Data-driven method for selecting replacement wells

Abstract

The present invention provides a data-driven alternative well screening method, comprising: Step 1: Collecting multi-source data on alternative wells and their surrounding oil and water wells in a target block; Step 2: Determining indicators characterizing the production performance of the alternative wells by analyzing their production characteristics; Step 3: Analyzing factors influencing the production performance of the alternative wells; Step 4: Determining sensitivity factors affecting the production performance of the alternative wells to form a sample library of alternative well screening models; Step 5: Determining corresponding comprehensive evaluation values of the production performance limits of the alternative wells; Step 6: Establishing an optimal alternative well screening model; and Step 7: Using the established alternative well screening model, screening potential alternative wells suitable for alternative production from low-yield, low-efficiency wells and shut-in oil wells in the target block. This data-driven alternative well screening method is highly practical and operationally feasible, and has far-reaching and significant significance for reducing oilfield costs, increasing efficiency, and extending the economic life of ultra-high water-cut reservoirs.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas field development, and in particular to a data-driven replacement well screening method. Background Art

[0002] Currently, domestic oilfields are managing remote, scattered wells, low-permeability wells with insufficient fluid supply, and long-idled wells. Due to the high operating costs of conventional oil production systems, in the current era of low oil prices, salvage recovery, also known as replacement oil production, is becoming a viable production method. This method, also known as shunting oil production, does not require three-stage extraction equipment, surface process flow, electrical circuits, transformers, or pumping maintenance. This significantly reduces production costs and effectively addresses the difficulty of recovering liquids from remote, scattered, and low-permeability wells.

[0003] Many wells in my country's onshore, long-term water-drive oilfields have reached the ultra-high water-cut development stage. When pumped using electric pumps, they produce high liquid volumes, extremely high water cuts, and very low oil yields. This places significant pressure on surface gathering and transportation systems and wastewater reinjection, while also resulting in poor economic returns. Under these circumstances, some oilfields have explored intermittent well drilling and replacement production. While oil production may be lower than pumping, this method offers significant economic benefits due to its low liquid production and low production costs. Therefore, research on replacement well development methods for ultra-high water-cut oilfields is of great significance. Especially in the current low oil price environment, the implementation of appropriate replacement production technologies for oilfields entering the ultra-high water-cut stage has become a key breakthrough in stabilizing production and increasing efficiency.

[0004] However, due to insufficient understanding of the reservoir's subsurface conditions, it's difficult to establish a reasonable standard and basis for converting conventional mechanical production wells to replacement wells, requiring only field experience to determine the appropriate criteria. Due to a lack of effective theoretical guidance, some replacement wells have failed to achieve adequate results and have been converted to pumping production, further limiting the development and widespread adoption of this technology.

[0005] Chinese patent application number CN202011431138.9 discloses a method for screening and evaluating favorable oil and gas accumulation zones, including the following steps: S1) conducting a regional data survey; S2) conducting various studies based on the regional data survey results, including: source rock evaluation and its hydrocarbon generation and expulsion history, reservoir and trap characteristics, source-reservoir-caprock assemblages, oil and gas migration and accumulation systems, oil and gas injection phases, and differential accumulation patterns of heavy and light oil; S3) identifying the primary controlling factors for oil and gas enrichment and accumulation based on these studies; S4) studying oil and gas distribution patterns and establishing an oil and gas accumulation model based on the analysis of the primary controlling factors for oil and gas enrichment and accumulation; and S5) identifying favorable oil and gas accumulation zones, identifying favorable accumulation zones, and evaluating favorable targets. This invention can clarify oil and gas accumulation conditions in an oilfield, evaluate favorable targets, improve drilling success rates, and accelerate capacity development, thereby enhancing the efficiency and economic benefits of oil reservoir exploration and development.

[0006] Chinese patent application number CN201310752321.2 describes a method for developing oil fields in the ultra-high water-cut period. The method exhaustively enumerates all possible stratum recombination schemes. A first-step screening optimization method statistically identifies recombination schemes whose pseudo-seepage resistance differentials fall within a set range. A second-step screening optimization method screens for single-control reserve limits, ensuring that the remaining reserves within a set of strata are economically viable. A third-step screening optimization method numerically simulates and predicts the schemes obtained from the second-step screening optimization, screening them based on technical indicators and increases in recovery rates. Finally, a fourth-step screening optimization method screens schemes based on economic indicators to determine the final stratum recombination scheme. This method is simple and easy to operate, effectively resolving the significant reduction in adaptability of conventional stratum combinations during ultra-high water-cut periods and achieving significant economic benefits.

[0007] Chinese patent application number CN201810646440.2 describes a technical method for well selection for stratified oil production. This method includes: 1) classifying oil and gas reservoir types; 2) determining target reservoir potential and stratified oil production type λ; 3) using evaluation parameters and stratified oil production type for quantitative scoring; 4) optimizing well locations; and 5) dynamic tracking and evaluation. This method fully utilizes all prior dynamic and static oilfield data. Based on reservoir geology, well logging, and reservoir engineering methods, it evaluates different reservoir types encountered vertically in a single well, identifies the potential of each reservoir type, determines stratified oil production evaluation parameters and quantitative scoring, and optimizes well locations and stratified oil production layers, thereby increasing the oilfield's crude oil production and economic benefits.

[0008] The above existing technologies are all significantly different from the present invention and fail to solve the technical problem we want to solve. Therefore, we have invented a new data-driven alternative well screening method. Summary of the Invention

[0009] The purpose of the present invention is to provide a data-driven replacement well screening method that can correctly and reasonably guide the conversion of mechanical oil production wells into replacement wells.

[0010] The object of the present invention can be achieved by the following technical measures: a data-driven alternative well screening method, the data-driven alternative well screening method comprising:

[0011] Step 1: Collect multi-source data on the replacement wells and surrounding oil and water wells in the target block;

[0012] Step 2: Determine the characterization index of the production effect of the replacement well by analyzing the production characteristics of the replacement well;

[0013] Step 3: Analyze the factors affecting the production effect of the replacement well;

[0014] Step 4: Determine the sensitivity factors affecting the production performance of the replacement wells and form a sample library for the replacement well screening model;

[0015] Step 5: Determine the corresponding comprehensive evaluation value of the production effect limit of the replacement well;

[0016] Step 6: Establish the best alternative well screening model;

[0017] Step 7: Use the established replacement well screening model to screen out potential replacement wells suitable for replacement production from low-yield and low-efficiency wells and shut-in oil wells in the target block.

[0018] The purpose of the present invention can also be achieved by the following technical measures:

[0019] In step 1, multi-source data on the replacement wells and surrounding oil and water wells in the target block are collected, including geological characteristics, historical production data, wellbore structure data, and fluid physical property data.

[0020] In step 1, the collected data include: effective thickness, effective porosity, effective permeability, casing pressure, pump hanging depth, dynamic liquid level depth, central depth of the replacement layer, monthly water injection volume of surrounding water wells, monthly oil production and monthly water production of surrounding oil wells, monthly oil production and monthly water production of the central replacement well, well location coordinates, tubing depth and outer diameter, and outer diameter and wall thickness of the oil layer casing.

[0021] In step 2, based on the relevant data collected in step 1, by analyzing the production characteristics of the replacement well, the characterization indicators of the replacement well production effect are determined to be the cumulative oil production of the replacement well in the first year of replacement production, the daily liquid level and the average replacement cycle.

[0022] In step 2, the calculation formulas for the replacement well production performance indicators, namely the replacement cumulative oil production, daily liquid level and average replacement cycle in the first year of replacement production, are as follows:

[0023]

[0024] Where N p_year is the cumulative oil production of the replacement in the first year of replacement production, t; L p_year The cumulative liquid production of the replacement in the first year of replacement production, t; q l_perday The daily liquid level in the first year of replacement production is calculated by dividing the cumulative liquid production of the replacement in the first year of replacement production by the number of days in the actual replacement month in a year, t·d -1 ; is the average replacement cycle in the first year of replacement production, d; q oi is the oil production of the i-th replacement in the first year of replacement production, t; q wi is the water production of the i-th replacement in the first year of replacement production, t; n is the total number of replacements in the first year of replacement production, an integer; m is the actual total number of replacement months in the first year of replacement production. Since replacement wells may be shut in some months or the wellhead equipment may be damaged and not replaced during the first year of replacement production, the actual number of replacement months may be less than 12, so the actual number of replacement months is used for calculation, an integer; T i is the interval between the i-th replacement and the i+1-th replacement in the first year of replacement production, d.

[0025] In step 3, the replacement well and its surrounding oil and water wells are regarded as a replacement well production unit. Based on reservoir engineering theory, multi-source data are integrated to analyze the factors affecting the replacement well production effect from six aspects: physical parameter characteristics of the replacement well production unit, pressure flow field characteristics, structural correspondence characteristics, residual potential characteristics, seepage environment characteristics, and injection-production connectivity characteristics.

