A low-permeability oil reservoir productivity prediction method based on an elliptical horizontal slit

Abstract

The application discloses a low-permeability oil reservoir productivity prediction method based on an elliptical horizontal fracture, and belongs to the field of oil reservoir productivity prediction, and comprises the following steps: S1, constructing a standardized parameter data set; S2, constructing an elliptical horizontal fracture single-phase percolation physical model and setting a boundary condition; S3, obtaining a dynamic productivity calculation model at different production stages; S4, obtaining an elastic development continuous productivity dynamic sequence under a multi-fracture condition; S5, establishing a dissolved gas drive material balance model, and obtaining an elastic development full-cycle productivity prediction result; S6, obtaining a productivity and water cut dynamic prediction result under a water injection development mode; S7, obtaining a final productivity prediction model suitable for a target oil reservoir; and S8, quantifying the influence weight of each parameter on the productivity, and outputting an optimal fracturing construction parameter interval and a development engineering parameter interval. The low-permeability oil reservoir productivity prediction method based on the elliptical horizontal fracture can simultaneously realize quantitative optimization of fracturing construction parameters and water injection development systems.

Description

Technical Field

[0001] This invention relates to the field of reservoir productivity prediction technology, and in particular to a method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures. Background Technology

[0002] Low-permeability sandstone reservoirs are characterized by a large proportion of reserves, abundant resource potential, ultra-low permeability, and low porosity. These reservoirs are tight, have extremely poor matrix permeability, and very low natural production capacity; without artificial intervention, they essentially have no natural production. Currently, the most efficient approach for developing these low-permeability reservoirs is fracturing, which involves creating artificial fractures to connect the tight reservoir, expand the drainage area, improve permeability channels, and achieve effective reservoir production.

[0003] Artificial fractures formed after in-situ fracturing are not ideal circular or radially equivalent fracture morphologies. Instead, they are generally elliptical horizontal fractures, controlled by differences in reservoir stress, lithology, and physical properties. Furthermore, reservoir development sequentially progresses through three major stages: elastic exploitation, dissolved gas drive two-phase exploitation, and water injection development. Accurate characterization of the seepage mechanism of elliptical horizontal fractures, quantification of multi-fracture interference, degassing phase transformation, and the coupling effects of heterogeneous water channeling are crucial for accurately predicting the productivity of low-permeability fracturing reservoirs and guiding the optimization of fracturing parameters and the control of water injection regimes. This has significant engineering implications for improving the development effect of tight, low-permeability reservoirs and increasing oil recovery.

[0004] Currently, the industry has developed various numerical and analytical models for predicting the production capacity of fractured low-permeability reservoirs, which can basically achieve dynamic prediction of the production capacity of fractured wells. However, existing technologies are generally suitable for conventional circular fractures, homogeneous reservoirs, and single development conditions, but cannot adapt to the complex geological development characteristics of elliptical horizontal fracture morphology, multi-fracture pressure interference, low-permeability non-Darcy flow, strong reservoir heterogeneity, and multiple development mode coupling. The prediction accuracy is low and the engineering guidance is poor. Summary of the Invention

[0005] The purpose of this invention is to provide a method for predicting the productivity of low-permeability oil reservoirs based on elliptical horizontal fractures, thereby solving the aforementioned technical problems.

[0006] To achieve the above objectives, this invention provides a method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures, comprising the following steps: S1. Collect multi-dimensional basic parameters of the target low-permeability reservoir and construct a standardized parameter dataset; among which, the basic parameters include reservoir physical property parameters, fluid physical property parameters, fracturing parameters and development engineering parameters; S2. Based on the standardized parameter dataset output by S1, construct a single-phase seepage physical model of an elliptical horizontal fracture and set boundary conditions. Derive the elliptical radial seepage sub-model in the horizontal direction and the linear seepage sub-model in the vertical direction respectively. Couple the two sub-models through the horizontal fracture node to obtain the steady-state productivity equation of a single elliptical horizontal fracture considering reservoir anisotropy and starting pressure gradient. S3. Based on the steady-state productivity equation of a single elliptical horizontal fracture output by S2, the skin coefficient and wellbore reservoir coefficient are introduced to construct an unsteady seepage mathematical model of the elliptical horizontal fracture. The Romberg numerical integration method is used to solve the dimensionless pressure response. Five seepage stages are divided into fracture radial flow, vertical linear flow, transition flow, hemispherical flow and reservoir radial flow to obtain dynamic productivity calculation models for different production stages. S4. Based on the dynamic production capacity calculation model output by S3, and combined with the number of fractures and longitudinal distribution parameters in the fracturing fracture parameters output by S1, a multi-horizontal fracture flow field coupling and pressure disturbance calculation model is constructed. Through flow field superposition and pressure drop correction, the dynamic sequence of continuous production capacity under multi-fracture conditions is obtained. S5. Based on the continuous production capacity dynamic sequence output by S4, combined with the standardized parameter dataset of S1, a dissolved gas drive material balance model is established to calculate the reservoir degassing characteristics after the formation pressure drops to the saturation pressure. The crude oil viscosity and relative permeability curve parameters are corrected to obtain the full-cycle production capacity prediction results for flexible development. S6. For the water injection development mode, based on the elliptical horizontal fracture seepage model framework output by S2, and combined with the injection-production well network, injection-production layer correspondence and reservoir heterogeneity parameters in the development engineering parameters output by S1, a horizontal fracture injection-production coupled seepage model is constructed. The type C water drive curve method is introduced to calculate the volumetric sweep efficiency, and the influence of reservoir heterogeneity and injection-production parameters on water channeling is coupled to obtain the dynamic prediction results of production capacity and water cut under the water injection development mode. S7. Based on the full-cycle production capacity prediction results of the flexible development output of S5 or the dynamic prediction results of production capacity and water cut under the water injection development mode output of S6, and combined with the actual production dynamic data of the target reservoir, the historical fitting inversion method is used to correct the model parameters of fracture conductivity, reservoir effective permeability, and permeability variation coefficient, so as to obtain the final production capacity prediction model adapted to the target reservoir. S8, based on the final production capacity prediction model output by S7, conducts multi-factor sensitivity analysis on fracture half-length, fracture shape factor, vertical permeability to horizontal permeability ratio, start-up pressure gradient, injection-production ratio, and injection pressure parameters, quantifies the influence weight of each parameter on production capacity, and outputs the optimal fracturing construction parameter range and development engineering parameter range.

[0007] Therefore, the present invention employs the above-mentioned method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures, which has the following beneficial effects: 1. Significantly improved accuracy in seepage solution: The Romberg high-precision integral combined with the Stefest Laplace inversion is used to solve the dimensionless pressure response of unsteady seepage. Based on the double logarithmic curve, five types of seepage stages are accurately divided, and a dedicated dynamic production capacity model is provided for each stage. The model fully considers the bidirectional starting pressure gradient of low-permeability reservoirs, reservoir anisotropy, and the geometric characteristics of elliptical fractures. Compared with the conventional well test model for circular fractures, the model closely matches the field fracturing morphology, and the production capacity calculation error in the early weak seepage stage is significantly reduced. 2. Achieve quantitative characterization of multi-fracture coupling interference: Introduce dynamic inter-fracture production reduction coefficient to quantify the production capacity reduction caused by the superposition of pressure fields of multiple elliptical horizontal fractures, differentiate between weak interference of two fractures and strong interference of three or more fractures, solve the defects of traditional single fracture model direct superposition without considering the overlap of oil drainage zone and pressure gradient attenuation, and make the dynamic prediction results of production capacity in the early stage of elastic development of multi-fracture fractured wells more consistent with the measured production data. 3. Covering the entire life cycle of reservoir development and production capacity calculation: Integrating three mainstream development modes: pure elastic drive, dissolved gas drive, two-phase flow, and water injection development, it quantitatively characterizes the dynamic changes in crude oil viscosity and phase permeability parameters caused by formation degassing through the material balance equation. Combined with the C-type water drive curve, it quantifies the volumetric sweep efficiency and couples the reservoir heterogeneous water channeling correction. It can output the production capacity and water cut time series prediction results of the entire process from elastic production to water injection development in one go, without the need to change the independent calculation model in segments. 4. Possesses reservoir adaptive correction capability: Based on actual on-site oil production and water cut dynamic data, a multi-objective error function is constructed, and the three core sensitive parameters of fracture conductivity, reservoir effective permeability, and permeability variation coefficient are iteratively inverted and corrected. In addition, the rationality verification of geological property boundaries is added to eliminate the empirical bias of the initial input parameters and obtain a production capacity prediction model that is adapted to the target block, with stronger cross-block applicability. 5. Quantitative guidance for on-site engineering optimization: Based on the calibrated accurate model, single-factor sensitivity analysis of six key parameters is carried out. The influence weight of each parameter on production capacity is quantified by the dimensionless sensitivity coefficient. The optimal fracturing construction parameter range and water injection development regime range are output respectively, taking into account multiple development goals such as stabilizing oil, controlling water, and delaying production capacity decline, and directly providing quantitative design basis for fracturing design and injection-production control. 6. Closed-loop computational logic and strong versatility: The entire method is progressively interconnected, with the output data of the preceding steps fully supplied as input to the subsequent model. The parameter system is unified and self-consistent, and all equations and algorithms are constructed around the mechanism of low-permeability elliptical horizontal fractured reservoirs. It takes into account steady-state, unsteady-state, single-phase, and two-phase flow scenarios, and is suitable for industrial-scale production capacity prediction of low-permeability fractured reservoirs.

