes increased each time is the unit change number of the number of nodes. Each time the number of nodes is increased, an indicator prediction is carried out and the recovery degree and water cut increase rate are calculated, and the error with the previous one is calculated;

[0046] S82: Set the node number error function to:

[0047]

[0048] S83: Set the minimum error E of the number of nodes JDmin , gradually increase the number of nodes until the calculated E is satisfied JD Less than E JDmin So far, the number of nodes calculated at this time is the optimal number of nodes.

[0049] Further, the S9 comprises the following steps:

[0050] S91: Based on the optimal number of flow tubes determined by S7 and the optimal number of nodes determined by S8, multiple groups of random parameter schemes are designed. In each scheme, each node of each flow tube is randomly assigned a value. Each time a value is assigned, an index prediction is carried out and the relationship between water content and recovery degree is calculated. After the calculation result is completed, the water content and recovery degree of this calculation result are normalized with those of all previous results, and the error between the normalized curve of water content and recovery degree of this calculation result and that of all previous results is calculated.

[0051] S92: Set the indicator error function to:

[0052]

[0053] S93: Set the minimum error E of the indicator ZBmin , the number of stepwise solutions until the calculated E is satisfied ZB Less than E ZBmin So far, the prediction results calculated at this time are reasonably accurate prediction results.

[0054] The advantages and positive effects of the present invention are:

[0055] The present invention adopts a multi-node multi-flow pipe network model to describe the heterogeneous properties of the reservoir, predicts the development index by calculating the flow of each node at different times, and determines the optimal number of flow pipes and the optimal number of nodes through an error function. The calculation speed is fast, and the calculated oil field development index is more reasonable and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention.

[0057] Figure 2 It is a relationship diagram between water content and recovery degree of the basic model of an embodiment of the present invention.

[0058] Figure 3 1 is a graph showing the relationship between water content and recovery degree under different flow tube numbers in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0061] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0062] The embodiments of the present invention are further described below in conjunction with the accompanying drawings:

[0063] like Figure 1 As shown, a method for predicting early development indicators of oil field development based on stochastic modeling includes the following steps:

[0064] S1: Conduct indoor relative permeability experimental research based on oilfield coring data. Specifically, S1 includes the following steps:

[0065] S11: Select typical coring wells for different layers and well areas in the oil field.

[0066] S12: The core samples are classified into different levels: high permeability, medium permeability and low permeability.

[0067] S13: Carry out conventional core tests and special core analysis tests respectively to obtain the porosity, permeability and relative permeability curve parameters of different cores.

[0068] S14: Ensure that the core test covers different reservoir conditions in the oil field and reflects the heterogeneity of the oil field.

[0069] S2: Based on the oilfield logging interpretation results, quantitative characterization of the permeability distribution law is carried out. Specifically, S2 includes the following steps:

[0070] S21: Utilize the logging interpretation results of oilfield development wells to compile statistical data on the permeability of the oilfield.

[0071] S22: Arrange the permeability data from small to large, and determine the permeability distribution function according to the permeability distribution characteristics.

[0072] S23: Obtain key parameters of the permeability distribution function through curve fitting.

[0073] S3: Establish the relationship between porosity and permeability based on the core analysis results. Specifically, S3 includes the following steps:

[0074] S31: Based on the porosity and permeability parameters obtained from rock tests, an intersection chart of porosity and permeability is established.

[0075] S32: Establish the relationship between porosity and permeability through least squares fitting.

[0076] S33: Determine the distribution law of porosity based on the permeability distribution function determined in S2.

[0077] S4: Normalize the phase permeability curve. Specifically, in S4, normalization is performed based on the phase permeability curves under different conditions obtained in S1 to form a representative curve for the entire oil field. The better the coverage of the original phase permeability curve, the stronger the representativeness of the normalized phase permeability, and the higher the prediction accuracy.

[0078] S5: Divide the flow tube nodes according to the actual well location and geological knowledge. Specifically, S5 includes the following steps:

[0079] S51: Divide the number of flow tubes, determine the lowest limit of the number of flow tubes according to the heterogeneity degree of the reservoir, and set the unit change number of the number of flow tubes.

