Method and system for predicting productivity of single well after pressing of volcanic reservoir

Abstract

The embodiment of the invention provides a volcanic rock reservoir after-pressing single well productivity prediction method and system, and belongs to the technical field of oil reservoir exploitation. The method comprises the following steps: classifying logging information based on each preset evaluation index to obtain a logging data set corresponding to each preset evaluation index; executing index parameter calculation corresponding to the preset evaluation index; correspondingly determining a reservoir quality index and an engineering quality index of the target prediction area based on the index parameter of each preset evaluation index, and constructing a corresponding comprehensive evaluation index; and constructing a relation model between the productivity of the scale well and the comprehensive evaluation index. According to the scheme, the blank of volcanic rock reservoir after-fracturing single well oil production prediction in the prior art is filled up, important technical support is provided for reasonable selection of fracturing layer sections, the success rate of fracturing transformation and the oil and gas field development efficiency can be improved, and the method has remarkable application value.

Description

Technical Field

[0001] This invention relates to the field of oil reservoir development technology, specifically to a method for predicting the production capacity of a single well after compression in a volcanic rock reservoir and a system for predicting the production capacity of a single well after compression in a volcanic rock reservoir. Background Technology

[0002] Volcanic reservoirs, due to their unique geological structure and complex lithological characteristics, have always been one of the challenges in oil and gas exploration and development. Volcanic reservoirs typically exhibit complex pore structures, diverse lithologies, and strong heterogeneity, leading to significant differences in permeability and storage capacity. This heterogeneity results in unsatisfactory conventional well testing in volcanic reservoirs, often demonstrating low production per well. Therefore, fracturing is a necessary means to obtain industrial oil and gas flow in volcanic reservoirs. However, how to rationally select fracturing intervals and accurately predict the post-fracturing production capacity of individual wells is of great significance for improving oil and gas field development efficiency and optimizing fracturing technology.

[0003] Existing reservoir productivity prediction methods primarily target natural gas reservoirs. They qualitatively determine reservoir productivity levels by constructing a reservoir productivity index and then use regression analysis to establish a quantitative interpretation model of reservoir productivity, combining parameters such as effective porosity and gas saturation. This method demonstrates a certain degree of accuracy in predicting productivity in some volcanic reservoirs. However, existing methods are mainly applied to natural gas reservoirs, and there is a lack of systematic research on predicting post-compression single-well oil production in volcanic reservoirs. The lithological diversity and complex pore structure of volcanic reservoirs make existing prediction methods difficult to effectively apply to post-compression single-well productivity prediction.

[0004] Furthermore, existing methods lack effective guidance tools for selecting fracturing intervals and cannot accurately assess post-fracturing productivity. Therefore, accurately predicting the post-fracturing productivity of a single well in a volcanic reservoir can not only provide a basis for the rational selection of fracturing intervals but also improve the success rate of fracturing stimulation and overall development efficiency. This deficiency in existing technology indicates an urgent need for a method that can effectively predict the productivity of volcanic reservoirs.

[0005] A model or method for calculating the single-well productivity after pressure is proposed to compensate for the shortcomings of existing solutions. Summary of the Invention

[0006] The purpose of this invention is to provide a method for predicting the production capacity of a single well after compression in volcanic rock reservoirs, so as to at least solve the problem that existing methods cannot predict the production capacity of volcanic rock reservoirs.

[0007] To achieve the above objectives, the first aspect of the present invention provides a method for predicting the post-compression single-well productivity of volcanic reservoirs. The method includes: collecting logging information of a target prediction area and classifying the logging information based on various preset evaluation indicators to obtain logging datasets corresponding to each preset evaluation indicator; calculating the index parameters of each preset evaluation indicator based on the logging datasets; determining the reservoir quality index and engineering quality index of the target prediction area based on the index parameters of each preset evaluation indicator, and constructing a corresponding comprehensive evaluation index based on the reservoir quality index and the engineering quality index; wherein the reservoir quality index is used to characterize the reservoir characteristics, and the engineering quality index is used to characterize the geological characteristics of the reservoir under engineering conditions; and constructing a relationship model between the calibrated well productivity and the comprehensive evaluation index based on the pre-determined calibrated well productivity of the target prediction area, using the relationship model as a post-compression single-well productivity prediction model for volcanic reservoirs, and performing single-well productivity prediction.

[0008] Optionally, before classifying the logging information based on each preset evaluation index, the method further includes performing preprocessing on the logging information. The preprocessing rules are as follows: performing irrelevant parameter filtering, missing value filling, outlier removal, and data normalization on the logging information in sequence.

[0009] Optionally, the step of classifying the logging information based on each preset evaluation index to obtain a logging dataset corresponding to each preset evaluation index includes: determining the corresponding clustering target cluster based on each preset evaluation index, and randomly determining the initial centroid of each cluster; traversing the logging information, calculating the Euclidean distance between each data point and each centroid, and classifying the data point to the nearest centroid; after completing one round of classification, updating the centroid position based on the mean of the data points in each cluster; and re-performing one round of clustering based on the updated centroid position.

[0010] The clustering process is repeated N times until the centroid of each cluster remains unchanged or the preset number of iterations is reached. Based on the latest clustering results, the well logging dataset for each preset evaluation index is determined.

[0011] Optionally, the preset evaluation indicators include any one or more of the following: lithological indicators, physical property indicators, electrical property indicators, and oil-bearing indicators; and any one or two of the following: fracture parameter indicators and brittleness parameter indicators.