[0026] In step 3, the factors affecting the replacement effect are analyzed based on the physical property parameter characteristics of the replacement well production unit. The average effective thickness, effective porosity, and effective permeability of the replacement layer are used to characterize the physical property characteristics of the replacement well production unit. The calculation formula is as follows:

[0027]

[0028] Where h j is the effective thickness of the replacement layer of the jth well, m; h ji is the effective thickness of the ith replacement layer in the jth well, m; l is the total number of small production layers in the replacement, an integer; is the average effective thickness of the replacement layer, m; φ j is the effective porosity of the replacement layer of the jth well, decimal; φ jiis the effective porosity of the ith replacement layer in the jth well, decimal; is the average effective porosity of the replacement layer, decimal; K j is the effective permeability of the replacement layer of the jth well, 10 -3 μm 2 ;K ji is the effective permeability of the ith replacement layer in the jth well, 10 -3 μm 2 ; is the average effective permeability of the replacement layer, 10 -3 μm 2 ; m is the number of oil wells around the replacement well before replacement, an integer; n is the number of water wells around the replacement well before replacement, an integer.

[0029] In step 3, the factors affecting the production effect of the replacement well are analyzed from the pressure flow field characteristics of the replacement well production unit, and the pressure flow field characteristics of the replacement well production unit are characterized by the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells and the cumulative injection-production ratio.

[0030] In step 3, the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells is calculated as follows:

[0031]

[0032] Where Δp i is the pressure difference between the central replacement well and the replacement layer of the surrounding well i, MPa; p is the replacement layer pressure of the central replacement well, MPa; p i is the casing pressure of the surrounding well i, MPa; is the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells, MPa.

[0033] The calculation formula of cumulative injection-production ratio is as follows:

[0034]

[0035] Where, IPR is the cumulative injection-production ratio before replacement, decimal; WI j is the cumulative water injection volume of the replacement layer of the jth water well around the central replacement well before replacement, 10 4 m 3 NP oi is the cumulative oil production of the replacement layer of the i-th oil well before replacement, 10 4 t; WP oi is the cumulative water production of the replacement layer of the surrounding i-th oil well before replacement, 10 4 m 3 NP is the cumulative oil production of the replacement layer of the central replacement well before replacement, 10 4t; WP is the cumulative water production of the replacement layer of the central replacement well before replacement, 10 4 m 3 θ oi is the inner angle of the polygon enclosed by the i-th oil well and other oil and water wells around the central replacement well, in degrees; θ wj The internal angle of the polygon enclosed by the j-th water well and other oil and water wells around the central replacement well, degrees; B o is the volume coefficient of formation crude oil, decimal; ρ o is the density of crude oil on the ground, g·cm -3 .

[0036] In step 3, the factors affecting the production effect of the replacement well are analyzed from the residual potential characteristics of the replacement well production unit. The dimensionless cumulative oil and water production of the central replacement well, the dimensionless cumulative oil and water production of the oil wells around the central replacement well, and the dimensionless cumulative water injection of the water wells around the central replacement well are used to characterize the residual potential characteristics of the replacement well production unit.

[0037] In step 3, the dimensionless cumulative oil and water production of the central replacement well are calculated as follows:

[0038]

[0039] Where NP′ is the dimensionless cumulative oil production of the central replacement well before replacement, decimal; WP′ is the dimensionless cumulative water production of the central replacement well before replacement, decimal; A is the polygonal area enclosed by the oil and water wells around the central replacement well, km 2 ; is the average effective thickness of the replacement layer, m; NP is the cumulative oil production of the replacement layer of the central replacement well before replacement, 10 4 t; is the average effective porosity of the replacement layer, decimal; B o is the volume coefficient of formation crude oil, decimal; ρ o is the density of crude oil on the ground, g·cm -3 .

[0040] The dimensionless cumulative oil and water production calculation formulas for the oil wells surrounding the central replacement well are as follows:

[0041]

[0042] Where NP o ′ is the dimensionless cumulative oil production of the oil wells around the central replacement well before replacement, decimal; WP o ′ is the dimensionless cumulative water production of the oil wells around the central replacement well before replacement, decimal; θ oi is the inner angle of the polygon enclosed by the i-th oil well and other oil and water wells around the central replacement well, in degrees; NP oi is the cumulative oil production of the replacement layer of the i-th oil well before replacement, 10 4 t;

[0043] The dimensionless cumulative water injection calculation formula for the wells surrounding the central replacement well is as follows:

[0044]

[0045] Where WI k ' is the dimensionless cumulative water injection of the surrounding wells of the central replacement well before replacement, decimal; WI j is the cumulative water injection volume of the replacement layer of the jth water well around the central replacement well before replacement, 10 4 m 3 θ wj The internal angle of the polygon enclosed by the j-th water well and other oil and water wells around the central replacement well, in degrees.

[0046] In step 3, the factors affecting the production effect of the replacement well are analyzed from the structural correspondence characteristics of the replacement well production unit, and the structural correspondence characteristics of the replacement well production unit are characterized by the average altitude depth difference between the central replacement well and the replacement layer of the surrounding oil and water wells, and the average formation dip of the replacement layer.

[0047] In step 3, the formula for calculating the average depth difference between the central replacement well and the replacement layers of the surrounding oil and water wells is as follows:

[0048]

[0049] Where Δz ij is the difference in altitude between the replacement well and the middle depth of the jth replacement layer in the surrounding i-th well, m; z j is the depth above sea level of the middle of the jth replacement layer in the replacement well, m; z ij is the depth above sea level of the middle of the jth replacement layer of the surrounding i-th well, m; Δz j is the average elevation depth difference between the replacement well and the surrounding oil and water wells in the middle of the jth replacement layer, m; is the average altitude depth difference between the replacement well and the replacement layer of the surrounding oil and water wells, m.

[0050] The calculation formula for the average formation dip angle of the replacement layer is as follows:

[0051]

[0052] Where, d ij is the distance between the replacement well and the jth replacement layer of the surrounding i-th well, m; θ ij is the formation dip angle between the replacement well and the jth replacement layer of the surrounding i-th well, Δz ij There are positive and negative values. When Δz ij When it is positive, the θ calculated by the inverse tangent ij Between 0-90 degrees, when Δz ij When it is negative, the inverse tangent calculated θij Between 90-180 degrees, degrees; θ j is the average dip angle of the jth replacement layer between the replacement well and the surrounding oil and water wells, degrees; is the average formation dip angle of the replacement layer, degrees.

[0053] In step 3, the factors affecting the production effect of the replacement well are analyzed based on the bottom hole seepage environment characteristics of the replacement well, and the bottom hole seepage environment characteristics of the replacement well are characterized by the annular space volume of the replacement well, the effective seepage area of the replacement layer, and the density of the interlayer.

[0054] In step 3, the calculation formula for the annular space volume of the replacement well is as follows:

[0055]

[0056] Where V is the volume of the casing space of the replacement well, m 3 ;h o is the depth of the oil pipe, m; d t is the outer diameter of the oil layer casing, mm; b t Oil layer casing wall thickness, mm; d o is the outer diameter of the oil pipe, mm; h b is the bottom depth of the deepest replacement layer, m.

[0057] The calculation formula for the effective seepage area of the replacement layer is as follows:

[0058]

[0059] Where S is the effective seepage area of the replacement layer, m 2 ;h i is the effective thickness of the ith replacement layer in the replacement well, m.

[0060] The calculation formula for the density of the interlayer in the replacement layer is as follows:

[0061]

[0062] Where ρ is the density of interlayers in the replacement layer, m -1 ; c is the number of interlayers in the replacement layer, an integer; h t is the top depth of the shallowest replacement layer, m.

[0063] In step 3, the factors affecting the production effect of the replacement well are analyzed based on the injection-production connectivity characteristics of the replacement well production unit. Using the time series data of the injection and production volume of oil and water wells, the dynamic time warping algorithm is used to calculate the similarity between the replacement well production volume and the injection volume change curve of the surrounding water wells to evaluate the dynamic connectivity between the injection and production wells; the greater the curve similarity, the better the connectivity.

[0064] In step 3, the comprehensive injection-production connectivity coefficient of the replacement well production unit is used to characterize the injection-production connectivity characteristics of the replacement well production unit. The calculation formula is as follows:

[0065]

[0066] Where, is the comprehensive injection-production connectivity coefficient of the replacement well production unit, decimal; DTW j It is the DTW distance between the replacement well and the jth water well around it, a decimal.

[0067] In step 4, the entropy method and comprehensive scoring method are used to calculate the comprehensive evaluation value of the replacement well production effect based on the replacement well production effect characterization indicators in step 2, and a Pearson correlation analysis is performed to determine the sensitivity influencing factors of the replacement well production effect, thereby forming a sample library of the replacement well screening model.