[0008] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0009] Figure 1 This is a flowchart of a method for predicting the productivity of low-permeability oil reservoirs based on elliptical horizontal fractures, according to the present invention. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0011] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, system, product, or server that includes a series of steps or units, not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0012] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0013] like Figure 1 As shown, a method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures includes the following steps: S1. Collect multi-dimensional basic parameters of the target low-permeability reservoir and construct a standardized parameter dataset; among which, the basic parameters include reservoir physical property parameters, fluid physical property parameters, fracturing parameters and development engineering parameters.

[0014] S2. Based on the standardized parameter dataset output by S1, construct a single-phase seepage physical model of an elliptical horizontal fracture and set boundary conditions. Derive the elliptical radial seepage sub-model in the horizontal direction and the linear seepage sub-model in the vertical direction respectively. Couple the two sub-models through the horizontal fracture node to obtain the steady-state productivity equation of a single elliptical horizontal fracture considering reservoir anisotropy and starting pressure gradient.

[0015] S3. Based on the steady-state productivity equation of a single elliptical horizontal fracture output by S2, the skin coefficient and wellbore reservoir coefficient are introduced to construct an unsteady seepage mathematical model of the elliptical horizontal fracture. The Romberg numerical integration method is used to solve the dimensionless pressure response. Five seepage stages are divided into fracture radial flow, vertical linear flow, transition flow, hemispherical flow and reservoir radial flow to obtain dynamic productivity calculation models for different production stages.

[0016] S4. Based on the dynamic production capacity calculation model output by S3, and combined with the number of fractures and longitudinal distribution parameters in the fracturing fracture parameters output by S1, a multi-horizontal fracture flow field coupling and pressure disturbance calculation model is constructed. Through flow field superposition and pressure drop correction, the dynamic sequence of continuous production capacity under multi-fracture conditions is obtained.

[0017] S5. Based on the continuous production capacity dynamic sequence output by S4, combined with the standardized parameter dataset of S1, a dissolved gas drive material balance model is established to calculate the reservoir degassing characteristics after the formation pressure drops to saturation pressure. The crude oil viscosity and relative permeability curve parameters are corrected to obtain the full-cycle production capacity prediction results for flexible development.

[0018] S6. For the water injection development mode, based on the elliptical horizontal fracture seepage model framework output by S2, and combined with the injection-production well network, injection-production layer correspondence and reservoir heterogeneity parameters in the development engineering parameters output by S1, a horizontal fracture injection-production coupled seepage model is constructed. The type C water drive curve method is introduced to calculate the volumetric sweep efficiency, and the influence of reservoir heterogeneity and injection-production parameters on water channeling is coupled to obtain the dynamic prediction results of production capacity and water cut under the water injection development mode.

[0019] S7. Based on the full-cycle production capacity prediction results of the flexible development output of S5 or the dynamic prediction results of production capacity and water cut under the water injection development mode output of S6, and combined with the actual production dynamic data of the target reservoir, the historical fitting inversion method is used to correct the model parameters of fracture conductivity, reservoir effective permeability, and permeability variation coefficient, so as to obtain the final production capacity prediction model adapted to the target reservoir.

[0020] S8, based on the final production capacity prediction model output by S7, conducts multi-factor sensitivity analysis on fracture half-length, fracture shape factor, vertical permeability to horizontal permeability ratio, start-up pressure gradient, injection-production ratio, and injection pressure parameters, quantifies the influence weight of each parameter on production capacity, and outputs the optimal fracturing construction parameter range and development engineering parameter range.

[0021] The reservoir physical property parameters mentioned in step S1 include basic reservoir physical property parameters, reservoir heterogeneity parameters, and reservoir sensitivity analysis parameters; wherein, the basic reservoir physical property parameters include horizontal permeability. Vertical permeability Effective porosity Original oil saturation Bound water saturation Effective reservoir thickness Original formation pressure and saturation pressure Reservoir heterogeneity parameters include the permeability variation coefficient. Permeability surge coefficient and penetration rate difference It is used to quantify the degree of vertical and planar heterogeneity of the reservoir; the reservoir sensitivity analysis parameters include water-sensitive, rate-sensitive, acid-sensitive, alkali-sensitive and salt-sensitive damage rates, which are used to constrain injection processes and stimulation measures.

[0022] Fluid properties include high-pressure crude oil properties and formation water and natural gas properties; among which, high-pressure crude oil properties include formation crude oil viscosity. Density of degassed crude oil on the ground Formation crude oil volume coefficient Original dissolved gas-oil ratio and crude oil compressibility Formation water and natural gas physical properties include formation water density. Formation water volume coefficient Formation water compressibility coefficient Natural gas volume coefficient .

[0023] Fracturing parameters include the geometric parameters of a single fracture and the distribution parameters of multiple fractures; among them, the geometric parameters of a single fracture include the semi-major axis of the elliptical fracture. short half shaft Crack height , fracture conductivity and epidermal coefficient The multi-fracture distribution parameters include the total number of elliptical horizontal fractures in a single well, the longitudinal spacing of the fractures, and the longitudinal distribution layers of the fractures.

[0024] Development engineering parameters include well pattern and well spacing parameters, and injection-production parameters; among which, well pattern and well spacing parameters include well pattern type (reverse nine-point, five-point, row-like) and injection-production well spacing. The relationship between injection and production layers; injection and production parameters include injection pressure of injection wells, daily injection volume, injection-production ratio, bottom-hole flowing pressure of production wells, daily oil production, water cut, cumulative oil production, and cumulative water production.

[0025] The above-collected multi-dimensional basic parameters are subjected to outlier removal, linear interpolation, and unit unification operations in sequence to form a standardized parameter dataset.

[0026] Step S2 specifically includes the following steps: S21. Based on the standardized parameter dataset output by S1, construct a physical model of single-phase seepage in an elliptical horizontal joint, and set the following boundary conditions: The reservoir fluid is a single-phase steady flow, and the horizontal fracture cross-section is elliptical.

[0027] The reservoir is symmetrically distributed above and below the horizontal fractured fracture.

[0028] The reservoir outer boundary pressure is constant. .

[0029] Consider reservoir anisotropy and initiation pressure gradient.

[0030] We can ignore the effects of gravity and capillary forces.

[0031] Ignoring pressure drop losses within the horizontal fractured area, the fracture possesses infinite conductivity, and the fracture pressure equals the bottom hole flowing pressure. .

[0032] S22. Derive the horizontal elliptical radial seepage sub-model.