[0012] Optionally, the calculation rules for the lithological index parameters are as follows:

[0013]

[0014] Among them, I lith GR represents the index parameter of lithology; GR is the natural gamma ray logging curve value; GR 基值1 is the natural gamma-ray value, AP; cn is the boundary value of the neutron logging curve; den is the boundary value of the density logging curve; DEN is the density logging curve value; CN is the neutron logging curve value; a and b are the neutron and density curve scale numbers, respectively.

[0015] Optionally, the calculation rules for the index parameters of physical properties are as follows:

[0016]

[0017] Among them, I por AC represents the physical property index parameter; AC is the acoustic transit time curve; AC 基值 The baseline value is the acoustic time difference baseline value; R LLd R LLs These represent the deep and shallow lateral resistivity, respectively; mf is the structural index; R mf The resistivity of mud filtrate under formation conditions.

[0018] Optionally, the calculation rules for the electrical performance parameters are as follows:

[0019]

[0020] Among them, I rt RT represents the electrical properties; RT is the formation resistivity curve; RT 基值 This is the base value for resistivity.

[0021] Optionally, the calculation rules for the oil content index parameters are as follows:

[0022]

[0023] Among them, I o QT is the index parameter for oil content; QT is the total hydrocarbon curve value within the layer. min This represents the minimum value of the total hydrocarbon curve within the stratum; QT max This represents the maximum value of the total hydrocarbon curve within the stratum.

[0024] Optionally, the calculation rules for the crack parameter index are as follows:

[0025] I frac =C×L;

[0026] Among them, I frac is the index parameter of the crack parameter index; C is the normalized crack density; L is the normalized crack length.

[0027] Optionally, the calculation rules for the brittleness parameter index are as follows: based on the well logging dataset of the brittleness parameter index, calculate the Young's modulus and Poisson's ratio of the reservoir section in the target prediction area respectively; perform normalization processing on the Young's modulus and the Poisson's ratio respectively; and calculate the brittleness parameter index based on the normalized Young's modulus and the normalized Poisson's ratio.

[0028] Optionally, the calculation rules for Young's modulus and Poisson's ratio of the reservoir interval in the target prediction area based on the well logging dataset of brittle parameters are as follows:

[0029]

[0030] Where E is the Young's modulus of the reservoir section; DT p and DT s These are the wave time differences for the longitudinal and transverse waves, respectively; ρ b 1 represents the density logging curve; μ represents the Poisson's ratio of the reservoir section.

[0031] Optionally, the rules for normalizing the Young's modulus and the Poisson's ratio are as follows:

[0032]

[0034] Among them, BI E E is the normalized Young's modulus; E is the Young's modulus of the reservoir section; E min E represents the minimum Young's modulus of the reservoir section. max BI represents the maximum Young's modulus of the reservoir section. PR The normalized Poisson's ratio is μ; μ is the Poisson's ratio of the reservoir section; μ min Minimum Poisson's ratio in the reservoir section; μ max The maximum Poisson's ratio of the reservoir section.

[0035] Optionally, the calculation rules for the brittleness parameter index based on the normalized Young's modulus and the normalized Poisson's ratio are as follows:

[0036]

[0037] Among them, I brit The brittleness parameter is the index parameter; BI E The normalized Young's modulus; BI PR This is the normalized Poisson's ratio.

[0038] Optionally, the reservoir quality index and engineering quality index of the target prediction area are determined based on the index parameters of each preset evaluation index, including: the index parameters of lithological index, physical property index, electrical property index and oil-bearing index, and the reservoir quality index calculation model, to calculate the reservoir quality index of the target prediction area; the index parameters of fracture parameter index and brittleness parameter index, and the engineering quality index calculation model, to calculate the engineering quality index of the target prediction area.

[0039] Optionally, the reservoir quality index calculation model is as follows:

[0040] I c =I lith ×I por ×I rt ×I o

[0041] Among them, I c The reservoir quality index for the target prediction region; I lith The index parameters for the lithological indices of the target prediction area; I por The index parameters of the physical properties of the target prediction area; I rt The index parameters for the electrical properties of the target prediction region; I o The index parameters are the oil content indicators for the target prediction area.

[0042] Optionally, the engineering quality index calculation model is as follows:

[0043] Ig = I frac *I brit

[0044] Where Ig is the engineering quality index of the target prediction area; I frac The index parameters of the crack parameters in the target prediction area; I brit The brittleness parameter index of the target prediction region.

[0045] Optionally, the rule for determining the production capacity of the calibration well is as follows: within the target area, obtain the production state information of multiple oil wells after fracturing; based on the production state information, identify the production state of each corresponding well; when the production state is identified as stable, re-collect the production state information of each oil well under stable conditions; based on the re-collected production state information, calculate the average oil production per unit time of each oil well, and use the average oil production per unit time of each oil well as the production capacity of the calibration well.

[0046] Optionally, the step of constructing a relationship model between the calibrated well productivity and the comprehensive evaluation index, and using the relationship model as a post-compression single-well productivity prediction model for volcanic reservoirs, includes: performing linear regression processing on the calibrated well productivity and the comprehensive evaluation index to obtain a linear relationship between the two; verifying the linear relationship based on a reserved validation dataset, and optimizing and adjusting it according to error indicators to obtain a final linear relationship model, and using the linear relationship model as a post-compression single-well productivity prediction model for volcanic reservoirs.