[0080] S52: dividing the number of flow tube nodes according to the distance between injection and production wells, determining the lowest limit of the number of nodes, and setting the unit change number of the number of nodes, wherein the number of flow tubes and the division of flow tube nodes are affected by the prediction accuracy and the prediction complexity. The more the number of flow tubes and flow tube nodes, the higher the calculation accuracy and the higher the calculation complexity; the fewer the number of flow tubes and flow tube nodes, the lower the calculation accuracy and the lower the calculation complexity.

[0081] S6: Set up the basic model and carry out oilfield index prediction. Specifically, S6 includes the following steps:

[0082] S61: Establish a flow pipe network model with the minimum number of flow pipes and the minimum number of nodes determined in S5.

[0083] S62: Setting physical property parameters according to the permeability distribution function and the relationship between porosity and permeability.

[0084] S63: Set injection and production parameters according to actual oil field production conditions.

[0085] S64: Predict development indicators to 98% water content and calculate the relationship between water content and recovery degree.

[0086] S7: Set different numbers of flow tubes to carry out index prediction, and determine the optimal number of flow tubes through the flow tube number error function. Specifically, S7 includes the following steps:

[0087] S71: Based on the basic model, the number of flow tubes is gradually increased. The number of flow tubes increased each time is the unit change number of the number of flow tubes. Each time it is increased, the development index is predicted once, and the relationship between the degree of recovery and the water cut increase rate is calculated, and the error compared with the previous time is calculated.

[0088] S72: Set the flow tube quantity error function to:

[0089]

[0090] S73: Set the minimum error E of the number of flow tubes LGmin , gradually increase the number of flow tubes until the calculated E is met LG Less than E LGmin So far, the number of flow tubes calculated at this time is the optimal number of flow tubes.

[0091] S8: Set different numbers of nodes to carry out indicator prediction, and determine the optimal number of nodes through the node number error function. Specifically, S8 includes the following steps:

[0092] S81: Based on the optimal number of flow tubes calculated in S7, the average values ​​of porosity and permeability, the number of nodes is used as the lowest limit to re-determine the basic model, and gradually increase the number of nodes. The number of nodes increased each time is the unit change number of the number of nodes. Each time an indicator prediction is carried out and the recovery degree and water cut increase rate are calculated, and the error compared with the previous time is calculated.

[0093] S82: Set the node number error function to:

[0094]

[0095] S83: Set the minimum error E of the number of nodesJDmin , gradually increase the number of nodes until the calculated E is satisfied JD Less than E JDmin So far, the number of nodes calculated at this time is the optimal number of nodes.

[0096] S9: Calculate the oilfield development index under different parameters by random assignment, and determine the oilfield development index by the index error function. Specifically, S9 includes the following steps:

[0097] S91: Based on the optimal number of flow tubes determined by S7 and the optimal number of nodes determined by S8, multiple sets of random parameter schemes are designed. In each scheme, each node of each flow tube is randomly assigned a value. Each time a value is assigned, an indicator prediction is carried out and the relationship between water content and recovery degree is calculated. After completion, the water content and recovery degree of this calculation result are first normalized with those of all previous results, and the error between the relationship between this water content and recovery degree and the normalized curve of water content and recovery degree of all previous results is calculated.

[0098] S92: Set the indicator error function to:

[0099]

[0100] S93: Set the minimum error E of the indicator ZBmin , the number of stepwise solutions until the calculated E is satisfied ZB Less than E ZBmin So far, the prediction results calculated at this time are reasonably accurate prediction results.

[0101] S10: Output development indicators. Specifically, according to the error judgment result of S9, output the latest development indicators, including daily oil production, daily water production, daily liquid production, annual oil production, annual water production, annual liquid production and other indicators.

[0102] The following takes Bohai B oilfield as an example to specifically illustrate the present invention:

[0103] S1: Bohai B Oilfield, which was put into production in 2019. The main production strata are the Dongying Formation and Shahejie Formation of the Paleogene. Cores were taken from exploration wells 1 and 5, and 5 core samples from different parts were selected. The permeabilities were 37mD, 89mD, 170mD, 583mD and 1062mD, respectively. Five relative permeability curves were measured.

[0104] S2: There are 29 development wells in Bohai B Oilfield, and 2240 logging interpretation sample points of the target layer were obtained, with a maximum value of 1300mD and a minimum value of 30mD. The permeability distribution characteristics conform to the log-normal distribution. Through fitting, the average value is 512 and the deviation coefficient is 0.608.