[0068] In step 4, the objective weighting method, the entropy method, is used to calculate the weights C of the three production effect characterization indicators of the replacement well. Then, the comprehensive scoring method is used to calculate the comprehensive evaluation value of the production effect of the replacement well. The calculation formula is as follows:

[0069]

[0070] Where Z is the comprehensive evaluation value of the production effect of the replacement well, a decimal; W i is the weight of the production effect characterization index of the replacement well i, a decimal; C i is the production effect characterization index value of the replacement well i, decimal;

[0071] Then, the Pearson correlation analysis method is used to calculate the correlation coefficient between the possible influencing factors of the replacement well production effect and the comprehensive evaluation value Z of the production effect, and to determine the sensitive influencing factors of the replacement well production effect; the Pearson correlation coefficient calculation formula is:

[0072]

[0073] Where r is the Pearson correlation coefficient between X and Y, a decimal; x i and y i are the observed values of X and Y, respectively, in decimal form; and are the average values of the observed values of X and Y, respectively, in decimal form; n is the number of samples, in integer form;

[0074] Sensitive factors affecting the comprehensive evaluation value of the replacement well production effect include: the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells Effective seepage area S of the replacement layer, average effective thickness of the replacement layer Average formation dip of the replacement layer Dimensionless cumulative oil production NP of the oil wells around the central replacement well before replacement o ′, cumulative injection-production ratio (IPR) before replacement, dimensionless cumulative water production (WP) of oil wells around the central replacement well before replacement o ′, the final formed replacement well screening model sample library includes the following parameters: S. NP o ′、IPR、WP o ′ and comprehensive evaluation value Z of production effect.

[0075] In step 5, based on the economic limit daily oil level of the replacement well, combined with the relationship between the daily liquid level in the first year of replacement production, the cumulative oil production of the stage replacement and the average replacement cycle in step 2, the corresponding comprehensive evaluation value of the replacement well production effect limit is determined.

[0076] In step 5, the economic limit daily oil level of the replacement well is set. Based on the water cut of the replacement well being basically maintained at 30%, the economic limit daily liquid level is calculated. Then, based on the relationship between the daily liquid level in the first year of replacement production, the cumulative oil production of the stage replacement, and the average replacement cycle, the specific relationship is as follows:

[0077]

[0078] Where a and b are regression coefficients, dimensionless; m and n are regression exponents; N p_year is the cumulative oil production of the replacement in the first year of replacement production, t; q l_perday The daily liquid level in the first year of replacement production is calculated by dividing the cumulative liquid production of the replacement in the first year of replacement production by the number of days in the actual replacement month in a year, t·d -1 ; is the average replacement cycle in the first year of replacement production, d;

[0079] The cumulative oil production of the stage replacement corresponding to the economic limit daily liquid level and the average replacement cycle can be calculated, and then the comprehensive evaluation value of the production effect limit under the set economic limit daily oil level of the replacement well can be calculated according to formula (14).

[0080] In step 6, the sample library obtained in step 4 is divided into a training set and a test set in a ratio of 8:2, and the multivariate linear regression algorithm is used for training. The coefficient of determination R between the predicted value and the actual value of the test set is 2 As an evaluation index of model accuracy, the polynomial characteristic order of the input data is optimized. After obtaining the optimal polynomial characteristic order, combined with the production effect limit evaluation value of the replacement well obtained in step 5, the best replacement well screening model is obtained.

[0081] In step 6, the coefficient of determination R 2 The calculation formula is as follows:

[0082]

[0083] Where R 2 The determination coefficient of the actual value and the predicted value for comprehensive evaluation of production effect, decimal; Z i is the actual value of the comprehensive evaluation of the production effect of the replacement well, a decimal; Z′ is the predicted value of the comprehensive evaluation of the production effect of the replacement well, a decimal; is the average value of the comprehensive evaluation of production effects of all replacement wells in the test set, a decimal; m is the number of test samples.

[0084] In step 7, find all low-yield, low-efficiency and shut-in oil wells in the target block and calculate the parameters required for the replacement well screening model: the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells Effective seepage area S of the replacement layer, average effective thickness of the replacement layer Average formation dip of the replacement layer Dimensionless cumulative oil production NP of the oil wells around the central replacement well before replacement o ′, cumulative injection-production ratio (IPR) before replacement, dimensionless cumulative water production (WP) of oil wells around the central replacement well before replacement o ′, a sample library of replacement wells to be screened is formed, which is substituted into the replacement well screening model. The calculated comprehensive evaluation value of the production effect of the replacement well is compared with the comprehensive evaluation value of the production effect of the replacement well to screen out potential replacement wells suitable for replacement.

[0085] The data-driven replacement well screening method in this invention utilizes data mining and machine learning techniques to reveal the main factors affecting the production performance of replacement wells from a large amount of historical multi-source data on replacement wells. Based on this, a replacement well screening model is established to accurately and rationally guide the conversion of mechanical oil production wells into replacement wells. This invention is highly practical and operationally feasible, and has far-reaching and important significance for reducing costs and increasing efficiency in oil fields and extending the economic life of ultra-high water-cut reservoirs. Compared with existing technologies, this invention can achieve the following technical effects:

[0086] 1) The present invention uses actual data from field replacement wells and, through data mining and machine learning, establishes a screening model for converting conventional mechanical oil production wells into replacement wells. This model has important application value in guiding the conversion of conventional mechanical oil production wells into replacement wells in ultra-high water-cut oil fields.

[0087] 2) In combination with reservoir engineering theory, this paper proposes a calculation method for 16 factors that affect the production effect of replacement wells, uses Pearson correlation analysis to identify 7 sensitive factors, and builds a sample library for replacement well screening models.

[0088] 3) The replacement well screening model constructed in the present invention can calculate the economic limit daily oil levels of different replacement wells according to the changes in actual oil prices, and then determine the corresponding comprehensive evaluation value of the replacement well production effect limit, thereby obtaining a replacement well screening model under different oil prices, which has strong applicability.

[0089] 4) The data required for the replacement well screening model of the present invention can be obtained on site, and the model screening accuracy is high and practical. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 A flowchart of a specific embodiment of the data-driven alternative well screening method of the present invention;

[0091] Figure 2 This is a schematic diagram of a replacement well production unit in a specific embodiment of the present invention;

[0092] Figure 3 Schematic diagram of calculation of the average depth difference above sea level between the central replacement well and the replacement layers of surrounding oil and water wells in a specific embodiment of the present invention;

[0093] Figure 4 Schematic diagram of calculation of average formation dip angle of alternate layers in a specific embodiment of the present invention;

[0094] Figure 5 Schematic diagram of calculation of characteristic parameters of bottom seepage environment of a replacement well in a specific embodiment of the present invention;

[0095] Figure 6 Schematic diagram of the prediction effect of the comprehensive evaluation value of production effect under different polynomial characteristic parameters in a specific embodiment of the present invention;

[0096] Figure 7 This is a distribution map of historical replacement wells and potential replacement wells in a target block in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0097] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0098] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations and / or combinations thereof.

[0099] like Figure 1 As shown, Figure 1 The flowchart of the data-driven alternative well screening method of the present invention is as follows. The data-driven alternative well screening method includes the following steps:

[0100] Step S1: Collect multi-source data of the replacement well and its surrounding oil and water wells in the target block, including geological characteristics, historical production data, wellbore structure data, and fluid physical property data;

[0101] The collected data include: effective thickness, effective porosity, effective permeability, casing pressure, pump hanging depth, dynamic liquid level depth, middle depth of the replacement layer, monthly water injection volume of surrounding water wells, monthly oil production and monthly water production of surrounding oil wells, monthly oil production and monthly water production of the central replacement well, well location coordinates, tubing depth and outer diameter, outer diameter and wall thickness of oil layer casing.

[0102] Step S2: Based on the relevant data collected in step S1, by analyzing the production characteristics of the replacement well, determining the characterizing indicators of the production effect of the replacement well as the cumulative oil production of the replacement well in the first year of replacement production, the daily liquid level and the average replacement cycle;

[0103] The calculation formulas for the replacement well production performance characterization indicators, the cumulative oil production of the replacement well in the first year of replacement production, the daily liquid level and the average replacement cycle are as follows:

[0104]

[0105] Where N p_year is the cumulative oil production of the replacement in the first year of replacement production, t; L p_year The cumulative liquid production of the replacement in the first year of replacement production, t; q l_perday is the daily liquid level in the first year of replacement production (dividing the cumulative liquid production of the replacement in the first year of replacement production by the number of days in the actual replacement month in a year), t·d -1 ; is the average replacement cycle in the first year of replacement production, d; q oi is the oil production of the i-th replacement in the first year of replacement production, t; q wi is the water production of the i-th replacement in the first year of replacement production, t; n is the total number of replacements in the first year of replacement production, an integer; m is the actual total number of replacement months in the first year of replacement production (since replacement wells may be shut in some months or the wellhead equipment may be damaged and not replaced in the first year of replacement production, the actual number of replacement months may be less than 12, so the actual number of replacement months is used for calculation), an integer; T i is the interval between the i-th replacement and the i+1-th replacement in the first year of replacement production, d.