[0047] A second aspect of this invention provides a post-compression single-well productivity prediction system for volcanic reservoirs. The system includes: a data acquisition unit for acquiring well logging information of a target prediction area and classifying the well logging information based on preset evaluation indicators to obtain well logging datasets corresponding to each preset evaluation indicator; a processing unit for performing indicator parameter calculations based on the well logging datasets of each preset evaluation indicator to obtain indicator parameters for each preset evaluation indicator; a comprehensive evaluation unit for determining a reservoir quality index and an engineering quality index for the target prediction area based on the indicator parameters of each preset evaluation indicator, and constructing a corresponding comprehensive evaluation index based on the reservoir quality index and the engineering quality index; wherein the reservoir quality index characterizes the reservoir characteristics, and the engineering quality index characterizes the geological characteristics of the reservoir under engineering conditions; and a prediction unit for constructing a relationship model between the calibrated well productivity and the comprehensive evaluation index based on a pre-determined calibrated well productivity of the target prediction area, using the relationship model as a post-compression single-well productivity prediction model for volcanic reservoirs, and performing the calculations.

[0048] Single-well production capacity forecast.

[0049] On the other hand, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described method for predicting the production capacity of a single well after compression in a volcanic reservoir.

[0050] Through the above technical solution, this invention effectively obtains various reservoir parameters by collecting well logging information from the target prediction area and classifying and calculating the information based on preset evaluation indicators. Furthermore, these parameters are used to construct reservoir quality and engineering quality indices, respectively characterizing the reservoir's properties and geological characteristics under engineering conditions, thereby generating a comprehensive evaluation index. This method quantifies various complex reservoir factors into operable evaluation indicators, improving the comprehensive understanding and analytical accuracy of volcanic reservoirs. By establishing a relationship model with calibrated well productivity data, this solution can provide a scientific basis for predicting the post-fracturing single-well productivity of volcanic reservoirs, achieving efficient and accurate productivity prediction. This not only fills the gap in existing technologies for predicting post-fracturing single-well oil production in volcanic reservoirs but also provides important technical support for the rational selection of fracturing intervals, helping to improve the success rate of fracturing stimulation and the efficiency of oil and gas field development, demonstrating significant application value.

[0051] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0052] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings:

[0053] Figure 1 This is a flowchart of the steps of a method for predicting the single-well productivity of volcanic rock reservoirs after compression, provided by one embodiment of the present invention.

[0054] Figure 2 This is a cross-plot of acoustic transit time and resistivity of rough breccia provided by one embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram of a post-compression single-well productivity prediction model for a rough volcanic rock reservoir provided by one embodiment of the present invention;

[0056] Figure 4 This is a system structure diagram of a single-well productivity prediction system for volcanic rock reservoirs after compression, provided in one embodiment of the present invention. Detailed Implementation

[0057] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0058] Currently, methods for predicting reservoir productivity using well logging data are not yet mature, but many scholars have conducted in-depth research on this issue. In summary, the main research methods fall into two categories: one category establishes a direct statistical relationship between reservoir productivity and extensive well testing data, well logging interpretation, and core analysis data. For example, Ouyang Jian proposed a method to evaluate reservoir productivity using effective permeability and oil saturation, while Mao Zhiqiang et al. established a mathematical model for predicting reservoir productivity based on effective permeability. The other category employs mathematical algorithms, such as principal component analysis, fuzzy mathematics, and neural network technology, to construct reservoir productivity prediction systems. However, these methods are mainly applicable to medium-to-high porosity and permeability sandstone reservoirs, and their productivity prediction models rely on planar radial flow models. For low-porosity and low-permeability reservoirs, due to the complex seepage mechanism and numerous influencing factors, the above models cannot be directly applied for productivity prediction.

[0059] In recent years, technologies for predicting the production capacity of low-porosity and ultra-low-permeability reservoirs have gradually developed. For example, Guo Haopeng et al. proposed a rapid prediction method based on five production capacity-sensitive parameters for the ultra-low permeability sandstone reservoirs of the Yanchang Formation in the Ordos Basin. Lin Zhongxia et al. proposed using the well logging curve coverage area index, combined with the oil production before single-well production, to establish a regional production capacity evaluation model. However, no well logging production capacity prediction methods for volcanic reservoirs have been reported. Therefore, this invention aims to propose a scheme for predicting the production capacity of volcanic reservoirs using well logging data, to fill this technological gap.

[0060] Figure 1 This is a flowchart of a method for predicting the single-well productivity of volcanic rock reservoirs after compression, provided by one embodiment of the present invention. Figure 1 As shown, this invention provides a method for predicting the post-compression single-well productivity of volcanic reservoirs, the method comprising:

[0061] Step S10: Collect well logging information of the target prediction area, and classify the well logging information based on each preset evaluation index to obtain a well logging dataset corresponding to each preset evaluation index.

[0062] Specifically, before classifying the logging information based on each preset evaluation index, the method further includes performing preprocessing on the logging information. The preprocessing rules are as follows: performing irrelevant parameter filtering, missing value filling, outlier removal, and data normalization on the logging information in sequence.

[0063] Furthermore, the step of classifying the logging information based on each preset evaluation index to obtain a logging dataset corresponding to each preset evaluation index includes: determining the corresponding clustering target cluster based on each preset evaluation index, and randomly determining the initial centroid of each cluster; traversing the logging information, calculating the Euclidean distance between each data point and each centroid, and classifying the data point to the nearest centroid; after completing one round of classification, updating the centroid position based on the mean of the data points in each cluster; re-performing one round of clustering based on the updated centroid position, repeating the clustering for N rounds until the centroid of each cluster remains unchanged or the preset iteration round is reached, and determining the logging dataset for each preset evaluation index based on the latest clustering result.