[0033] S221, Considering horizontal starting pressure gradient Establish the horizontal non-Darcy flow control equations and set the inner and outer boundary conditions: ; Internal boundary condition: The pressure at the boundary of the elliptical fracture is equal to the bottom hole flowing pressure. The semi-major axis of the ellipse is The minor half-axis is .

[0034] External boundary conditions: The pressure at the outer boundary of the elliptical oil drain region is... The long half-axis of the oil drain area is The minor half-axis is .

[0035] In the formula: Two-dimensional location of the reservoir plane The pressure in the formation.

[0036] By combining the S222 and S221 plane non-Darcy flow control equations with the inner and outer boundary conditions of the elliptical fracture, the two-dimensional formation pressure field in the plane can be obtained. Substituting the pressure gradient into the non-Darcy flow velocity equation with a horizontal starting pressure gradient, and integrating along the elliptical fracture boundary, we obtain the flow rate of the matrix flowing horizontally towards the elliptical horizontal fracture. : ; In the formula: The equivalent path length for horizontal seepage; This refers to the viscosity of the crude oil in the formation.

[0037] S23. Derive the vertical linear seepage sub-model.

[0038] S231. Establish the vertical one-dimensional non-Darcy seepage control equations and the outer and inner boundary conditions: ; External boundary conditions: The pressure at the upper and lower boundaries of the reservoir is ; Internal boundary conditions: The pressure at the upper and lower surfaces of the horizontal seam is ; In the formula: The vertical starting pressure gradient; Vertical permeability of the reservoir; The coordinates are in the vertical thickness direction of the reservoir.

[0039] Solving the vertical one-dimensional non-Darcy flow control equations of S232 and S231, along with the vertical upper and lower inner and outer boundary conditions of the reservoir, yields the vertical formation pressure distribution. Substituting the vertical pressure gradient into the non-Darcy seepage velocity equation with a vertical initiation pressure gradient, and integrating along the longitudinal surface of the horizontal joint, the flow rate of the matrix flowing vertically towards the elliptical horizontal joint is obtained. : ; S24. By coupling the horizontal elliptical radial seepage sub-model and the vertical linear seepage sub-model at the horizontal joint node, the steady-state production capacity equation of a single elliptical horizontal joint is constructed: ; In the formula: This represents the total daily oil production of the oil well.

[0040] The coupling condition is as follows: based on the infinite conductivity assumption of S21, the pressure inside the horizontal fracture is uniformly equal to the bottom hole flowing pressure. Total oil production is the sum of the horizontal flow rate and the vertical flow rate.

[0041] Step S3 specifically includes the following steps: S31, Introducing the epidermal coefficient This corrects for the additional pressure drop at the bottom of the well, characterizing the effect of fracture wall contamination or increased production; simultaneously, it introduces the wellbore reservoir coefficient. To correct the wellbore storage effect and characterize the influence of fluid elastic compression within the wellbore on early pressure.

[0042] S32. Establish the mathematical model and boundary conditions for unsteady seepage in an elliptical horizontal joint: ; Inner boundary conditions: When producing at a constant output, ; External boundary conditions: for infinitely large strata, ; In the formula: Dimensionless formation pressure; All are elliptic coordinate components; For Laplace variables; These are the elliptical coordinates corresponding to the elliptical crack.

[0043] S33. Solve the dimensionless pressure response using the Romberg numerical integration method.

[0044] S331. Obtain the trapezoidal integral sequence through recursion using variable step size: ; In the formula: For the first The integration step size of the next recursion; For the Laplace space integrand function of seepage through an elliptical horizontal slit; This is the lower limit of integration; For the first Approximate value of the composite trapezoidal integral corresponding to the next recursion; For the first Approximate value of the composite trapezoidal integral corresponding to the next recursion; For discrete integral node counting variables; This represents the total number of newly added encrypted points nodes in this round.

[0045] S332, First-order extrapolation: Based on two adjacent sets of trapezoidal integral sequences Perform Richardson extrapolation and eliminate Fourth-order error, generating a fourth-order accurate Simpson integral sequence: ; In the formula: For the first The high-precision Simpson integral value.

[0046] S333, Based on two adjacent sets of Simpson sequences Secondary extrapolation, elimination Order error, generating a sixth-order accurate Cortes integral sequence: ; In the formula: For the first High-precision integral values ​​from the Cortes; For the first The high-precision Simpson integral value.

[0047] S334, Based on two adjacent sets of Cortes sequences Ultimate extrapolation, elimination Order error, generating eighth-order ultra-high precision Romberg integral results: ; In the formula: For the first The results of the Strömberg ultra-high precision integration; For the first High-precision integral values ​​from the Tescos system.

[0048] S335. After obtaining the Laplace dimensionless pressure analytical solution in the complex frequency domain through Romberg integration, the Stefest numerical inversion algorithm is used to convert the complex frequency domain solution into the real time domain dynamic pressure response. : ; In the formula: This refers to dimensionless production time. The order of truncation for the inversion of discrete terms; These are the Stefest weighting coefficients; The result is the Romberg integral, i.e., the dimensionless pressure solution of the elliptic horizontal slit in the Laplace complex frequency domain. These are the sampling points for discrete Laplace complex variables.

[0049] S336, Continuous-time domain dimensionless pressure response sequence based on the output of S335 The dimensionless pressure derivative is calculated using the general logarithmic difference method for well testing, eliminating time-scale interference and accurately identifying the characteristics of weak seepage stages. ; In the formula: The derivative of pressure is dimensionless; It is the first derivative of dimensionless pressure with respect to dimensionless time.

[0050] S337. Construct a dimensionless pressure-pressure derivative double logarithmic curve dataset: Take the common logarithm of dimensionless time, dimensionless pressure, and dimensionless pressure derivative respectively to obtain three sets of mapping data: x-axis: , dimensionless logarithmic time value.

[0051] Vertical axis 1: , dimensionless logarithmic pressure value.

[0052] Vertical axis 2: The logarithmic value of the dimensionless pressure derivative.

[0053] S34. Based on the dimensionless pressure-pressure derivative double logarithmic curve obtained in S33, five seepage stages are divided and a dynamic production capacity calculation model for each seepage stage is derived.

[0054] Radial flow stage in fractures: In the early stage of seepage, flow occurs within the fractures, and the slope of the dimensionless pressure-pressure derivative double logarithmic curve is 1, corresponding to elastic storage and radial flow within the fractures; and the instantaneous dynamic production capacity expression for this stage is as follows: ; In the formula: For the dimensionless dynamic production capacity of the radial flow stage of the crack; The radial flow stage of the crack represents the dimensionless pressure change rate.

[0055] Vertical linear flow stage: Fluid in the matrix flows vertically into the fracture from above and below the horizontal fracture. The slope of the dimensionless pressure-pressure derivative double logarithmic curve is 1 / 2, and the length of the straight segment is limited by the reservoir thickness and fracture size. The instantaneous dynamic productivity expression for this stage is as follows: ; In the formula: For the dimensionless dynamic capacity of the vertical linear flow stage; Let be the dimensionless time at which the linear flow begins.

[0056] Transitional flow stage: The seepage flow gradually transitions from vertically dominant to horizontally dominant, and a transition segment appears in the dimensionless pressure-pressure derivative double logarithmic curve; and the instantaneous dynamic production capacity expression for this stage is as follows: ; In the formula: The dimensionless dynamic capacity during the transition phase; The transition weight coefficients are dynamically decaying over dimensionless time, and , This is the real-time calculated value of the dimensionless pressure derivative at the current moment. The steady-state dimensionless pressure derivative value for the vertical linear flow stage. This represents the steady-state dimensionless pressure derivative during the radial flow stage of the crack.

[0057] Hemispherical flow stage: This stage occurs when the reservoir is relatively thick and the fracture half-length is relatively short. Fluid flows into the fracture from both vertical and horizontal directions, and the dimensionless pressure-pressure derivative double logarithmic curve exhibits a decreasing slope of -1 / 2. The instantaneous dynamic productivity expression for this stage is as follows: ; In the formula, The dimensionless dynamic capacity of the hemispherical flow stage; The diffusion correction coefficient for the hemispherical flow field is obtained by fitting the characteristic of the -1 / 2 slope of the dimensionless pressure-pressure derivative double logarithmic curve.