[0106] Step S3: The replacement well and its surrounding oil and water wells are considered as a replacement well production unit. Based on reservoir engineering theory, multi-source data are integrated to analyze the factors affecting the replacement well production performance from six aspects: physical parameter characteristics, pressure and flow field characteristics, structural correspondence characteristics, residual potential characteristics, seepage environment characteristics, and injection-production connectivity characteristics of the replacement well production unit;

[0107] The factors affecting the replacement effect are analyzed from the physical property parameter characteristics of the replacement well production unit. The average effective thickness, effective porosity, and effective permeability of the replacement layer are used to characterize the physical property characteristics of the replacement well production unit. The calculation formula is as follows:

[0108]

[0109] Where h j is the effective thickness of the replacement layer of the jth well, m; h ji is the effective thickness of the ith replacement layer in the jth well, m; l is the total number of small production layers in the replacement, an integer; is the average effective thickness of the replacement layer, m; φ j is the effective porosity of the replacement layer of the jth well, decimal; φ ji is the effective porosity of the ith replacement layer in the jth well, decimal; is the average effective porosity of the replacement layer, decimal; K j is the effective permeability of the replacement layer of the jth well, 10 -3 μm 2 ;K ji is the effective permeability of the ith replacement layer in the jth well, 10 -3 μm 2 ; is the average effective permeability of the replacement layer, 10 -3 μm 2 ; m is the number of oil wells around the replacement well before replacement, an integer; n is the number of water wells around the replacement well before replacement, an integer.

[0110] The factors influencing the production performance of the replacement well are analyzed based on the pressure flow field characteristics of the replacement well production unit. The pressure flow field characteristics of the replacement well production unit are characterized by the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells (before replacement) and the cumulative injection-production ratio (before replacement). The formula for calculating the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells is as follows:

[0111]

[0112] Where Δp i is the pressure difference between the central replacement well and the replacement layer of the surrounding well i, MPa; p is the replacement layer pressure of the central replacement well, MPa; p i is the casing pressure of the surrounding well i, MPa; is the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells, MPa.

[0113] The calculation formula of cumulative injection-production ratio is as follows:

[0114]

[0115] Where, IPR is the cumulative injection-production ratio before replacement, decimal; WI j is the cumulative water injection volume of the replacement layer of the jth water well around the central replacement well before replacement, 10 4 m 3 NP oi is the cumulative oil production of the replacement layer of the i-th oil well before replacement, 10 4 t; WP oi is the cumulative water production of the replacement layer of the surrounding i-th oil well before replacement, 10 4 m 3 NP is the cumulative oil production of the replacement layer of the central replacement well before replacement, 10 4 t; WP is the cumulative water production of the replacement layer of the central replacement well before replacement, 10 4 m 3 θ oi is the inner angle of the polygon enclosed by the i-th oil well and other oil and water wells around the central replacement well, in degrees; θ wj The internal angle of the polygon enclosed by the j-th water well and other oil and water wells around the central replacement well, degrees; B o is the volume coefficient of formation crude oil, decimal; ρ o is the density of crude oil on the ground, g·cm -3 .

[0116] The factors influencing the production performance of the replacement wells were analyzed based on the residual potential characteristics of the replacement well production unit. The dimensionless cumulative oil and water production of the central replacement well (before replacement), the dimensionless cumulative oil and water production of the oil wells surrounding the central replacement well (before replacement), and the dimensionless cumulative water injection of the water wells surrounding the central replacement well (before replacement) were used to characterize the residual potential characteristics of the replacement well production unit. The dimensionless cumulative oil and water production (before replacement) of the central replacement well were calculated as follows:

[0117]

[0118] Where NP′ is the dimensionless cumulative oil production of the central replacement well before replacement, decimal; WP′ is the dimensionless cumulative water production of the central replacement well before replacement, decimal; A is the polygonal area enclosed by the oil and water wells around the central replacement well, km 2 .

[0119] The dimensionless cumulative oil and water production (before replacement) of the oil wells surrounding the central replacement well is calculated as follows:

[0120]

[0121] Where NP o ′ is the dimensionless cumulative oil production of the oil wells around the central replacement well before replacement, decimal; WP o ′ is the dimensionless cumulative water production of the oil wells around the central replacement well before replacement, a decimal.

[0122] The calculation formula for dimensionless cumulative water injection (before replacement) of the wells surrounding the central replacement well is as follows:

[0123]

[0124] Where WI k ′ is the dimensionless cumulative water injection of the wells around the central replacement well before replacement, a decimal.

[0125] The factors influencing the production performance of the replacement wells were analyzed based on the structural correspondence characteristics of the replacement well production unit. The structural correspondence characteristics of the replacement well production unit were characterized by the average depth difference above sea level between the central replacement well and the replacement layers of the surrounding oil and water wells, as well as the average formation dip of the replacement layers. The formula for calculating the average depth difference above sea level between the central replacement well and the replacement layers of the surrounding oil and water wells is as follows:

[0126]

[0127] Where Δz ij is the difference in altitude between the replacement well and the middle depth of the jth replacement layer in the surrounding i-th well, m; z j is the depth above sea level of the middle of the jth replacement layer in the replacement well, m; z ij is the depth above sea level of the middle of the jth replacement layer of the surrounding i-th well, m; Δz j is the average elevation depth difference between the replacement well and the surrounding oil and water wells in the middle of the jth replacement layer, m; is the average altitude depth difference between the replacement well and the replacement layer of the surrounding oil and water wells, m.

[0128] The calculation formula for the average formation dip angle of the replacement layer is as follows:

[0129]

[0130] Where, d ij is the distance between the replacement well and the jth replacement layer of the surrounding i-th well, m; θ ij is the formation dip angle (Δz ij There are positive and negative values. When Δz ij When it is positive, the θ calculated by the inverse tangent ij Between 0-90 degrees, when Δz ij When it is negative, the inverse tangent calculated θ ij between 90 and 180 degrees), degrees; θj is the average dip angle of the jth replacement layer between the replacement well and the surrounding oil and water wells, degrees; is the average formation dip angle of the replacement layer, degrees.

[0131] The factors affecting the production effect of the replacement well are analyzed based on the bottom hole seepage environment characteristics of the replacement well. The bottom hole seepage environment characteristics of the replacement well are characterized by the replacement well annulus space volume, the effective seepage area of the replacement layer, and the density of the interlayer. The calculation formula for the replacement well annulus space volume is as follows:

[0132]

[0133] Where V is the volume of the casing space of the replacement well, m 3 ;h o is the depth of the oil pipe, m; d t is the outer diameter of the oil layer casing, mm; b t Oil layer casing wall thickness, mm; d o is the outer diameter of the oil pipe, mm; h b is the bottom depth of the deepest replacement layer, m.

[0134] The calculation formula for the effective seepage area of the replacement layer is as follows:

[0135]

[0136] Where S is the effective seepage area of the replacement layer, m 2 ;h i is the effective thickness of the ith replacement layer in the replacement well, m.

[0137] The calculation formula for the density of the interlayer in the replacement layer is as follows:

[0138]

[0139] Where ρ is the density of interlayers in the replacement layer, m -1 ; c is the number of interlayers in the replacement layer, an integer; h t is the top depth of the shallowest replacement layer, m.

[0140] The factors influencing the production performance of the replacement wells were analyzed based on the injection-production connectivity characteristics of the replacement well production unit. Using time series data of injection and production fluid volumes from oil and water wells, the Dynamic Time Warping (DTW) algorithm was used to calculate the similarity between the replacement well production volume and the injection volume curves of surrounding water wells, thereby evaluating the dynamic connectivity between the injection and production wells. The greater the curve similarity, the better the connectivity. The comprehensive injection-production connectivity coefficient of the replacement well production unit was used to characterize the injection-production connectivity characteristics of the replacement well production unit. The calculation formula is as follows:

[0141]

[0142] Where, is the comprehensive injection-production connectivity coefficient of the replacement well production unit, decimal; DTW j It is the DTW distance between the replacement well and the jth water well around it, a decimal.

[0143] Step S4: Calculate the comprehensive evaluation value of the replacement well production effect based on the replacement well production effect characterization index in step S2 using the entropy method and the comprehensive scoring method, and perform Pearson correlation analysis to determine the sensitivity influencing factors of the replacement well production effect, thereby forming a sample library of the replacement well screening model;

[0144] The objective weighting method - entropy method is used to calculate the weights C of the three production effect characterization indicators of the replacement well. Then, the comprehensive scoring method is used to calculate the comprehensive evaluation value of the production effect of the replacement well. The calculation formula is as follows:

[0145]

[0146] Where Z is the comprehensive evaluation value of the production effect of the replacement well, a decimal; W i is the weight of the production effect characterization index of the replacement well i, a decimal; C i is the value of the production effect characterization index of the replacement well, a decimal (here only the values of each production effect evaluation index are taken, the larger the value, the better the production effect, regardless of the dimension).