[0064] In this embodiment of the invention, the solution first collects well logging information for the target prediction area and then performs data normalization during the preprocessing stage to ensure data quality and consistency. The preprocessing steps include irrelevant parameter filtering, missing value imputation, outlier removal, and data normalization. Irrelevant parameter filtering aims to remove well logging data that has little or no impact on the prediction model, improving computational efficiency and model accuracy. Missing value imputation uses appropriate interpolation methods or model inference to fill in missing values ​​in the well logging data, reducing the impact of incomplete data on the model. Outlier removal uses thresholds or statistical methods to remove extreme or unreasonable data points, avoiding interference with the prediction model. Data normalization adjusts various types of well logging data to the same dimension or range, ensuring that the distance metric in subsequent clustering calculations reflects the true differences between different types of data.

[0065] Further, after data preprocessing, the next step is to classify the logging information to obtain logging datasets corresponding to each preset evaluation index. Specifically, this step uses the K-means clustering algorithm to divide the logging data into multiple clusters based on the preset evaluation index. First, the number of target clusters corresponding to each preset evaluation index is determined, and the centroid position of each cluster is randomly initialized. Then, the logging information is traversed, and the Euclidean distance between each data point and each centroid is calculated. Each data point is classified into the corresponding cluster according to the principle of the shortest distance. After completing one round of data point classification, the mean of the data points in each cluster is calculated, and the position of the centroid is updated. Next, the clustering step is repeated based on the updated centroid. Through multiple iterations, the centroid no longer changes or the preset number of iterations is reached, and finally, the logging dataset corresponding to each preset evaluation index is determined.

[0066] This invention employs rigorous data preprocessing to ensure high-quality and consistent input data, eliminating interference from irrelevant factors and outliers, thereby improving the accuracy and effectiveness of classification. Simultaneously, the use of K-means clustering to classify different preset evaluation indicators enables the model to fully capture potential patterns and features in well logging data, effectively distinguishing different physical properties within the reservoir. This process not only simplifies the complex management of well logging information data but also provides more accurate data support for subsequent reservoir evaluation and production prediction.

[0067] Step S20: Based on the well logging dataset of each preset evaluation index, perform the index parameter calculation of the corresponding preset evaluation index to obtain the index parameters of each preset evaluation index.

[0068] Specifically, the preset evaluation indicators include any one or more of the following: lithological indicators, physical property indicators, electrical property indicators, and oil-bearing indicators; and any one or two of the following: fracture parameter indicators and brittleness parameter indicators. Among them, lithological indicators, physical property indicators, electrical property indicators, and oil-bearing indicators are related to reservoir quality indices, while fracture parameter indicators and brittleness parameter indicators are related to engineering quality indices.

[0069] Specifically, as the SiO2 content in volcanic rocks gradually increases, indicating a transition from basic to intermediate-acidic lithology, the natural gamma value gradually increases, while the neutron and density values ​​gradually decrease. Trachytic volcanic reservoirs exhibit significantly high natural gamma, low neutron, and low density characteristics. At a specific scale, the neutron-density curve displays a "mirror image," meaning the neutron-density envelope area gradually increases. Therefore, [the following is a possible interpretation / reference]:

[0070] Lithological indices are established using natural gamma, neutron, and density curves. The calculation rules for the corresponding lithological index parameters are as follows:

[0071]

[0072] Among them, I lith GR represents the index parameter of lithology; GR is the natural gamma ray logging curve value; GR 基值 1 is the natural gamma-ray value, AP; cn is the boundary value of the neutron logging curve; den is the boundary value of the density logging curve; DEN is the density logging curve value; CN is the neutron logging curve value; a and b are the neutron and density curve scale numbers, respectively.

[0073] In this embodiment of the invention, the natural gamma-ray value, i.e., the minimum GR value of the trachytic volcanic reservoir, is taken as 100 API based on the existing lithology identification standards within the Sha-3 Member of the Eastern Depression; the boundary values ​​of the neutron logging curve and the boundary values ​​of the density logging curve are expressed in % and g / cm³, respectively. 3 The unit of density logging curves is g / cm³.3 The unit of neutron logging curve values ​​is %. The larger the value of the lithological index parameter, the higher the SiO2 content. That is, when the lithology is trachyte, the value is positive, and when the lithology is basalt, the value is negative.

[0074] Furthermore, according to statistics, trachytic volcanic rock reservoirs with certain post-compression productivity generally possess dual reservoir spaces of pores and fractures. Porosity evaluation should assess both matrix porosity and fracture porosity separately. Acoustic transit-time curves reflect primary porosity; therefore, acoustic transit-time curves are used to evaluate matrix porosity. Fracture porosity is constructed using both deep and shallow lateral resistivity. The calculation rules for the corresponding physical property parameters are as follows:

[0075]

[0076] Among them, I por AC represents the physical property index parameter; AC is the acoustic transit time curve; AC 基值 The baseline value is the acoustic time difference baseline value; R LLd R LLs These represent the deep and shallow lateral resistivity, respectively; mf is the structural index; R mf The resistivity of mud filtrate under formation conditions.