[0105] S3: Through least squares fitting, porosity and permeability conform to a linear relationship, and the relationship is as follows:

[0106] y=0.00276x+24.32.

[0108] S4: The five measured phase permeabilities are normalized to obtain a phase permeability curve representing the entire oil field. After normalization, the bound water saturation is 0.289, the residual oil saturation is 0.258, and the water phase permeability under residual oil is 0.489.

[0109] S5: According to the actual situation of the oil field, the minimum limit of the number of flow tubes is set to 100, the unit change number of the number of flow tubes is 20, the minimum limit of the number of nodes is set to 20, and the unit change number of the number of nodes is set to 5.

[0110] S6: The basic model is set according to the actual situation of the oil field. The number of flow tubes is 100, and the number of nodes of each flow tube is 20. According to the permeability distribution function of the oil field, 100 groups of random permeability values ​​with a mean of 512mD and a deviation coefficient of 0.608 are randomly generated. The porosity parameters are generated according to the relationship between porosity and permeability. Each flow tube is assigned a permeability value and a corresponding porosity value. The output is 163 cubic meters / day, the injection volume is 187 cubic meters / day, and the injection-production balance is achieved. The water cut is predicted to be 98%, as shown in Table 1, and the relationship between water cut and recovery degree is calculated, as shown in Figure 2 shown.

[0111] S7: Figure 3 As shown in the figure, based on the basic model, 20 flow tubes are added each time, and other parameters remain unchanged. Each time the number of flow tubes is increased, the development index is predicted, the oil field accuracy requirements are estimated, and the minimum error E of the number of flow tubes is set. LGmin When the number of flow tubes is 220, the calculated E LG is 4.66%, which is less than E LGmin , and determined that the optimal number of flow tubes is 220.

[0112] Table 1 Prediction results of the basic model indicators of Bohai X Oilfield

[0113]

[0114]

[0115] S8: Based on the 220 flow tubes and 20 nodes determined in S7, 5 nodes are added each time, and other parameters remain unchanged. Each time the number is increased, the development index is predicted, the oil field accuracy requirements are estimated, and the minimum error E of the number of nodes is set. JDmin When the number of nodes is 90, the calculated E JD is 2.86%, which is less than E JDmin , and determined that the optimal number of nodes is 90.

[0116] S9: Based on the 220 flow tubes and 90 nodes determined in S8, 19,800 permeability values ​​that conform to the logarithmic normal distribution are randomly generated according to the permeability distribution function, and the corresponding porosity value is formed according to the relationship between porosity and permeability. Each node of each flow tube is randomly assigned a value. Each time a value is assigned, an indicator prediction is carried out and the relationship between water content and recovery degree is calculated. The error between the relationship between water content and recovery degree this time and the normalized curve of water content and recovery degree of all previous results is calculated; the minimum error E of the indicator is set ZBmin is 5%. If the E ZB Less than E ZBmin The calculation is stopped, otherwise the relationship between the water content and the recovery degree calculated this time is normalized with the relationship between the water content and the recovery degree of all previous results. ZB The value is 4.47%, and the calculation is completed.

[0117] S10: According to the error judgment result of S9, the last development index is output, including daily oil production, daily water production, daily liquid production, annual oil production, annual water production, annual liquid production and other indicators, as shown in Table 2.

[0118] Table 2 Prediction results of the final model index of Bohai X Oilfield

[0119]

[0120]

[0121] The advantages and positive effects of the present invention are:

[0122] The present invention adopts a multi-node multi-flow pipe network model to describe the heterogeneous properties of the reservoir, predicts the development index by calculating the flow of each node at different times, and determines the optimal number of flow pipes and the optimal number of nodes through an error function. The calculation speed is fast, and the calculated oil field development index is more reasonable and reliable.

[0123] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for predicting early development indicators of oil field development based on stochastic modeling, characterized by: The following steps are included: S1: Conduct indoor relative permeability experimental research based on oilfield coring data; S2: Quantitative characterization of permeability distribution law is carried out based on the well logging interpretation results of the oil field; S3: Establish the relationship between porosity and permeability based on core analysis results; S4: normalize the phase permeability curve; S5: Divide the flow tube nodes according to the actual well location and geological knowledge; S6: Set up the basic model and carry out oilfield indicator prediction; S7: setting different numbers of flow tubes to carry out index prediction, and determining the optimal number of flow tubes through the error function of the number of flow tubes; S8: Set different numbers of nodes to carry out indicator prediction and determine the optimal number of nodes through the node number error function; S9: Calculate the oilfield development index under different parameters by random assignment, and determine the oilfield development index by the index error function; S10: Output development indicators.

2. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1, characterized in that: The S1 comprises the following steps, S11: Select typical coring wells in different layers and well areas of the oil field; S12: Classify the cores into different levels of high permeability, medium permeability and low permeability; S13: Carry out conventional core tests and special core analysis tests respectively to obtain the porosity, permeability and relative permeability curve parameters of different cores; S14: Ensure that the core test covers different reservoir conditions in the oil field and reflects the heterogeneity of the oil field.

3. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: The S2 comprises the following steps, S21: Utilize the logging interpretation results of oilfield development wells to calculate the permeability data of the oilfield; S22: arranging the permeability data from small to large, and determining a permeability distribution function according to the permeability distribution characteristics; S23: Obtain key parameters of the permeability distribution function through curve fitting.

4. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: The S3 comprises the following steps, S31: Based on the porosity and permeability parameters obtained from rock tests, establish an intersection chart of porosity and permeability; S32: Establish the relationship between porosity and permeability through least squares fitting; S33: Determine the distribution law of porosity based on the permeability distribution function determined in S2.

5. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: In S4, normalization is performed based on the phase permeability curves under different conditions obtained in S1 to form a representative curve for the entire oil field.

6. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: The S5 comprises the following steps: S51: dividing the number of flow tubes, determining the lowest limit of the number of flow tubes according to the heterogeneity of the reservoir, and setting a unit change number of the number of flow tubes; S52: dividing the number of flow tube nodes according to the distance between injection and production wells, determining the lowest limit of the number of nodes, and setting the unit change number of the number of nodes, wherein the number of flow tubes and the division of flow tube nodes are affected by the prediction accuracy and the prediction complexity.

7. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: The S6 comprises the following steps, S61: Establishing a flow pipe network model with the minimum number of flow pipes and the minimum number of nodes determined in S5; S62: setting physical property parameters according to the permeability distribution function and the relationship between porosity and permeability; S63: setting injection and production parameters according to actual oil field production conditions; S64: Predict development indicators to 98% water content and calculate the relationship between water content and recovery degree.

8. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: The S7 comprises the following steps, S71: Based on the basic model, the number of flow tubes is gradually increased, and the number of flow tubes increased each time is the unit change number of the number of flow tubes. Each time the number is increased, the development index is predicted once, and the relationship between the recovery degree and the water cut increase rate is calculated, and the error compared with the previous time is calculated; S72: Set the flow tube quantity error function to: S73: Set the minimum error E of the number of flow tubes LGmin , gradually increase the number of flow tubes until the calculated E is met LG Less than E LGmin So far, the number of flow tubes calculated at this time is the optimal number of flow tubes.

9. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: The S8 comprises the following steps, S81: Based on the optimal number of flow tubes calculated in S7, the number of nodes is used as the lowest limit to determine the basic model again, and the number of nodes is gradually increased. The number of nodes increased each time is the unit change number of the number of nodes. Each time the number of nodes is increased, an indicator prediction is carried out and the recovery degree and water cut increase rate are calculated, and the error with the previous one is calculated; S82: Set the node number error function to: S83: Set the minimum error E of the number of nodes JDmin , gradually increase the number of nodes until the calculated E is satisfied JD Less than E JDmin So far, the number of nodes calculated at this time is the optimal number of nodes.

10. The method for predicting early development indicators of oil field development based on stochastic modeling according to claim 1 or 2, characterized in that: The S9 comprises the following steps: S91: Based on the optimal number of flow tubes determined by S7 and the optimal number of nodes determined by S8, multiple groups of random parameter schemes are designed. In each scheme, each node of each flow tube is randomly assigned a value. Each time a value is assigned, an index prediction is carried out and the relationship between water content and recovery degree is calculated. After the calculation result is completed, the water content and recovery degree of this calculation result are normalized with those of all previous results, and the error between the normalized curve of water content and recovery degree of this calculation result and that of all previous results is calculated. S92: Set the indicator error function to: S93: Set the minimum error E of the indicator ZBmin , the number of stepwise solutions until the calculated E is satisfied ZB Less than E ZBmin So far, the prediction results calculated at this time are reasonably accurate prediction results.

Images