[0147] Then, the Pearson correlation analysis method was used to calculate the correlation coefficient between the possible influencing factors of the production effect of the 16 replacement wells and the comprehensive evaluation value of the production effect Z to determine the sensitive influencing factors of the production effect of the replacement wells. The Pearson correlation coefficient calculation formula is:

[0148]

[0149] Where r is the Pearson correlation coefficient between X and Y, a decimal; x i and y i are the observed values of X and Y, respectively, in decimal form; and are the average values of the observed values of X and Y, respectively, as decimals; n is the number of samples, as an integer.

[0150] The sensitive factors affecting the comprehensive evaluation value of the replacement well production effect include: S. NP o ′、IPR、WP o ′, the final formed replacement well screening model sample library includes the following parameters: S. NP o ′、IPR、WP o′ and Z.

[0151] Step S5: Determine the corresponding comprehensive evaluation value of the replacement well's production effect limit based on the replacement well's economic limit daily oil level and the relationship between the replacement well's daily liquid level in the first year of production, the cumulative oil production of the replacement well in each stage, and the average replacement cycle in step S2;

[0152] The economic limit daily oil level of the replacement well is set. Based on the water cut of the replacement well being basically maintained at around 30%, the economic limit daily liquid level is calculated. Then, based on the relationship between the daily liquid level in the first year of replacement production, the cumulative oil production of the stage replacement, and the average replacement cycle, the specific relationship is as follows:

[0153]

[0154] Where a and b are regression coefficients, dimensionless; m and n are regression exponents; N p_year is the cumulative oil production of the replacement in the first year of replacement production, t; q l_perday The daily liquid level in the first year of replacement production is calculated by dividing the cumulative liquid production of the replacement in the first year of replacement production by the number of days in the actual replacement month in a year, t·d -1 ; is the average replacement cycle in the first year of replacement production, d;

[0155] The cumulative oil production of the stage replacement corresponding to the economic limit daily liquid level and the average replacement cycle can be calculated, and then the comprehensive evaluation value of the production effect limit under the set economic limit daily oil level of the replacement well can be calculated according to formula (14).

[0156] Step S6: Divide the sample library obtained in step S4 into a training set and a test set in a ratio of 8:2, and use the multivariate linear regression algorithm for training. The coefficient of determination R between the predicted value and the actual value of the test set is 2 As an evaluation index of model accuracy, the polynomial characteristic order of the input data is optimized. After obtaining the optimal polynomial characteristic order, the optimal replacement well screening model is obtained by combining it with the production effect limit evaluation value of the replacement well obtained in step S5.

[0157] Coefficient of determination R 2 The calculation formula is as follows:

[0158]

[0159] Where R 2 The determination coefficient of the actual value and the predicted value for comprehensive evaluation of production effect, decimal; Z i is the actual value of the comprehensive evaluation of the production effect of the replacement well, a decimal; Z′ is the predicted value of the comprehensive evaluation of the production effect of the replacement well, a decimal; is the average value of the comprehensive evaluation of production effects of all replacement wells in the test set, a decimal; m is the number of test samples.

[0160] Step S7: Using the established replacement well screening model, potential replacement wells suitable for replacement production are screened from low-yield and low-efficiency wells and shut-in oil wells in the target block.

[0161] Find all low-yield, low-efficiency (monthly oil production less than 30 tons) and shut-in oil wells in the target block, and calculate the parameters required for the replacement well screening model: S. NP o ′、IPR、WP o ′, a sample library of replacement wells to be screened is formed, which is substituted into the replacement well screening model. The calculated comprehensive evaluation value of the production effect of the replacement well is compared with the comprehensive evaluation value of the production effect of the replacement well to screen out potential replacement wells suitable for replacement.

[0162] In a specific embodiment of the present invention, the data-driven alternative well screening method includes the following steps:

[0163] Step S1: Collect multi-source data of the replacement well and its surrounding oil and water wells in the target block, including geological characteristics, historical production data, wellbore structure data, and fluid physical property data;

[0164] Step S2: Based on the relevant data collected in step S1, by analyzing the production characteristics of the replacement well, the characterizing indicators of the replacement well production effect are determined as the replacement well cumulative oil production, daily liquid level and average replacement cycle in the first year of replacement production.

[0165] The calculation formula is as follows:

[0166]

[0167] Where N p_year is the cumulative oil production of the replacement in the first year of replacement production, t; L p_year The cumulative liquid production of the replacement in the first year of replacement production, t; q l_perday is the daily liquid level in the first year of replacement production (dividing the cumulative liquid production of the replacement in the first year of replacement production by the number of days in the actual replacement month in a year), t·d -1 ; is the average replacement cycle in the first year of replacement production, d; q oi is the oil production of the i-th replacement in the first year of replacement production, t; q wiis the water production of the i-th replacement in the first year of replacement production, t; n is the total number of replacements in the first year of replacement production, an integer; m is the actual total number of replacement months in the first year of replacement production (since replacement wells may be shut in some months or the wellhead equipment may be damaged and not replaced in the first year of replacement production, the actual number of replacement months may be less than 12, so the actual number of replacement months is used for calculation), an integer; T i is the interval between the i-th replacement and the i+1-th replacement in the first year of replacement production, d.

[0168] Step S3: The replacement well and its surrounding oil and water wells are regarded as a replacement well production unit (see Figure 2 ), based on reservoir engineering theory, multi-source data were integrated to analyze the influencing factors of the production effect of replacement wells from six aspects: physical parameter characteristics of the production unit of replacement wells, pressure flow field characteristics, structural correspondence characteristics, residual potential characteristics, seepage environment characteristics, and injection-production connectivity characteristics.

[0169] The factors affecting the replacement effect are analyzed from the physical property parameter characteristics of the replacement well production unit. The average effective thickness, effective porosity, and effective permeability of the replacement layer are used to characterize the physical property characteristics of the replacement well production unit. The calculation formula is as follows:

[0170]

[0171] Where h j is the effective thickness of the replacement layer of the jth well, m; h ji is the effective thickness of the ith replacement layer in the jth well, m; l is the total number of small production layers in the replacement, an integer; is the average effective thickness of the replacement layer, m; φ j is the effective porosity of the replacement layer of the jth well, decimal; φ ji is the effective porosity of the ith replacement layer in the jth well, decimal; is the average effective porosity of the replacement layer, decimal; K j is the effective permeability of the replacement layer of the jth well, 10 -3 μm 2 ;K ji is the effective permeability of the ith replacement layer in the jth well, 10 -3 μm 2 ; is the average effective permeability of the replacement layer, 10 -3 μm 2 ; m is the number of oil wells around the replacement well before replacement, an integer; n is the number of water wells around the replacement well before replacement, an integer.

[0172] The factors influencing the production performance of the replacement well are analyzed based on the pressure flow field characteristics of the replacement well production unit. The pressure flow field characteristics of the replacement well production unit are characterized by the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells (before replacement) and the cumulative injection-production ratio (before replacement). The formula for calculating the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells is as follows:

[0173]

[0174] Where Δp i is the pressure difference between the central replacement well and the replacement layer of the surrounding well i, MPa; p is the replacement layer pressure of the central replacement well, MPa; p i is the casing pressure of the surrounding well i, MPa; is the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells, MPa.

[0175] The calculation formula of cumulative injection-production ratio is as follows:

[0176]

[0177] Where, IPR is the cumulative injection-production ratio before replacement, decimal; WI j is the cumulative water injection volume of the replacement layer of the jth water well around the central replacement well before replacement, 10 4 m 3 NP oi is the cumulative oil production of the replacement layer of the i-th oil well before replacement, 10 4 t; WP oi is the cumulative water production of the replacement layer of the surrounding i-th oil well before replacement, 10 4 m 3 NP is the cumulative oil production of the replacement layer of the central replacement well before replacement, 10 4 t; WP is the cumulative water production of the replacement layer of the central replacement well before replacement, 10 4 m 3 θ oi is the inner angle of the polygon enclosed by the i-th oil well and other oil and water wells around the central replacement well, in degrees; θ wj The internal angle of the polygon enclosed by the j-th water well and other oil and water wells around the central replacement well, degrees; B o is the volume coefficient of formation crude oil, decimal; ρ o is the density of crude oil on the ground, g·cm -3 .

[0178] The factors influencing the production performance of the replacement wells were analyzed based on the residual potential characteristics of the replacement well production unit. The dimensionless cumulative oil and water production of the central replacement well (before replacement), the dimensionless cumulative oil and water production of the oil wells surrounding the central replacement well (before replacement), and the dimensionless cumulative water injection of the water wells surrounding the central replacement well (before replacement) were used to characterize the residual potential characteristics of the replacement well production unit. The dimensionless cumulative oil and water production (before replacement) of the central replacement well were calculated as follows:

[0179]

[0180] Where NP′ is the dimensionless cumulative oil production of the central replacement well before replacement, decimal; WP′ is the dimensionless cumulative water production of the central replacement well before replacement, decimal; A is the polygonal area enclosed by the oil and water wells around the central replacement well, km 2 .