[0077] In this embodiment of the invention, the unit of the acoustic transit time baseline is μs / ft; the unit of deep and shallow lateral resistivity is Ω·m; and the unit of the structural index, the resistivity of mud filtrate under formation conditions between 1.0 and 2.0, is Ω·m. The transit time baseline parameter is defined to eliminate the influence of dry layers in trachyte volcanic rocks, and this value is determined based on the distribution range of dry layer data points in the acoustic transit time-resistivity cross-plot. Taking an acoustic transit time-resistivity cross-plot of a trachyte breccia as an example, as follows... Figure 2 The acoustic transit time values ​​of the dry layer points during oil testing or production are all lower than the lower limit of acoustic transit time. Therefore, the AC base value for this block is taken as 61.5 μs / ft.

[0078] Furthermore, the resistivity curves of trachytic volcanic rock reservoirs often exhibit high values. Electrical properties are defined by establishing a baseline resistivity value, and the calculation rules for the corresponding electrical property parameters are as follows:

[0079]

[0080] Among them, I rt RT represents the electrical properties; RT is the formation resistivity curve; RT 基值 This is the base value for resistivity.

[0081] In this embodiment of the invention, the units for the formation resistivity curve and the base resistivity value are Ω·m. Taking the sonic transit time-resistivity cross-plot of a certain section of trachyte breccia as an example, as follows... Figure 2The lower limit of oil layer resistivity is selected as the base value of resistivity, that is, the base value of RT is 30Ω·m.

[0082] Furthermore, the total hydrocarbon value in gas logging can reflect the oil-bearing capacity of a reservoir to a certain extent. Therefore, an oil-bearing capacity index Io is defined, and the calculation rules for the corresponding index parameters are as follows:

[0083]

[0084] Among them, I o QT is the index parameter for oil content; QT is the total hydrocarbon curve value within the layer. min This represents the minimum value of the total hydrocarbon curve within the stratum; QT max This represents the maximum value of the total hydrocarbon curve within the stratum.

[0085] Furthermore, factors influencing reservoir engineering quality include fractures, brittleness, geostress differences, and rock fracture toughness. Among these, the presence of natural fractures is a favorable condition for volumetric fracturing; therefore, the degree of development of natural fractures (fracture density, fracture length, etc.) is an important indicator of engineering quality. Thus, an engineering quality index evaluation model is established using fractures as the primary parameter and considering formation brittleness parameters.

[0086] Specifically, fracture density is the total number of fractures observed per unit well section, and fracture length is the sum of fracture lengths observed per square meter of well wall. Fracture parameter indices are constructed based on these indices, and the calculation rules for the corresponding fracture parameter indices are as follows:

[0087] I frac =C×L;

[0088] Among them, I frac Here, C represents the index parameter of the crack parameter index; C is the normalized crack density; and L is the normalized crack length. The index parameter value of the crack parameter index can reflect the degree of crack development to a certain extent; the higher the value, the more developed the cracks.

[0089] Furthermore, rock brittleness refers to the property of plastic deformation that is not detected before fracturing, that is, the property of rocks to easily break under external forces (such as hydraulic fracturing). In the volumetric fracturing design of volcanic reservoirs, rock brittleness is one of the important factors to consider. This study uses dipole array sonic logging data to calculate brittleness parameters based on rock elastic parameters. The calculation rules for the corresponding brittleness parameter indices are as follows: Based on the logging dataset of brittleness parameter indices, calculate the Young's modulus and Poisson's ratio of the reservoir section in the target prediction area; perform normalization processing on the Young's modulus and Poisson's ratio respectively; and calculate the brittleness parameter indices based on the normalized Young's modulus and normalized Poisson's ratio.

[0090] Preferably, the calculation rules for Young's modulus and Poisson's ratio of the reservoir interval in the target prediction area based on the well logging dataset of brittle parameter indices are as follows:

[0091]

[0092] Where E is the Young's modulus of the reservoir section; DT p and DT s These are the wave time differences for the longitudinal and transverse waves, respectively; ρ b 1 represents the density logging curve; μ represents the Poisson's ratio of the reservoir section.

[0093] Furthermore, the rules for normalizing the Young's modulus and the Poisson's ratio are as follows:

[0094]

[0095] Among them, BI E E is the normalized Young's modulus; E is the Young's modulus of the reservoir section; E min E represents the minimum Young's modulus of the reservoir section. max BI represents the maximum Young's modulus of the reservoir section. PR The normalized Poisson's ratio is μ; μ is the Poisson's ratio of the reservoir section; μ min Minimum Poisson's ratio in the reservoir section; μ max The maximum Poisson's ratio of the reservoir section.

[0096] Furthermore, the calculation rules for the brittleness parameter index based on the normalized Young's modulus and the normalized Poisson's ratio are as follows:

[0097]

[0098] Among them, I brit The brittleness parameter is the index parameter; BI E The normalized Young's modulus; BI PR This is the normalized Poisson's ratio.

[0099] Step S30: Determine the reservoir quality index and engineering quality index of the target prediction area based on the index parameters of each preset evaluation index, and construct the corresponding comprehensive evaluation index based on the reservoir quality index and the engineering quality index.

[0100] Specifically, the reservoir quality index and engineering quality index of the target prediction area are determined based on the index parameters of each preset evaluation index. This includes: calculating the reservoir quality index of the target prediction area based on the index parameters of lithology, physical properties, electrical properties, and oil-bearing properties, as well as the reservoir quality index calculation model; and calculating the engineering quality index of the target prediction area based on the index parameters of fracture parameters and brittleness parameters, as well as the engineering quality index calculation model.