[0058] Reservoir radial flow stage: In the later stage of seepage, radial flow forms throughout the entire reservoir. The dimensionless pressure-time semi-logarithmic coordinates show a linear segment, and the semi-logarithmic linear formula is: ; In the formula: This is the dimensionless bottom hole pressure.

[0059] Furthermore, the instantaneous dynamic capacity expression for this stage is as follows: ; In the formula: This is to ensure stable dimensionless dynamic production capacity during the radial flow stage of the reservoir.

[0060] S35. The segmented dynamic production capacity models of the five seepage stages in S34 are continuously spliced ​​together to eliminate the abrupt error of stage connection and form a complete dynamic production capacity calculation model that covers the entire production cycle of the reservoir and is adapted to the five-stage seepage characteristics of the elliptical horizontal fracture.

[0061] Step S4 specifically includes the following steps: S41. Divide the seepage of a single horizontal joint into three stages; Initial vertical linear flow stage: Only the matrix fluid in the vertical direction above and below the horizontal crack overcomes the vertical initiation pressure gradient and flows vertically into the elliptical horizontal crack in a parallel linear manner. The flow field morphology is regular and there is no radial diffusion characteristic.

[0062] Radial flow stage perpendicular to the fracture surface: The fluid flows radially into the wellbore with the elliptical fracture as the center, perpendicular to the fracture end face, forming a radial flow field of the fracture. The flow field is transformed from vertical linear flow to radial flow in the fracture plane.

[0063] Full-area pressure drop funnel seepage stage: The horizontal and vertical seepage of the reservoir are fully coupled, forming a full-area elliptical pressure drop funnel centered on the horizontal fractured wellbore, and the seepage enters the radial flow stage of the stable reservoir.

[0064] S42. Based on the total number of elliptical horizontal fractures, longitudinal spacing of fractures, and longitudinal stratigraphic distribution parameters of fractures collected in S1, determine the intensity of interference from multiple fracture superpositions and calculate the additional pressure drop loss caused by inter-fracture superposition: ; In the formula: Dimensionless time The additional pressure drop caused by inter-slit interference is an extra pressure loss caused by the superposition of multiple slits; Dimensionless time The average formation pressure under single-fracture independent seepage conditions; Dimensionless time The average regional formation pressure is superimposed under multi-slit coupled seepage conditions.

[0065] S43. Based on the inter-fracture additional pressure drop solved in S42, construct a dynamic productivity reduction factor suitable for low-permeability reservoirs with elliptical horizontal fractures: ; In the formula: Dimensionless time The corresponding dynamic inter-seam capacity reduction coefficient has a value range of (0,1). The stronger the interference, the smaller the value, and the greater the capacity reduction. This represents the effective displacement pressure differential of the reservoir under single-fracture independent seepage conditions.

[0066] S44. Based on the differentiated interference mechanism of double-slit and multi-slit systems, the benchmark single-slit capacity is dynamically reduced and corrected. The formula for the dynamic capacity after single-slit correction is as follows: ; In the formula: For the first under multi-slit interference conditions The dimensionless dynamic production capacity after correction of an elliptical horizontal crack; For the first The dynamic production capacity of the non-interference benchmark for cracks.

[0067] S45. The corrected single-crack capacity of all longitudinally distributed cracks is time-series superimposed to construct the multi-crack coupled total dynamic capacity sequence for the elastic development stage: ; In the formula: The dimensionless total dynamic production capacity of oil wells under multi-slit coupling interference conditions constitutes the dynamic sequence of initial production capacity for flexible development. This represents the total number of elliptical horizontal fractures in a single well.

[0068] S46. Through full-domain time-series iterative calculation, a continuous dynamic sequence of production capacity is obtained, covering the entire stage of elastic development and considering the inter-gap pressure interference and flow field superposition effect.

[0069] Step S5 specifically includes the following steps: S51. Retrieve the continuous production capacity dynamic sequence output by S4, and simultaneously read the reservoir physical property parameters collected by S1 to distinguish between the purely elastic drive stage and the dissolved gas drive stage, thereby achieving staged modeling: ; In the formula: This represents the current average formation pressure.

[0070] S52. Based on the comprehensive compressibility characteristics of reservoir rocks and fluids, an elastic displacement material balance model is established under pressure higher than saturation pressure to characterize the relationship between formation pressure and cumulative oil production in the elastic stage, and to obtain the comprehensive overall compressibility coefficient of the formation.

[0071] The formula for calculating the mass balance of elastic drive is as follows: ; The formula for calculating the overall compressibility coefficient of a formation is as follows: ; In the formula: This represents the cumulative oil production of the reservoir. This represents the original geological reserves of crude oil in the reservoir. This represents the crude oil volume factor under the original formation pressure. The overall compressibility coefficient of the reservoir; The compression coefficient of crude oil; The rock skeleton compression coefficient; This represents the original formation pressure of the oil reservoir.

[0072] S53. When the formation pressure falls below the saturation pressure, dissolved gas continuously precipitates from the crude oil, and the reservoir changes from single-phase flow to two-phase flow of oil and gas. Based on the characteristics of low-permeability oil reservoirs with no gas cap and weak edge and bottom water, a dissolved gas drive material balance model is established: ; ; In the formula: This represents the combined volumetric coefficient of oil and gas under the current pressure. This is the combined volumetric coefficient of oil and gas under saturated pressure, and it is equivalent to the volumetric coefficient of crude oil under saturated pressure. ; To calculate the cumulative gas-oil ratio; The original dissolved gas-oil ratio under saturated pressure; This represents the real-time dissolved gas-oil ratio under the current formation pressure. This represents the volume coefficient of natural gas under the current pressure.

[0073] Given the current cumulative oil production of the reservoir Cumulative production of gas-oil ratio , United and ,by The process iterates repeatedly with the only unknown variable in the iteration. After convergence, the real-time dissolved gas-oil ratio corresponding to the current formation pressure is output, thus obtaining the full-time dissolved gas-oil ratio decay sequence.

[0074] Based on the real-time dissolved gas-oil ratio obtained through iteration, and combined with the original differences in dissolved gas endowment and pore volume conservation, the real-time free gas saturation of the reservoir after pressure drop and dissolved gas precipitation is calculated: ; In the formula: This represents the real-time gas saturation of the formation under the current formation pressure. This represents the cumulative volume of free gas released per cubic meter of crude oil.

[0075] S54. Based on the real-time dissolved gas-oil ratio decay law solved in S53, the decrease in formation pressure and the precipitation of dissolved gas will lead to a significant increase in crude oil viscosity. Combining the degassing viscosity characteristics of low-permeability reservoirs, a dynamic correction model for crude oil viscosity is established to replace the initial constant viscosity parameter and realize dynamic updating of physical properties.

[0076] The dynamic correction expression for crude oil viscosity is as follows: ; In the formula: This represents the dynamic crude oil viscosity under the current formation pressure. The base viscosity of crude oil under saturated pressure conditions; It is the viscosity sensitivity coefficient for degassing in low-permeability reservoirs.

[0077] S55. Based on the real-time formation gas saturation generated by degassing, dynamically update the relative permeability of the oil and gas phases: ; In the formula: Real-time dynamic relative permeability of the oil phase; Real-time dynamic relative permeability of the gas phase; The relative permeability of the original oil phase during the initial single-phase seepage stage; The initial relative permeability of the gas phase when the initial free gas has not yet developed is approximately 0. Real-time formation gas saturation obtained for the degassing evolution of S53; The maximum gas saturation level confined within the reservoir; , , are the permeability variation indices of the oil phase and gas phase, respectively, and the fitting constants for the physical properties of low-permeability reservoirs.