[0181] The dimensionless cumulative oil and water production (before replacement) of the oil wells surrounding the central replacement well is calculated as follows:

[0182]

[0183] Where NP o ′ is the dimensionless cumulative oil production of the oil wells around the central replacement well before replacement, decimal; WP o ′ is the dimensionless cumulative water production of the oil wells around the central replacement well before replacement, a decimal.

[0184] The calculation formula for dimensionless cumulative water injection (before replacement) of the wells surrounding the central replacement well is as follows:

[0185]

[0186] Where WI k ′ is the dimensionless cumulative water injection of the wells around the central replacement well before replacement, a decimal.

[0187] The factors affecting the production effect of the replacement well are analyzed from the structural correspondence characteristics of the replacement well production unit. The structural correspondence characteristics of the replacement well production unit are characterized by the average depth difference in the middle of the replacement layer of the central replacement well and the surrounding oil and water wells, and the average formation dip of the replacement layer. The schematic diagram of the calculation of the average depth difference in the central replacement well and the surrounding oil and water wells is shown in the figure. Figure 3 As shown, the calculation formula is as follows:

[0188]

[0189] Where Δz ij is the difference in altitude between the replacement well and the middle depth of the jth replacement layer in the surrounding i-th well, m; z j is the depth above sea level of the middle of the jth replacement layer in the replacement well, m; z ij is the depth above sea level of the middle of the jth replacement layer of the surrounding i-th well, m; Δzj is the average elevation depth difference between the replacement well and the surrounding oil and water wells in the middle of the jth replacement layer, m; is the average altitude depth difference between the replacement well and the replacement layer of the surrounding oil and water wells, m.

[0190] The schematic diagram of calculating the average formation dip angle of the replacement layer is as follows: Figure 4 As shown, the calculation formula is as follows:

[0191]

[0192] Where, d ij is the distance between the replacement well and the jth replacement layer of the surrounding i-th well, m; θ ij is the formation dip angle (Δz ij There are positive and negative values. When Δz ij When it is positive, the θ calculated by the inverse tangent ij Between 0-90 degrees, when Δz ij When it is negative, the inverse tangent calculated θ ij between 90 and 180 degrees), degrees; θ j is the average dip angle of the jth replacement layer between the replacement well and the surrounding oil and water wells, degrees; is the average formation dip angle of the replacement layer, degrees.

[0193] The factors affecting the production effect of the replacement well are analyzed from the bottom seepage environment characteristics of the replacement well. The bottom seepage environment characteristics of the replacement well are characterized by the replacement well annulus space volume, the replacement layer effective seepage area, and the interlayer density. Figure 5 As shown, the calculation formula is as follows:

[0194]

[0195] Where V is the volume of the casing space of the replacement well, m 3 ;h o is the depth of the oil pipe, m; d t is the outer diameter of the oil layer casing, mm; b t Oil layer casing wall thickness, mm; d o is the outer diameter of the oil pipe, mm; h b is the bottom depth of the deepest replacement layer, m.

[0196] The schematic diagram of calculating the effective seepage area of the replacement layer is as follows: Figure 5 As shown, the calculation formula is as follows:

[0197]

[0198] Where S is the effective seepage area of the replacement layer, m 2 ;hi is the effective thickness of the ith replacement layer in the replacement well, m.

[0199] Schematic diagram of calculation of interlayer density of replacement layer Figure 5 As shown, the calculation formula is as follows:

[0200]

[0201] Where ρ is the density of interlayers in the replacement layer, m -1 ; c is the number of interlayers in the replacement layer, an integer; h t is the top depth of the shallowest replacement layer, m.

[0202] The factors influencing the production performance of the replacement wells were analyzed based on the injection-production connectivity characteristics of the replacement well production unit. Using time series data of injection and production fluid volumes from oil and water wells, the Dynamic Time Warping (DTW) algorithm was used to calculate the similarity between the replacement well production volume and the injection volume curves of surrounding water wells, thereby evaluating the dynamic connectivity between the injection and production wells. The greater the curve similarity, the better the connectivity. The comprehensive injection-production connectivity coefficient of the replacement well production unit was used to characterize the injection-production connectivity characteristics of the replacement well production unit. The calculation formula is as follows:

[0203]

[0204] Where, is the comprehensive injection-production connectivity coefficient of the replacement well production unit, decimal; DTW j It is the DTW distance between the replacement well and the jth water well around it, a decimal.

[0205] Step S4: The objective weighting method, the entropy method, is used to calculate the weights C of the three production effect characterization indicators of the replacement well. Then, the comprehensive scoring method is used to calculate the comprehensive evaluation value of the production effect of the replacement well. The calculation formula is as follows:

[0206]

[0207] Where Z is the comprehensive evaluation value of the production effect of the replacement well, a decimal; W i is the weight of the production effect characterization index of the replacement well i, a decimal; C i is the value of the production effect characterization index of the replacement well, a decimal (here only the values of each production effect evaluation index are taken, the larger the value, the better the production effect, regardless of the dimension).

[0208] Then, the Pearson correlation analysis method is used to calculate the correlation coefficient between the possible influencing factors of the production effect of the 16 replacement wells in step S3 and the comprehensive evaluation value Z of the production effect, and determine the sensitive influencing factors of the production effect of the replacement wells. The Pearson correlation coefficient calculation formula is:

[0209]

[0210] Where r is the Pearson correlation coefficient between X and Y, a decimal; x i and y i are the observed values of X and Y, respectively, in decimal form; and are the average values of the observed values of X and Y, respectively, as decimals; n is the number of samples, as an integer.

[0211] The sensitive factors affecting the comprehensive evaluation value of the replacement well production effect include: S. NP o ′、IPR、WP o ′, the final formed replacement well screening model sample library includes the following parameters: S. NP o ′、IPR、WP o ′ and Z.

[0212] Step S5: The economic limit daily oil level of the replacement well is 0.5t·d, which is determined based on the current oil price and the replacement production cost. -1 Since the water content of the replacement well is basically maintained at around 30%, the economic limit of daily liquid level is 0.71t·d -1 According to the relationship between the daily liquid level in the first year of replacement production, the cumulative oil production of the stage replacement and the average replacement cycle, the specific relationship is as follows:

[0213]

[0214] When the economic limit daily liquid level is 0.71t·d -1 When the average replacement cycle and the cumulative oil production of the stage replacement are calculated by formula (16), they are 4.24d and 104.48t respectively. Then, the comprehensive evaluation value of the production effect limit is calculated by formula (14) as 49. Therefore, the economic limit daily oil level is considered to be 0.5t·d -1 When the comprehensive evaluation value of production effect is greater than or equal to 49, replacement production is feasible; otherwise, replacement production is not feasible.

[0215] Step S6: Using the alternative well screening model sample library, based on the spyder development environment, using Python language and combining with the third-party machine learning library scikit-learn, according to the basic principles of multiple linear regression, write a program to divide the alternative well screening sample library into a training set and a test set in an 8:2 ratio for learning and training, and use the determination coefficient R between the predicted value and the actual value of the test set. 2 As an evaluation index of model accuracy, the polynomial characteristic order of the input data is optimized, such as Figure 6As shown, combined with the comprehensive evaluation value of the replacement well production effect limit obtained in step S5, the best replacement well screening model is obtained. 2 The calculation formula is as follows:

[0216]

[0217] Where R 2 The determination coefficient of the actual value and the predicted value for comprehensive evaluation of production effect, decimal; Z i is the actual value of the comprehensive evaluation of the production effect of the replacement well, a decimal; Z′ is the predicted value of the comprehensive evaluation of the production effect of the replacement well, a decimal; is the average value of the comprehensive evaluation of production effects of all replacement wells in the test set, a decimal; m is the number of test samples.

[0218] The economic limit of daily oil production for the replacement well is 0.5t·d -1 When , the optimal replacement well screening model is:

[0219]

[0220] Step S7: Based on the economic limit of the replacement well daily oil level of 0.5t·d -1 The established replacement well screening model formula (18) is used to screen the oil wells in the target block and select potential replacement wells suitable for replacement. From the target block, all low-yield and low-efficiency (monthly oil production less than 30t) and shut-in oil wells, a total of 17 wells, are found. According to the required parameters of the replacement well screening model and the calculation method of the factors affecting the production effect of replacement wells in step S3, a sample library of 17 replacement wells to be screened in the target block is calculated and sorted out. The sample library of replacement wells to be screened is substituted into the replacement well screening model to obtain the comprehensive evaluation value of the production effect of each replacement well to be screened. The sample library of replacement wells to be screened in the target block and the comprehensive evaluation value of production effect are shown in Table 1. As can be seen from Table 1, the comprehensive evaluation value of production effect of each replacement well to be screened is different. According to the replacement well screening model, only those with a comprehensive evaluation value of production effect Z greater than or equal to 49 are suitable for replacement production. In Table 1, there are 11 replacement wells to be screened whose comprehensive production effect evaluation values are greater than or equal to 49. It is believed that replacement production of these 11 wells is feasible and these 11 wells are called potential replacement wells.