[0101] The preferred reservoir quality index calculation model is as follows:

[0102] I c =I lith ×I por ×I rt ×I o

[0103] Among them, I c The reservoir quality index for the target prediction region; I lith The index parameters for the lithological indices of the target prediction area; I por The index parameters of the physical properties of the target prediction area; I rt The index parameters for the electrical properties of the target prediction region; I o These are the index parameters for the oil-bearing properties of the target prediction area. A higher reservoir quality index indicates a better quality trachytic volcanic rock reservoir.

[0104] Furthermore, the calculation model for the engineering quality index is as follows:

[0105] Ig = I frac *I brit

[0106] Where Ig is the engineering quality index of the target prediction area; I frac The index parameters of the crack parameters in the target prediction area; I brit The brittleness parameter index of the target prediction region.

[0107] Preferably, the higher the value of the engineering quality index parameter, the better the engineering quality of the trachytic volcanic rock reservoir. It should be noted that not all wells have undergone electrical imaging and array sonic logging. Therefore, in specific calculations, when a well lacks electrical imaging or array sonic logging data, the corresponding fracture or brittleness parameter index is assigned a default value of 1.

[0108] Specifically, a comprehensive evaluation index for trachytic volcanic rock reservoirs is constructed by combining the reservoir quality index with the engineering quality index; the determination rules are as follows:

[0109] I q =IC *I g

[0110] Among them, I q I is the comprehensive reservoir evaluation index. C I is the reservoir quality index. g This is an engineering quality index, dimensionless. The higher the value, the better the quality and engineering quality of the trachytic volcanic rock reservoir, meaning better oil production capacity.

[0111] Step S40: Based on the pre-determined target prediction area's calibrated well productivity, construct a relationship model between the calibrated well productivity and the comprehensive evaluation index, and use the relationship model as a post-compression single-well productivity prediction model for volcanic rock reservoirs to perform single-well productivity prediction.

[0112] Specifically, the rule for determining the production capacity of a calibration well is as follows: within the target area, the production state information of multiple oil wells after fracturing is collected; based on the production state information, the production state of each corresponding well is identified; when the production state is identified as stable, the production state information of each oil well is re-collected under stable conditions; based on the re-collected production state information, the average oil production per unit time of each oil well is calculated, and the average oil production per unit time of each oil well is used as the production capacity of the calibration well.

[0113] In this embodiment of the invention, a relationship model is first constructed based on the relationship between the production capacity of calibrated wells and the comprehensive evaluation index in the target prediction area. Determining the production capacity of the calibrated wells is a crucial foundation for the prediction model. Specifically, firstly, production data after fracturing is collected from multiple oil wells within the target area, with particular attention paid to the fluid production status. Next, the collected fluid production information is analyzed to determine whether the fluid production status of each well has reached a stable state. When the fluid production status of a well is identified as stable, production data for all wells is collected again under this stable state. Subsequently, based on the production information collected under these stable states, the average oil production per unit time (e.g., within one day) of each well is calculated. Finally, these average values ​​are used as the production capacity of the calibrated wells for model construction.

[0114] Furthermore, the step of constructing a relationship model between the calibrated well productivity and the comprehensive evaluation index, and using the relationship model as a post-compression single-well productivity prediction model for volcanic reservoirs, includes: performing linear regression processing on the calibrated well productivity and the comprehensive evaluation index to obtain a linear relationship between the two; verifying the linear relationship based on a reserved validation dataset, and optimizing and adjusting it according to error indicators to obtain the final linear relationship model, and using the linear relationship model as a post-compression single-well productivity prediction model for volcanic reservoirs.

[0115] In this embodiment of the invention, the solution first performs linear regression processing based on the collected well productivity data and the comprehensive evaluation index of the reservoir. The purpose of linear regression is to find the optimal linear relationship between well productivity and the comprehensive evaluation index, thereby establishing a simple and effective mathematical model between the two. This model quantifies the relationship between reservoir characteristics and productivity by fitting the well productivity data. Next, the constructed linear relationship is validated using a reserved validation dataset. The validation dataset is a portion separated from the well logging and productivity data of the target area, specifically used to test the accuracy and generalization ability of the model. During the validation phase, error indices, such as mean squared error (MSE), are calculated by comparing the differences between the model's predicted values ​​and the actual measured values. If the error is large, the model can be optimized and adjusted based on these error indices to improve the model's fitting effect and ensure that it maintains high prediction accuracy under different production conditions.

[0116] Furthermore, once the optimization is complete and the final linear relationship model is obtained, this model can be used as a post-furlough single-well productivity prediction model for volcanic reservoirs. The model's application is not limited to the initial target area; it can also quickly predict future single-well production as logging data is updated in real time. In addition, the model is adaptable and can be extended to other volcanic reservoirs. Reliable productivity prediction can be achieved simply by adjusting the comprehensive evaluation index of the new reservoir accordingly.

[0117] In one possible implementation, production capacity prediction can be divided into production forecasting and production capacity index prediction (i.e., production capacity per unit thickness or production capacity per unit thickness and unit pressure difference). This invention primarily focuses on production forecasting, specifically using the actual daily production after fracturing to calibrate the reservoir comprehensive evaluation index, and then predicting the production capacity of the fracturing well section. The specific production capacity selection criterion is to select the production capacity of the calibrated well after the fluid production stabilizes during the post-fracturing blowout stage and the oil production stabilizes after water cut. A production capacity prediction model is established by cross-refractory analysis of the reservoir comprehensive evaluation index and the post-fracturing production capacity, such as... Figure 3 The corresponding model is:

[0118] Q = 5.448 * I q -15.606

[0119] Where Q is the daily production rate after well pressure treatment, m 3 / t, I q This is a dimensionless comprehensive reservoir evaluation index.