[0078] S56. Based on the multi-slit coupled continuous production dynamic sequence output by S4, a production prediction model is constructed in stages: In the pure elastic drive stage, the formation pressure is higher than the saturation pressure, and there is no degassing or two-phase flow effect, so the production dynamic sequence of S4 is directly used; In the dissolved gas drive stage, the formation pressure is lower than the saturation pressure, and the dynamic correction coefficient of crude oil viscosity from S54 and the real-time dynamic relative permeability of the oil phase from S55 are substituted, and the production capacity decay rate is corrected by coupling the dissolved gas expansion and energy replenishment characteristics, and the final full-cycle production capacity is calculated: ; in, ; ; In the formula: To adjust the dynamic capacity for full-cycle flexible development; This is the continuous dynamic sequence of production capacity output by S4; This is the dynamic correction coefficient for crude oil viscosity. Degassing and increased viscosity result in a coefficient less than 1, leading to reduced production capacity. This is a correction factor for two-phase flow capacity, which decreases continuously with increasing gas saturation. This serves as the energy supply coefficient for dissolved gas drive, provides elastic energy for free gas expansion, and mitigates the decline in production capacity. This represents the viscosity of the benchmark crude oil under saturated pressure.

[0079] Step S6 specifically includes the following steps: S61, a single-phase seepage model framework for elliptical horizontal fractures constructed based on S2, retaining the geometry of the elliptical fractures, reservoir anisotropy, initiation pressure gradient, and the core mechanism of horizontal-vertical coupled seepage; retrieving the injection-production well pattern type and injection-production well spacing collected in S1. Correspondence between injection and production layers, and coefficient of variation of permeability. Penetration rate differences And injection and production parameters.

[0080] S62. Based on the S2 elliptical horizontal fracture single-phase seepage model framework and combined with the actual injection-production correspondence, the seepage evolution law of the three stages of water injection development is divided, and an injection-production coupled dynamic seepage model adapted to fracturing low-permeability reservoirs is established: In the early stage of development without water seepage: the injected water has not yet affected the artificial fractures, and the reservoir fluid only undergoes vertical linear seepage from the matrix to the elliptical horizontal fractures and radial seepage from the fractures. The water cut of the oil well is 0, and the production capacity is maintained by the formation elastic energy and the initial water injection energy.

[0081] During the mid-stage of development, the water injection energy continuously replenishes the formation pressure, and the water drive front gradually advances towards the elliptical fracture. The reservoir simultaneously exhibits vertical matrix flow, radial planar flow, and water drive propulsion flow, with stable displacement of the oil and water phases and a slow increase in water cut.

[0082] Late-stage water channeling: Due to reservoir heterogeneity and high-pressure water injection, elliptical fractures and high-permeability bands form dominant water flow channels, resulting in fracture water channeling, obvious water drive surge characteristics, rapid increase in water cut, and rapid decrease in oil well productivity.

[0083] S63. The classic C-type water drive curve of the oilfield is used to characterize the water drive law of the whole process of water injection development. The curve intercept and slope parameters are obtained by fitting production dynamic data. The classic C-type water drive curve expression for oilfields is as follows: ; In the formula: This represents the cumulative liquid production of the reservoir. The intercept of the fitting curve for the classic Type C waterflooding curve in the oilfield characterizes the initial waterflooding difficulty of the reservoir. The slope of the fitted curve for the classic Type C water drive in the oilfield represents the rate of decrease in water cut and production capacity during the later stages of water drive.

[0084] S64. Based on the curve intercept and slope parameters obtained from the S63 fitting, and combined with the formation crude oil volume factor and geological reserves, a quantitative calculation formula for water cut and volumetric sweep efficiency is established: ; In the formula: Moisture content The reservoir volume sweep efficiency is used to characterize the proportion of reservoir volume effectively displaced by water injection.

[0085] S65. Combining the reservoir heterogeneity parameters and water injection engineering parameters output from S1, and considering the deviations in productivity and water cut caused by non-uniform water drive and fracture water channeling, correct the dynamic productivity and water cut: ; ; in, ; ; ; In the formula: The dynamic production capacity during the water injection stage after multi-factor correction; This refers to the uncorrected basic water injection capacity. This is the corrected moisture content; Predict water cut based on classic type C water drive curves for oilfields; The dynamic volumetric sweep efficiency is calculated after correction for reservoir heterogeneity. The water injection sensitivity coefficient for low-permeability elliptical fractured reservoirs; This refers to the water cross-contamination correction factor for injection and production parameters; Inject pressure to average the formation; The steady-state injection-production ratio at the site; This is the correction coefficient for injection-production layer matching; This is a layer connectivity correction constant; The effective vertical overlap thickness of the injection and production layers.

[0086] S66. Based on the phased seepage characteristics of water injection development, establish a segmented exponentially decreasing capacity model to match the production pattern of rapid reduction from high capacity in the early stage and slow reduction from stable water injection in the middle and later stages: Early stage of rapid capacity reduction: ; Production capacity in the mid-to-late stage of slow decline: ; In the formula: These represent the daily oil production of oil wells in the early and middle-to-late stages, respectively. These represent the initial production capacity for the two production phases. These represent the production decline rates in the early and middle / late stages, respectively. This refers to the time required for water injection development and production.

[0087] By combining S67, S66 segmented decreasing basic capacity, S64 time-varying water-drive volumetric sweep efficiency, and S65 multi-dimensional water channeling heterogeneity correction coefficient, explicit prediction formulas for dynamic capacity and water cut throughout the entire water injection development cycle are established. The final outputs are the dynamic sequence of capacity and dynamic change sequence of water cut throughout the entire water injection development cycle. ; ; in, ; In the formula: for Final production capacity at any given moment; To reduce basic production capacity in stages; for Dynamic volumetric sweep efficiency after reservoir heterogeneity correction at all times; for Moisture content at any given time; for Predicted water cut based on classic C-type water drive curve in Shike Oilfield.

[0088] Step S7 specifically includes the following steps: S71. Based on the actual development method of the reservoir, select the calculation results of the previous model as the fitting basis: in the elastic development stage, retrieve the full-cycle production capacity prediction sequence output by S5, and in the water injection development stage, retrieve the dynamic prediction sequence of production capacity and water cut output by S6; simultaneously collect the time-series production data measured on-site in the target reservoir, including actual daily oil production, actual water cut, production time series, etc., and construct a historical fitting sample dataset.

[0089] S72. Establish a multi-parameter coupled error objective function with the goal of minimizing the deviation between model predictions and actual field values: ; In the formula: To account for the overall fitting error, the smaller the value, the higher the model fitting accuracy; The production weighting coefficient represents the proportion of the production error in the overall fit. The moisture content fitting weight coefficient represents the proportion of the moisture content error in the overall fitting. For the first Time-based models predict daily oil production; For the first Real-time daily oil production of the reservoir; For the first Moisture content is predicted by a time-based model; For the first Real-time reservoir water cut during production; This represents the total number of time-series samples in the historical fitted data.

[0090] S73. Combining the seepage control mechanism of low-permeability elliptical horizontal fracture reservoirs and the dynamic response characteristics of water injection development, three types of sensitive parameters to be corrected—those with the highest sensitivity to dynamic changes in well productivity and water cut, the greatest on-site uncertainty, and the most significant contribution to modeling error—were selected as the sole iterative correction variables: fracture conductivity. Effective reservoir permeability and permeability variation coefficient .

[0091] All other reservoir physical parameters, fracture geometric parameters, fluid high-pressure physical parameters, and static geological parameters are fixed to the measured values ​​of S1 and S2 and do not participate in the inversion iteration, ensuring the uniqueness of the parameter inversion solution, the rationality of the physical geology, and the convergence stability of the model.

[0092] S74. Using the error objective function minimization iterative algorithm, iteratively optimize the three types of sensitive parameters to be corrected in S73 until the desired result is achieved. , For the first Round iteration comprehensive error; For the first Round iteration comprehensive error; This is the preset fitting accuracy threshold.

[0093] After iterative convergence, the corrected optimal parameter set is output: corrected fracture conductivity, corrected reservoir effective permeability, and corrected permeability variation coefficient.

[0094] S75. Combining reservoir geological characteristics and the range of fracturing parameters, the optimal parameter set obtained by inversion in S74 is checked for boundary constraints: the parameter values ​​fall within the reservoir physical property range, the fracture modification characteristic range, and the heterogeneous distribution range.