[0221] Table 1 Sample library of replacement wells to be screened in the target block and comprehensive evaluation values of production effects

[0222]

[0223] In order to observe the distribution of 42 historical replacement wells and 11 potential replacement wells in the target block, a well location distribution map was drawn according to their well location coordinates. Figure 7It can be seen that the 11 potential replacement wells selected by the screening model are mainly distributed near the historical replacement wells, and some are distributed in the marginal areas of the target block.

[0224] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art may modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features therein. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

[0225] Except for the technical features described in the specification, all other technical features are known technologies to those skilled in the art.

Claims

1. A data-driven alternative well screening method, characterized in that: The data-driven alternative well screening method includes: Step 1: Collect multi-source data on the replacement wells and surrounding oil and water wells in the target block; Step 2: Determine the characterization index of the production effect of the replacement well by analyzing the production characteristics of the replacement well; Step 3: Analyze the factors affecting the production effect of the replacement well; Step 4: Determine the sensitivity factors affecting the production performance of the replacement wells and form a sample library for the replacement well screening model; Step 5: Determine the corresponding comprehensive evaluation value of the production effect limit of the replacement well; Step 6: Establish the best alternative well screening model; Step 7: Use the established replacement well screening model to select potential replacement wells suitable for replacement production from low-yield and low-efficiency wells and shut-in oil wells in the target block; In step 2, based on the relevant data collected in step 1, by analyzing the production characteristics of the replacement well, the characterizing indicators of the production effect of the replacement well are determined to be the cumulative oil production of the replacement well in the first year of replacement production, the daily liquid level and the average replacement cycle; In step 3, the replacement well and its surrounding oil and water wells are considered as a replacement well production unit. Based on reservoir engineering theory, multi-source data are integrated to analyze the factors affecting the replacement well production effect from six aspects: physical parameter characteristics of the replacement well production unit, pressure and flow field characteristics, structural correspondence characteristics, residual potential characteristics, seepage environment characteristics, and injection-production connectivity characteristics. In step 4, the entropy method and comprehensive scoring method are used to calculate the comprehensive evaluation value of the replacement well production effect based on the replacement well production effect characterization index in step 2, and a Pearson correlation analysis is performed to determine the sensitivity influencing factors of the replacement well production effect, thereby forming a sample library of the replacement well screening model; In step 4, the objective weighting method, the entropy method, is used to calculate the weights C of the three production effect characterization indicators of the replacement well. Then, the comprehensive scoring method is used to calculate the comprehensive evaluation value of the production effect of the replacement well. The calculation formula is as follows: Where Z is the comprehensive evaluation value of the production effect of the replacement well, a decimal; W i is the weight of the production effect characterization index of the replacement well i, a decimal; C i is the production effect characterization index value of the replacement well i, decimal; Then, the Pearson correlation analysis method is used to calculate the correlation coefficient between the possible influencing factors of the production effect of the replacement well and the comprehensive evaluation value Z of the production effect, and to determine the sensitive influencing factors of the production effect of the replacement well; In step 5, based on the economic limit daily oil level of the replacement well, combined with the relationship between the daily liquid level in the first year of replacement production, the cumulative oil production of the stage replacement and the average replacement cycle in step 2, the corresponding comprehensive evaluation value of the replacement well production effect limit is determined; In step 6, the sample library obtained in step 4 is divided into a training set and a test set in a ratio of 8:2, and the multivariate linear regression algorithm is used for training. The coefficient of determination R between the predicted value and the actual value of the test set is 2 As an evaluation index of model accuracy, the polynomial characteristic order of the input data is optimized. After obtaining the optimal polynomial characteristic order, combined with the production effect limit evaluation value of the replacement well obtained in step 5, the best replacement well screening model is obtained.

2. The data-driven alternative well screening method according to claim 1, characterized in that: In step 1, multi-source data on the replacement wells and surrounding oil and water wells in the target block are collected, including geological characteristics, historical production data, wellbore structure data, and fluid physical property data.

3. The data-driven alternative well screening method according to claim 2, characterized in that: In step 1, the collected data include: effective thickness, effective porosity, effective permeability, casing pressure, pump hanging depth, dynamic liquid level depth, central depth of the replacement layer, monthly water injection volume of surrounding water wells, monthly oil production and monthly water production of surrounding oil wells, monthly oil production and monthly water production of the central replacement well, well location coordinates, tubing depth and outer diameter, and outer diameter and wall thickness of the oil layer casing.

4. The data-driven alternative well screening method according to claim 1, characterized in that: In step 2, the calculation formulas for the replacement well production performance indicators, namely the replacement cumulative oil production, daily liquid level and average replacement cycle in the first year of replacement production, are as follows: Where N p_year is the cumulative oil production of the replacement in the first year of replacement production, t; L p_year The cumulative liquid production of the replacement in the first year of replacement production, t; q l_perday The daily liquid level in the first year of replacement production is calculated by dividing the cumulative liquid production of the replacement in the first year of replacement production by the number of days in the actual replacement month in a year, t·d -1 ; is the average replacement cycle in the first year of replacement production, d; q oi is the oil production of the i-th replacement in the first year of replacement production, t; q wi is the water production of the i-th replacement in the first year of replacement production, t; n is the total number of replacements in the first year of replacement production, an integer; m is the actual total number of replacement months in the first year of replacement production. Since replacement wells may be shut in some months or the wellhead equipment may be damaged and not replaced during the first year of replacement production, the actual number of replacement months may be less than 12, so the actual number of replacement months is used for calculation, an integer; T i is the interval between the i-th replacement and the i+1-th replacement in the first year of replacement production, d.

5. The data-driven alternative well screening method according to claim 1, characterized in that: In step 3, the factors affecting the replacement effect are analyzed based on the physical property parameter characteristics of the replacement well production unit. The average effective thickness, effective porosity, and effective permeability of the replacement layer are used to characterize the physical property characteristics of the replacement well production unit. The calculation formula is as follows: Where h j is the effective thickness of the replacement layer of the jth well, m; h ji is the effective thickness of the ith replacement layer in the jth well, m; l is the total number of small production layers in the replacement, an integer; is the average effective thickness of the replacement layer, m; φ j is the effective porosity of the replacement layer of the jth well, decimal; φ ji is the effective porosity of the ith replacement layer in the jth well, decimal; is the average effective porosity of the replacement layer, decimal; K j is the effective permeability of the replacement layer of the jth well, 10 -3 μm 2 ;K ji is the effective permeability of the ith replacement layer in the jth well, 10 -3 μm 2 ; is the average effective permeability of the replacement layer, 10 -3 μm 2 ; m is the number of oil wells around the replacement well before replacement, an integer; n is the number of water wells around the replacement well before replacement, an integer.

6. The data-driven alternative well screening method according to claim 1, characterized in that: In step 3, the factors affecting the production effect of the replacement well are analyzed from the pressure flow field characteristics of the replacement well production unit, and the pressure flow field characteristics of the replacement well production unit are characterized by the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells and the cumulative injection-production ratio.

7. The data-driven alternative well screening method according to claim 6, characterized in that: In step 3, the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells is calculated as follows: Where Δp i is the pressure difference between the central replacement well and the replacement layer of the surrounding well i, MPa; p is the pressure of the replacement layer in the central replacement well, MPa; p i is the casing pressure of the surrounding well i, MPa; is the average pressure difference between the central replacement well and the surrounding oil and water well replacement layers, MPa; The calculation formula of cumulative injection-production ratio is as follows: Where, IPR is the cumulative injection-production ratio before replacement, decimal; WI j is the cumulative water injection volume of the replacement layer of the jth water well around the central replacement well before replacement, 10 4 m 3 ; NP oi is the cumulative oil production of the replacement layer of the i-th oil well before replacement, 10 4 t; WP oi is the cumulative water production of the replacement layer of the surrounding i-th oil well before replacement, 10 4 m 3 NP is the cumulative oil production of the replacement layer of the central replacement well before replacement, 10 4 t; WP is the cumulative water production of the replacement layer of the central replacement well before replacement, 10 4 m 3 ; θ oi is the interior angle of the polygon enclosed by the i-th oil well and other oil and water wells around the central replacement well, in degrees; θ wj The internal angle of the polygon enclosed by the j-th water well and other oil and water wells around the central replacement well, degrees; B o is the formation crude oil volume coefficient, decimal; ρ o is the density of crude oil on the ground, g·cm -3 .

8. The data-driven alternative well screening method according to claim 1, characterized in that: In step 3, the factors affecting the production effect of the replacement well are analyzed from the residual potential characteristics of the replacement well production unit. The dimensionless cumulative oil and water production of the central replacement well, the dimensionless cumulative oil and water production of the oil wells around the central replacement well, and the dimensionless cumulative water injection of the water wells around the central replacement well are used to characterize the residual potential characteristics of the replacement well production unit.