[0120] Figure 4 This is a system structure diagram of a post-compression single-well productivity prediction system for volcanic reservoirs provided in one embodiment of the present invention. Figure 4As shown, this invention provides a post-compression single-well productivity prediction system for volcanic reservoirs. The system includes: a data acquisition unit for acquiring well logging information of a target prediction area and classifying the well logging information based on various preset evaluation indicators to obtain well logging datasets corresponding to each preset evaluation indicator; a processing unit for performing indicator parameter calculations based on the well logging datasets of each preset evaluation indicator to obtain indicator parameters for each preset evaluation indicator; a comprehensive evaluation unit for determining the reservoir quality index and engineering quality index of the target prediction area based on the indicator parameters of each preset evaluation indicator, and constructing a corresponding comprehensive evaluation index based on the reservoir quality index and the engineering quality index; wherein the reservoir quality index is used to characterize the reservoir characteristics, and the engineering quality index is used to characterize the geological characteristics of the reservoir under engineering conditions; and a prediction unit for constructing a relationship model between the calibrated well productivity and the comprehensive evaluation index based on the pre-determined calibrated well productivity of the target prediction area, so as to use the relationship model as a post-compression single-well productivity prediction model for volcanic reservoirs and perform single-well productivity prediction.

[0121] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described method for predicting the single-well productivity of volcanic reservoirs after compression.

[0122] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a microcontroller, chip, or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0123] The optional embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details described above. Within the scope of the technical concept of the embodiments of the present invention, various simple modifications can be made to the technical solutions of the embodiments of the present invention, and these simple modifications all fall within the protection scope of the embodiments of the present invention. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the embodiments of the present invention will not further describe the various possible combinations.

[0124] Furthermore, various different embodiments of the present invention can be combined in any way, as long as they do not violate the spirit of the embodiments of the present invention, they should also be regarded as the content disclosed by the embodiments of the present invention.

Claims

1. A method for predicting single well deliverability of a volcanic reservoir after fracturing, characterized in that, The method comprises: collecting well logging information of a target prediction area, and classifying the well logging information based on each preset evaluation index to obtain a well logging data set corresponding to each preset evaluation index; based on the well logging data set of each preset evaluation index, performing index parameter calculation corresponding to the preset evaluation index to obtain an index parameter of each preset evaluation index; based on the index parameter of each preset evaluation index, determining a reservoir quality index and an engineering quality index of the target prediction area, and constructing a corresponding comprehensive evaluation index based on the reservoir quality index and the engineering quality index; wherein, the reservoir quality index is used to represent the reservoir characteristics of the reservoir, and the engineering quality index is used to represent the geological characteristics of the reservoir under engineering conditions; based on a predetermined scale well productivity of the target prediction area, a relationship model between the scale well productivity and the comprehensive evaluation index is constructed, so as to take the relationship model as a single well productivity prediction model of the volcanic reservoir after pressure, and perform single well productivity prediction.

2. The method of claim 1, wherein, Before classifying the well logging information based on each preset evaluation index, the method further comprises performing preprocessing on the well logging information, and the preprocessing rule is: the well logging information is sequentially subjected to irrelevant parameter filtering processing, missing value filling processing, abnormal value removal processing and data normalization processing.

3. The method of claim 1, wherein, The well logging information is classified based on each preset evaluation index to obtain a well logging data set corresponding to each preset evaluation index, which comprises: based on each preset evaluation index, a corresponding clustering target cluster is determined, and the initial centroid of each cluster is randomly determined; the well logging information is traversed, the Euclidean distance between each data point and each centroid is calculated, and the data point is classified into the nearest centroid; after one round of classification is completed, the centroid position is updated based on the mean value of the data points of each cluster; based on the updated centroid position, a round of clustering is re-performed, and the clustering is repeatedly performed for N rounds until the cluster centroid is unchanged or a preset iteration round is reached, and the well logging data set of each preset evaluation index is determined based on the latest clustering result.

4. The method of claim 1, wherein, The preset evaluation index comprises: any one or more of a lithology index, a physical property index, an electrical property index and an oil-bearing property index; and any one or both of a fracture parameter index and a brittleness parameter index.

5. The method of claim 4, wherein, The index parameter calculation rule of the lithology index is: wherein I lith is an index parameter of lithology index; GR is the natural gamma ray logging curve value; GR 基值 GR is the natural gamma base value, AP. cn is the neutron logging curve calibration boundary value; den is the density logging curve calibration boundary value; DEN is the density logging curve value; CN is the neutron logging curve value; a and b are the neutron and density curve calibration numbers respectively.

6. The method of claim 4, wherein, The index parameter calculation rule of the physical property index is: wherein I por is an index parameter of a physical property index; AC is the acoustic time difference curve; AC 基值 The base value is the acoustic travel time base value. R LLd , R LLs are the deep and shallow lateral resistivities, respectively; mf is the structure index; R mf Rf is the formation fluid resistivity.

7. The method of claim 4, wherein, The index parameter calculation rule of the electrical property index is: wherein I rt is an index parameter of the electrical index; RT is the formation resistivity curve; RT 基值 is the resistivity base value.

8. The method of claim 4, wherein, The index parameter calculation rule of the oil-bearing property index is: I o is an index parameter of an oiliness index; QT is the gas logging total hydrocarbon curve value in the interval; QT min is the minimum value of the gas logging total hydrocarbon curve value in the interval QT max The maximum value of the gas logging total hydrocarbon curve in the interval.