9. The data-driven alternative well screening method according to claim 8, characterized in that: In step 3, the dimensionless cumulative oil and water production of the central replacement well are calculated as follows: Where NP′ is the dimensionless cumulative oil production of the central replacement well before replacement, decimal; WP′ is the dimensionless cumulative water production of the central replacement well before replacement, decimal; A is the polygonal area enclosed by the oil and water wells around the central replacement well, km 2 ; is the average effective thickness of the replacement layer, m; NP is the cumulative oil production of the replacement layer of the central replacement well before replacement, 10 4 t; is the average effective porosity of the replacement layer, decimal; B o is the formation crude oil volume coefficient, decimal; ρ o is the density of crude oil on the ground, g·cm -3 ; The dimensionless cumulative oil and water production calculation formulas for the oil wells surrounding the central replacement well are as follows: Where NP o ′ is the dimensionless cumulative oil production of the oil wells around the central replacement well before replacement, decimal; WP o ′ is the dimensionless cumulative water production of the oil wells around the central replacement well before replacement, decimal; θ oi is the interior angle of the polygon enclosed by the i-th oil well and other oil and water wells around the central replacement well, in degrees; NP oi is the cumulative oil production of the replacement layer of the i-th oil well before replacement, 10 4 t; The dimensionless cumulative water injection calculation formula for the wells surrounding the central replacement well is as follows: Where WI k ' is the dimensionless cumulative water injection of the surrounding wells of the central replacement well before replacement, decimal; WI j is the cumulative water injection volume of the replacement layer of the jth water well around the central replacement well before replacement, 10 4 m 3 θ wj The internal angle of the polygon enclosed by the j-th water well and other oil and water wells around the central replacement well, in degrees.

10. The data-driven alternative well screening method according to claim 1, characterized in that: In step 3, the factors affecting the production effect of the replacement well are analyzed from the structural correspondence characteristics of the replacement well production unit, and the structural correspondence characteristics of the replacement well production unit are characterized by the average altitude depth difference between the central replacement well and the replacement layer of the surrounding oil and water wells, and the average formation dip of the replacement layer.

11. The data-driven alternative well screening method according to claim 10, characterized in that: In step 3, the formula for calculating the average depth difference between the central replacement well and the replacement layers of the surrounding oil and water wells is as follows: Where Δz ij is the difference in altitude between the replacement well and the middle depth of the jth replacement layer in the surrounding i-th well, m; z j is the depth above sea level of the middle of the jth replacement layer in the replacement well, m; z ij is the depth above sea level of the middle of the jth replacement layer of the surrounding i-th well, m; Δz j is the average elevation depth difference between the replacement well and the surrounding oil and water wells in the middle of the jth replacement layer, m; is the average elevation depth difference between the replacement well and the replacement layer of the surrounding oil and water wells, m; The calculation formula for the average formation dip angle of the replacement layer is as follows: Where, d ij is the distance between the replacement well and the jth replacement layer of the surrounding i-th well, m; θ ij is the formation dip angle between the replacement well and the jth replacement layer of the surrounding i-th well, Δz ij There are positive and negative values. When Δz ij When it is positive, the θ calculated by the inverse tangent ij Between 0-90 degrees, when Δz ij When it is negative, the inverse tangent calculated θ ij Between 90-180 degrees, degrees; θ j is the average dip angle of the jth replacement layer between the replacement well and the surrounding oil and water wells, degrees; θ is the average dip angle of the replacement layer, degrees.

12. The data-driven alternative well screening method according to claim 1, characterized in that: In step 3, the factors affecting the production effect of the replacement well are analyzed based on the bottom hole seepage environment characteristics of the replacement well, and the bottom hole seepage environment characteristics of the replacement well are characterized by the annular space volume of the replacement well, the effective seepage area of the replacement layer, and the density of the interlayer.

13. The data-driven alternative well screening method according to claim 12, characterized in that: In step 3, the calculation formula for the annular space volume of the replacement well is as follows: Where V is the volume of the casing space of the replacement well, m 3 ;h o is the depth of the oil pipe, m; d t is the outer diameter of the oil layer casing, mm; b t Oil layer casing wall thickness, mm; d o is the outer diameter of the oil pipe, mm; h b is the bottom depth of the deepest replacement layer, m; The calculation formula for the effective seepage area of the replacement layer is as follows: Where S is the effective seepage area of the replacement layer, m 2 ; h i is the effective thickness of the ith replacement layer in the replacement well, m; The calculation formula for the density of the interlayer in the replacement layer is as follows: Where ρ is the density of interlayers in the replacement layer, m -1 ; c is the number of interlayers in the replacement layer, an integer; h t is the top depth of the shallowest replacement layer, m.

14. The data-driven alternative well screening method according to claim 1, characterized in that: In step 3, the factors affecting the production performance of the replacement well are analyzed based on the injection-production connectivity characteristics of the replacement well production unit. The dynamic time warping algorithm is used to calculate the similarity between the replacement well production rate and the injection rate change curve of the surrounding water wells using the time series data of the injection and production volume of the oil and water wells, and the dynamic connectivity between the injection and production wells is evaluated. The greater the curve similarity, the better the connectivity.

15. The data-driven alternative well screening method according to claim 14, characterized in that: In step 3, the comprehensive injection-production connectivity coefficient of the replacement well production unit is used to characterize the injection-production connectivity characteristics of the replacement well production unit. The calculation formula is as follows: Where, is the comprehensive injection-production connectivity coefficient of the replacement well production unit, decimal; DTW j It is the DTW distance between the replacement well and the jth water well around it, a decimal.

16. The data-driven alternative well screening method according to claim 1, characterized in that: The formula for calculating the Pearson correlation coefficient is: Where r is the Pearson correlation coefficient between X and Y, a decimal; x i and y i are the observed values of X and Y, respectively, in decimal form; and are the average values of the observed values of X and Y, respectively, in decimal form; n is the number of samples, in integer form; Sensitive factors affecting the comprehensive evaluation value of the replacement well production effect include: the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells Effective seepage area S of the replacement layer, average effective thickness of the replacement layer Average formation dip of the replacement layer Dimensionless cumulative oil production NP of the oil wells around the central replacement well before replacement o ′, cumulative injection-production ratio (IPR) before replacement, dimensionless cumulative water production (WP) of oil wells around the central replacement well before replacement o ′, the final formed replacement well screening model sample library includes the following parameters: S. NP o ′、IPR、WP o ′ and comprehensive evaluation value Z of production effect.

17. The data-driven alternative well screening method according to claim 1, characterized in that: In step 5, the economic limit daily oil level of the replacement well is set. Based on the water cut of the replacement well being maintained at 30%, the economic limit daily liquid level is calculated. Then, based on the relationship between the daily liquid level in the first year of replacement production, the cumulative oil production of the stage replacement, and the average replacement cycle, the specific relationship is as follows: Where a and b are regression coefficients, dimensionless; m and n are regression exponents; N p_year is the cumulative oil production of the replacement in the first year of replacement production, t; q l_perday The daily liquid level in the first year of replacement production is calculated by dividing the cumulative liquid production of the replacement in the first year of replacement production by the number of days in the actual replacement month in a year, t·d -1 ; is the average replacement cycle in the first year of replacement production, d; The cumulative oil production of the stage replacement corresponding to the economic limit daily liquid level and the average replacement cycle are calculated, and then the comprehensive evaluation value of the production effect limit under the set economic limit daily oil level of the replacement well is calculated according to formula (14).

18. The data-driven alternative well screening method according to claim 1, characterized in that: In step 6, the coefficient of determination R 2 The calculation formula is as follows: Where R 2 The determination coefficient of the actual value and the predicted value for comprehensive evaluation of production effect, decimal; Z i is the actual value of the comprehensive evaluation of the production effect of the replacement well, a decimal; Z′ is the predicted value of the comprehensive evaluation of the production effect of the replacement well, a decimal; is the average value of the comprehensive evaluation of production effects of all replacement wells in the test set, a decimal; m is the number of test samples.

19. The data-driven alternative well screening method according to claim 1, characterized in that: In step 7, find all low-yield, low-efficiency and shut-in oil wells in the target block and calculate the parameters required for the replacement well screening model: the average pressure difference between the central replacement well and the replacement layers of the surrounding oil and water wells Effective seepage area S of the replacement layer, average effective thickness of the replacement layer Average formation dip of the replacement layer Dimensionless cumulative oil production NP of the oil wells around the central replacement well before replacement o ′, cumulative injection-production ratio (IPR) before replacement, dimensionless cumulative water production (WP) of oil wells around the central replacement well before replacement o ′, a sample library of replacement wells to be screened is formed, which is substituted into the replacement well screening model. The calculated comprehensive evaluation value of the production effect of the replacement well is compared with the comprehensive evaluation value of the production effect of the replacement well to screen out potential replacement wells suitable for replacement.

Images