9. The method of claim 4, wherein, The index parameter calculation rule of the fracture parameter index is: I frac = C x L; where I frac is an index parameter for the crack parameter index; C is the normalized fracture density; L is the normalized fracture length.

10. The method of claim 4, wherein, The index parameter calculation rule of the brittleness parameter index is: the Young's modulus and the Poisson's ratio of the reservoir interval of the target prediction area are calculated based on the well logging data set of the brittleness parameter index; the Young's modulus and the Poisson's ratio are subjected to normalization processing respectively; The brittle parameter index is calculated based on the normalized Young's modulus and the normalized Poisson's ratio.

11. The method of claim 10, wherein, The calculation rules of the Young's modulus and the Poisson's ratio of the reservoir section of the target prediction area based on the brittle parameter index are as follows: E is the Young's modulus of the reservoir section; DT p and DT s are the travel times of the longitudinal and transverse waves, respectively; p b Density log; DEX; DEXY; DEXYI; DEXYII; DEXYIII; μ is the Poisson's ratio of the reservoir section.

12. The method of claim 10, wherein, The normalization rules of the Young's modulus and the Poisson's ratio are as follows: wherein BI E is the normalized Young's modulus; E is the Young's modulus of the reservoir section; E min E is the minimum value of the Young's modulus of the reservoir section; E max E is the maximum Young's modulus of the reservoir section; BI PR Pn is the normalized Poisson's ratio; μ is the Poisson's ratio of the reservoir section. μ min minimum Poisson's ratio of the reservoir interval; μ max Maximum Poisson's ratio of the reservoir interval.

13. The method of claim 10, wherein, The calculation rules of the brittle parameter index based on the normalized Young's modulus and the normalized Poisson's ratio are as follows: wherein I brit is an index parameter of the brittleness parameter index; BI E E is the normalized Young's modulus; BI PR is the normalized Poisson's ratio.

14. The method of claim 4, wherein, The reservoir quality index and the engineering quality index of the target prediction area are determined based on the index parameters of each preset evaluation index, including: The reservoir quality index of the target prediction area is calculated based on the index parameters of the lithology index, the index parameters of the physical property index, the index parameters of the electrical property index, and the index parameters of the oil-bearing property index, and a reservoir quality index calculation model; The engineering quality index of the target prediction area is calculated based on the index parameters of the fracture parameter index and the index parameters of the brittle parameter index, and an engineering quality index calculation model.

15. The method of claim 14, wherein, The reservoir quality index calculation model is as follows: I c = I lith x I por x I rt x I o where I c is the reservoir quality index of the target prediction region; I lith index parameters for lithology indicators of the target prediction region; I por an index parameter for a property index of a target prediction region; I rt an index parameter for predicting an electrical property of a region of interest; I o An index parameter that targets an index of oiliness of the prediction region.

16. The method of claim 14, wherein, The engineering quality index calculation model is as follows: Ig = I frac * I brit Ig is the engineering quality index of the target prediction area; I frac an indicator parameter of a crack parameter indicator of the target prediction region; I brit an indicator parameter of a target prediction region's brittleness parameter indicator.

17. The method of claim 1, wherein, The determination rules of the scale well productivity are as follows: In the target area, the production fluid state information of multiple production wells after fracturing is obtained; Based on the production fluid state information, the production fluid state of each well is identified, and when the production fluid state is identified as a stable state, the production fluid state information of each production well is re-collected in the stable state; Based on the re-collected production fluid state information, the average oil production per unit time of each production well is calculated, and the average oil production per unit time of each production well is taken as the scale well productivity.

18. The method of claim 1, wherein, The relationship model between the scale well productivity and the comprehensive evaluation index is constructed, and the relationship model is taken as a post-fracturing single-well productivity prediction model for volcanic reservoirs, including: Linear regression processing is performed on the scale well productivity and the comprehensive evaluation index to obtain their linear relationship; The linear relationship is verified based on a reserved verification data set, and is optimized and adjusted according to an error index to obtain a final linear relationship model, which is taken as a post-fracturing single-well productivity prediction model for volcanic reservoirs.

19. A system for predicting single well deliverability of a volcanic reservoir after compaction, the system comprising: The system includes: A collection unit is configured to collect logging information of a target prediction area, and classify the logging information based on each preset evaluation index to obtain logging data sets corresponding to each preset evaluation index; A processing unit is configured to calculate index parameters corresponding to each preset evaluation index based on the logging data sets of each preset evaluation index; A comprehensive evaluation unit is configured to determine a reservoir quality index and an engineering quality index of the target prediction area based on the index parameters of each preset evaluation index, and construct a corresponding comprehensive evaluation index based on the reservoir quality index and the engineering quality index; wherein The reservoir quality index is used to represent the reservoir characteristics of the reservoir, and the engineering quality index is used to represent the geological characteristics of the reservoir under engineering conditions; The prediction unit is configured to construct a relationship model between the scale well productivity and the comprehensive evaluation index based on the scale well productivity of a predetermined target prediction area, and to execute single well productivity prediction by taking the relationship model as a single well productivity prediction model of the volcanic reservoir after pressure.

20. A computer-readable storage medium, characterized in that, The computer readable storage medium stores instructions, which, when executed on the computer, cause the computer to execute the method for predicting single well productivity of the volcanic reservoir after pressure according to any one of claims 1-18.

Images