[0095] S76. Substitute the optimal parameters that meet the boundary constraints back into the single-phase seepage physical model of the elliptical horizontal fracture in S2, the calculation model of flow field coupling and pressure disturbance of multiple horizontal fractures in S4, the dissolved gas drive material balance model in S5, and S6. Replace the initial empirical parameters, complete the calibration of the full-link model parameters, and obtain the final production capacity prediction model that is adapted to the geological characteristics, fracturing stimulation characteristics, and production dynamics of the target reservoir.

[0096] Step S8 specifically includes the following steps: S81. Using the final production capacity prediction model after historical fitting correction in S7 as the basis for numerical simulation, and combining the controllable parameters of fracturing construction, reservoir dynamic parameters, and water injection development engineering parameters, the following sensitivity analysis parameters are determined: fracture half-length, fracture shape factor, vertical to horizontal permeability ratio, initiation pressure gradient, injection-production ratio, and injection pressure.

[0097] S82. Based on the measured basic parameters of S1, and combined with the feasible range of on-site fracturing construction and water injection development project, upper and lower limit fluctuation ranges are set for each analysis parameter, and multiple sets of gradient values ​​are evenly set within the upper and lower limit fluctuation ranges.

[0098] S83. Keeping the other parameters constant, only change the gradient value of the sensitivity analysis parameter, substitute it into the final production capacity prediction model of S7, and calculate the steady-state production capacity, cumulative oil production, production capacity deceleration rate and water cut increase rate of the reservoir under different sensitivity analysis parameter values. Obtain the dynamic response law of production capacity corresponding to the independent change of each sensitivity analysis parameter, and obtain the positive and negative influence trend and change range of the single parameter on production capacity.

[0099] S84. Calculate the parameter sensitivity coefficient: ; In the formula: For the first The capacity sensitivity coefficient of the parameter, the larger the absolute value, the more significant the parameter's impact on capacity; This represents the change in capacity forecast indicators. The initial production capacity is based on the baseline parameters. For the first The change in the values ​​of the sensitivity analysis parameters; For the first The baseline initial values ​​for the sensitivity analysis parameters.

[0100] S85. Based on the absolute values ​​of the production capacity sensitivity coefficients of each parameter calculated in S84, the influence weights of fracture half-length, fracture shape factor, vertical to horizontal permeability ratio, start-up pressure gradient, injection-production ratio, and injection pressure are ranked in order of level.

[0101] Furthermore, it distinguishes between promoting and inhibiting parameters: a positive sensitivity coefficient indicates that increasing the parameter is beneficial to improving production capacity; a negative sensitivity coefficient indicates that increasing the parameter will exacerbate the decline in production capacity and the increase in water content.

[0102] S86. Based on the sensitivity analysis results, and combining the objectives of maximizing production capacity, minimizing production capacity decline, and maximizing the stable production period, a comprehensive optimization objective function for fracturing parameters is constructed: ; In the formula: To optimize the target values ​​for fracturing parameters; , , These are the weighting coefficients for capacity level, capacity deceleration rate, and length of stable production period, respectively. This represents the average effective production capacity corresponding to the current combination of fracturing parameters; The baseline production capacity corresponds to the initial baseline fracturing parameters; The initial rate of decline in production capacity under the baseline fracturing parameters; This represents the overall capacity reduction rate under the current fracturing parameters; This represents the effective stable production period length corresponding to the current fracturing parameters; This represents the baseline stable production period length corresponding to the baseline fracturing parameters.

[0103] Iterate through all gradient values ​​of fracture half-length, fracture shape factor, and fracturing stimulation strength parameters, and calculate each group sequentially. and take The largest interval is taken as the optimal interval for fracturing parameters: ; In the formula: and These represent the lower and upper limits of the optimal range of values ​​for a single type of fracturing parameter, respectively. The preset optimal parameter selection threshold is used.

[0104] S87. Combining the suppression of water channeling in fractures, reduction of the negative impact of heterogeneity, and ensuring the stable termination of water drive replacement energy, a comprehensive optimization objective function for water injection parameters is constructed: ; Constraints: ; In the formula: The target values ​​for comprehensive optimization of water injection parameters; , , These are the weighting coefficients for production capacity maintenance, water control effectiveness, and resistance to water channeling, respectively. This represents the average stable production capacity corresponding to the current water injection parameters; Base production capacity under benchmark water injection parameters; This represents the average moisture content over the production cycle. The minimum allowable water channeling resistance threshold; and These are the upper and lower limits of the injection pressure engineering, respectively; Average injection pressure; and These are the lower and upper limits of the reasonable range for the injection-production ratio, respectively. This represents the steady-state injection-production ratio.

[0105] By iterating through the values ​​of injection pressure and injection-production ratio gradient, the parameter range corresponding to the maximum value of the objective function is taken as the optimal range of water injection development parameters.

[0106] S88. Summarize the sensitivity weight ranking results of each parameter, the influence law of single parameter, and the coupling and adaptation relationship of multiple parameters, and finally output two sets of core optimal parameter ranges: one is the optimal range of fracturing parameters that adapts to the reservoir seepage characteristics, and the other is the optimal range of water injection development parameters that adapts to water injection to stabilize oil and control water.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures, characterized in that: Includes the following steps: S1. Collect multi-dimensional basic parameters of the target low-permeability reservoir and construct a standardized parameter dataset; among which, the basic parameters include reservoir physical property parameters, fluid physical property parameters, fracturing parameters and development engineering parameters; S2. Based on the standardized parameter dataset output by S1, construct a single-phase seepage physical model of an elliptical horizontal fracture and set boundary conditions. Derive the elliptical radial seepage sub-model in the horizontal direction and the linear seepage sub-model in the vertical direction respectively. Couple the two sub-models through the horizontal fracture node to obtain the steady-state productivity equation of a single elliptical horizontal fracture considering reservoir anisotropy and starting pressure gradient. S3. Based on the steady-state productivity equation of a single elliptical horizontal fracture output by S2, the skin coefficient and wellbore reservoir coefficient are introduced to construct an unsteady seepage mathematical model of the elliptical horizontal fracture. The Romberg numerical integration method is used to solve the dimensionless pressure response. Five seepage stages are divided into fracture radial flow, vertical linear flow, transition flow, hemispherical flow and reservoir radial flow to obtain dynamic productivity calculation models for different production stages. S4. Based on the dynamic production capacity calculation model output by S3, and combined with the number of fractures and longitudinal distribution parameters in the fracturing fracture parameters output by S1, a multi-horizontal fracture flow field coupling and pressure disturbance calculation model is constructed. Through flow field superposition and pressure drop correction, the dynamic sequence of continuous production capacity under multi-fracture conditions is obtained. S5. Based on the continuous production capacity dynamic sequence output by S4, combined with the standardized parameter dataset of S1, a dissolved gas drive material balance model is established to calculate the reservoir degassing characteristics after the formation pressure drops to the saturation pressure. The crude oil viscosity and relative permeability curve parameters are corrected to obtain the full-cycle production capacity prediction results for flexible development. S6. For the water injection development mode, based on the elliptical horizontal fracture seepage model framework output by S2, and combined with the injection-production well network, injection-production layer correspondence and reservoir heterogeneity parameters in the development engineering parameters output by S1, a horizontal fracture injection-production coupled seepage model is constructed. The type C water drive curve method is introduced to calculate the volumetric sweep efficiency, and the influence of reservoir heterogeneity and injection-production parameters on water channeling is coupled to obtain the dynamic prediction results of production capacity and water cut under the water injection development mode. S7. Based on the full-cycle production capacity prediction results of the flexible development output of S5 or the dynamic prediction results of production capacity and water cut under the water injection development mode output of S6, and combined with the actual production dynamic data of the target reservoir, the historical fitting inversion method is used to correct the model parameters of fracture conductivity, reservoir effective permeability, and permeability variation coefficient, so as to obtain the final production capacity prediction model adapted to the target reservoir. S8, based on the final production capacity prediction model output by S7, conducts multi-factor sensitivity analysis on fracture half-length, fracture shape factor, vertical permeability to horizontal permeability ratio, start-up pressure gradient, injection-production ratio, and injection pressure parameters, quantifies the influence weight of each parameter on production capacity, and outputs the optimal fracturing construction parameter range and development engineering parameter range.

2. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 1, characterized in that: The reservoir physical property parameters mentioned in step S1 include basic reservoir physical property parameters, reservoir heterogeneity parameters, and reservoir sensitivity analysis parameters; wherein, the basic reservoir physical property parameters include horizontal permeability. Vertical permeability Effective porosity Original oil saturation Bound water saturation Effective reservoir thickness Original formation pressure and saturation pressure Reservoir heterogeneity parameters include the permeability variation coefficient. Permeability surge coefficient and penetration rate difference It is used to quantify the degree of vertical and planar heterogeneity of the reservoir; the reservoir sensitivity analysis parameters include water-sensitive, rate-sensitive, acid-sensitive, alkali-sensitive and salt-sensitive damage rates, which are used to constrain the injection process and stimulation measures; Fluid properties include high-pressure crude oil properties and formation water and natural gas properties; among which, high-pressure crude oil properties include formation crude oil viscosity. Density of degassed crude oil on the ground Formation crude oil volume coefficient Original dissolved gas-oil ratio and crude oil compressibility Formation water and natural gas physical properties include formation water density. Formation water volume coefficient Formation water compressibility coefficient Natural gas volume coefficient ; Fracturing parameters include the geometric parameters of a single fracture and the distribution parameters of multiple fractures; among them, the geometric parameters of a single fracture include the semi-major axis of the elliptical fracture. short half shaft Crack height , fracture conductivity and epidermal coefficient Multi-fracture distribution parameters include the total number of elliptical horizontal fractures in a single well, the longitudinal spacing of the fractures, and the longitudinal distribution of the fractures across different strata. Development engineering parameters include well pattern and well spacing parameters and injection-production parameters; among which, well pattern and well spacing parameters include well pattern type and injection-production well spacing. Correspondence between injection and production layers; injection and production parameters include injection pressure of injection wells, daily injection rate, injection-production ratio, bottomhole flowing pressure of production wells, daily oil production, water cut, cumulative oil production, and cumulative water production; The above-collected multi-dimensional basic parameters are subjected to outlier removal, linear interpolation, and unit unification operations in sequence to form a standardized parameter dataset.

3. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 2, characterized in that: Step S2 specifically includes the following steps: S21. Based on the standardized parameter dataset output by S1, construct a physical model of single-phase seepage in an elliptical horizontal joint, and set the following boundary conditions: The reservoir fluid is a single-phase steady flow, and the horizontal fracture cross-section is elliptical. The reservoir is symmetrically distributed vertically around the horizontal fractured fractures. The reservoir outer boundary pressure is constant. ; Considering reservoir anisotropy and initiation pressure gradient; Neglecting the effects of gravity and capillary forces; Ignoring pressure drop losses within the horizontal fractured area, the fracture possesses infinite conductivity, and the fracture pressure equals the bottom hole flowing pressure. ; S22. Derive the horizontally elliptical radial seepage sub-model; S221, Considering horizontal starting pressure gradient Establish the horizontal non-Darcy flow control equations and set the inner and outer boundary conditions: ； Internal boundary condition: The pressure at the boundary of the elliptical fracture is equal to the bottom hole flowing pressure. The semi-major axis of the ellipse is The minor half-axis is ; External boundary conditions: The pressure at the outer boundary of the elliptical oil drain region is... The long half-axis of the oil drain area is The minor half-axis is ; In the formula: Two-dimensional location of the reservoir plane Formation pressure; By combining the S222 and S221 plane non-Darcy flow control equations with the inner and outer boundary conditions of the elliptical fracture, the two-dimensional formation pressure field in the plane can be obtained. Substituting the pressure gradient into the non-Darcy flow velocity equation with a horizontal starting pressure gradient, and integrating along the elliptical fracture boundary, we obtain the flow rate of the matrix flowing horizontally towards the elliptical horizontal fracture. : ； In the formula: The equivalent path length for horizontal seepage; The viscosity of the formation crude oil; S23. Derive the vertical linear seepage sub-model; S231. Establish the vertical one-dimensional non-Darcy seepage control equations and the outer and inner boundary conditions: ； External boundary conditions: The pressure at the upper and lower boundaries of the reservoir is ; Internal boundary conditions: The pressure at the upper and lower surfaces of the horizontal seam is ; In the formula: The vertical starting pressure gradient; Vertical permeability of the reservoir; The coordinates are in the vertical thickness direction of the reservoir. Solving the vertical one-dimensional non-Darcy flow control equations of S232 and S231, along with the vertical upper and lower inner and outer boundary conditions of the reservoir, yields the vertical formation pressure distribution. Substituting the vertical pressure gradient into the non-Darcy seepage velocity equation with a vertical initiation pressure gradient, and integrating along the longitudinal surface of the horizontal joint, the flow rate of the matrix flowing vertically towards the elliptical horizontal joint is obtained. : ； S24. By coupling the horizontal elliptical radial seepage sub-model and the vertical linear seepage sub-model at the horizontal joint node, the steady-state production capacity equation of a single elliptical horizontal joint is constructed: ； In the formula: This represents the total daily oil production of the oil well. The coupling condition is as follows: based on the infinite conductivity assumption of S21, the pressure inside the horizontal fracture is uniformly equal to the bottom hole flowing pressure. Total oil production is the sum of the horizontal flow rate and the vertical flow rate.

4. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 3, characterized in that: Step S3 specifically includes the following steps: S31, Introducing the epidermal coefficient This corrects for the additional pressure drop at the bottom of the well, characterizing the effect of fracture wall contamination or increased production; simultaneously, it introduces the wellbore reservoir coefficient. To correct the wellbore storage effect and characterize the influence of fluid elastic compression within the wellbore on early pressure; S32. Establish the mathematical model and boundary conditions for unsteady seepage in an elliptical horizontal joint: ； Inner boundary conditions: When producing at a constant output, ; External boundary conditions: for infinitely large strata, ; In the formula: Dimensionless formation pressure; All are elliptic coordinate components; For Laplace variables; These are the elliptical coordinates corresponding to the elliptical crack. S33. Solve the dimensionless pressure response using the Romberg numerical integration method; S331. Obtain the trapezoidal integral sequence through recursion using variable step size: ； In the formula: For the first The integration step size of the next recursion; For the Laplace space integrand function of seepage through an elliptical horizontal slit; This is the lower limit of integration; For the first Approximate value of the composite trapezoidal integral corresponding to the next recursion; For the first Approximate value of the composite trapezoidal integral corresponding to the next recursion; For discrete integral node counting variables; This represents the total number of newly added encrypted points nodes in this round; S332, First-order extrapolation: Based on two adjacent sets of trapezoidal integral sequences Perform Richardson extrapolation and eliminate Fourth-order error, generating a fourth-order accurate Simpson integral sequence: ； In the formula: For the first Simpson high-precision integral value; S333, Based on two adjacent sets of Simpson sequences Secondary extrapolation, elimination Order error, generating a sixth-order accurate Cortes integral sequence: ； In the formula: For the first High-precision integral values ​​from the Cortes; For the first Simpson high-precision integral value; S334, Based on two adjacent sets of Cortes sequences Ultimate extrapolation, elimination Order error, generating eighth-order ultra-high precision Romberg integral results: ； In the formula: For the first The results of the Strömberg ultra-high precision integration; For the first High-precision integral values ​​from the Cortes; S335. After obtaining the Laplace dimensionless pressure analytical solution in the complex frequency domain through Romberg integration, the Stefest numerical inversion algorithm is used to convert the complex frequency domain solution into the real time domain dynamic pressure response. : ； In the formula: This refers to dimensionless production time. The order of truncation for the inversion of discrete terms; These are the Stefest weighting coefficients; The result of Romberg's integral; For discrete Laplace complex variables, sampling points are used. S336, Continuous-time domain dimensionless pressure response sequence based on the output of S335 The dimensionless pressure derivative was calculated using the general logarithmic difference method for well testing: ； In the formula: The pressure derivative is dimensionless. This is the first derivative of dimensionless pressure with respect to dimensionless time. S337. Construct a dimensionless pressure-pressure derivative double logarithmic curve dataset: Take the common logarithm of dimensionless time, dimensionless pressure, and dimensionless pressure derivative respectively to obtain three sets of mapping data: x-axis: ; Vertical axis 1: ; Vertical axis 2: ; S34. Based on the dimensionless pressure-pressure derivative double logarithmic curve obtained in S33, five seepage stages are divided and a dynamic production capacity calculation model for each seepage stage is derived. Radial flow stage in the fracture: Flow occurs within the fracture, and the slope of the dimensionless pressure-pressure derivative double logarithmic curve is 1, corresponding to elastic storage and radial flow within the fracture; the instantaneous dynamic capacity expression for this stage is as follows: ； In the formula: For the dimensionless dynamic production capacity of the radial flow stage of the crack; The rate of change of dimensionless pressure during the radial flow stage of the crack; Vertical linear flow stage: Fluid in the matrix flows vertically into the fracture from above and below the horizontal fracture, and the slope of the dimensionless pressure-pressure derivative double logarithmic curve is 1 / 2; and the instantaneous dynamic production capacity expression for this stage is as follows: ； In the formula: The dimensionless dynamic capacity of the vertical linear flow stage; The dimensionless time at the start of the linear flow; Transitional flow stage: The seepage flow gradually transitions from vertically dominant to horizontally dominant, and a transition segment appears in the dimensionless pressure-pressure derivative double logarithmic curve; and the instantaneous dynamic production capacity expression for this stage is as follows: ； In the formula: The dimensionless dynamic capacity during the transition phase; The transition weight coefficients are dynamically decaying over dimensionless time, and , This is the real-time calculated value of the dimensionless pressure derivative at the current moment. The steady-state dimensionless pressure derivative value for the vertical linear flow stage. The steady-state dimensionless pressure derivative during the radial flow stage of the crack; Hemispherical flow stage: Fluid flows into the crack from both vertical and horizontal directions, and the dimensionless pressure-pressure derivative double logarithmic curve exhibits a decreasing slope of -1 / 2; and the instantaneous dynamic production capacity expression for this stage is as follows: ； In the formula, The dimensionless dynamic capacity of the hemispherical flow stage; This is the diffusion correction factor for the hemispherical flow field; Reservoir radial flow stage: Radial flow forms throughout the entire reservoir, and the dimensionless pressure-time semi-logarithmic coordinates show a straight line segment, with the semi-logarithmic linear formula as follows: ； In the formula: The dimensionless bottom hole pressure; Furthermore, the instantaneous dynamic capacity expression for this stage is as follows: ； In the formula: To achieve stable dimensionless dynamic productivity during the radial flow stage of the reservoir; S35. The segmented dynamic production capacity models of the five seepage stages in S34 are continuously spliced ​​together to eliminate the abrupt error of stage connection and form a complete dynamic production capacity calculation model that covers the entire production cycle of the reservoir and is adapted to the five-stage seepage characteristics of the elliptical horizontal fracture.

5. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 4, characterized in that: Step S4 Specifically, the following steps are included: S41. Divide the seepage of a single horizontal joint into three stages; Initial vertical linear flow stage: Only the matrix fluid in the vertical direction above and below the horizontal crack overcomes the vertical initiation pressure gradient and flows vertically into the elliptical horizontal crack in a parallel linear manner, and the flow field morphology is regular and has no radial diffusion characteristics. Radial flow stage perpendicular to the fracture surface: The fluid flows radially into the wellbore with the elliptical fracture as the center, perpendicular to the fracture end face, forming a radial flow field of the fracture. The flow field is transformed from vertical linear flow to radial flow in the fracture plane. Full-domain pressure drop funnel seepage stage: The horizontal and vertical seepage of the reservoir are fully coupled, forming a full-domain elliptical pressure drop funnel centered on the horizontal fractured wellbore, and the seepage enters the radial flow stage of the stable reservoir; S42. Based on the total number of elliptical horizontal fractures, longitudinal spacing of fractures, and longitudinal stratigraphic distribution parameters of fractures collected in S1, determine the intensity of interference from multiple fracture superpositions and calculate the additional pressure drop loss caused by inter-fracture superposition: ； In the formula: Dimensionless time Additional pressure drop due to inter-slit interference; Dimensionless time The average formation pressure under single-fracture independent seepage conditions; Dimensionless time The average regional formation pressure is superimposed under multi-slit coupled seepage conditions. S43. Based on the inter-fracture additional pressure drop solved in S42, construct a dynamic productivity reduction factor suitable for low-permeability reservoirs with elliptical horizontal fractures: ； In the formula: Dimensionless time The corresponding dynamic inter-seam capacity reduction factor; This represents the effective displacement pressure differential of the reservoir under single-fracture independent seepage conditions. S44. Based on the differentiated interference mechanism of double-slit and multi-slit systems, the benchmark single-slit capacity is dynamically reduced and corrected. The formula for the dynamic capacity after single-slit correction is as follows: ； In the formula: For the first under multi-slit interference conditions The dimensionless dynamic production capacity after correction of an elliptical horizontal crack; For the first The dynamic production capacity of the benchmark without interference from the crack; S45. The corrected single-crack capacity of all longitudinally distributed cracks is time-series superimposed to construct the multi-crack coupled total dynamic capacity sequence for the elastic development stage: ； In the formula: The dimensionless total dynamic production capacity of oil wells under multi-slit coupling interference conditions; This represents the total number of elliptical horizontal fractures in a single well. S46. Through full-domain time-series iterative calculation, a continuous dynamic sequence of production capacity is obtained, covering the entire stage of elastic development and considering the inter-gap pressure interference and flow field superposition effect.

6. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 5, characterized in that: Step S5 specifically includes the following steps: S51. Retrieve the continuous production capacity dynamic sequence output by S4, and simultaneously read the reservoir physical property parameters collected by S1 to distinguish between the purely elastic drive stage and the dissolved gas drive stage, thereby achieving staged modeling: ； In the formula: This represents the current average formation pressure. S52. Based on the comprehensive compressibility characteristics of reservoir rocks and fluids, an elastic flooding material balance model is established under the condition that the pressure is higher than the saturation pressure. The relationship between formation pressure and cumulative oil production in the elastic stage is characterized, and the overall comprehensive compressibility coefficient of the formation is obtained. The formula for calculating the mass balance of elastic drive is as follows: ； The formula for calculating the overall compressibility coefficient of a formation is as follows: ； In the formula: This represents the cumulative oil production of the reservoir. This represents the original geological reserves of crude oil in the reservoir. This represents the crude oil volume factor under the original formation pressure. The overall compressibility coefficient of the reservoir; The compression coefficient of crude oil; The rock skeleton compression coefficient; This represents the original formation pressure of the oil reservoir. S53. When the formation pressure falls below the saturation pressure, dissolved gas continuously precipitates from the crude oil, and the reservoir changes from single-phase flow to two-phase flow of oil and gas. Based on the characteristics of low-permeability oil reservoirs with no gas cap and weak edge and bottom water, a dissolved gas drive material balance model is established: ； ； In the formula: This represents the combined volumetric coefficient of oil and gas under the current pressure. This is the combined volumetric coefficient of oil and gas under saturated pressure, and it is equivalent to the volumetric coefficient of crude oil under saturated pressure. ; To calculate the cumulative gas-oil ratio; The original dissolved gas-oil ratio under saturated pressure; This represents the real-time dissolved gas-oil ratio under the current formation pressure. This represents the natural gas volume factor under the current pressure. Given the current cumulative oil production of the reservoir Cumulative production of gas-oil ratio , United and ,by The process involves iterating repeatedly with the only unknown variable in the iteration. After convergence, the real-time dissolved gas-oil ratio corresponding to the current formation pressure is output to obtain the full-time dissolved gas-oil ratio decay sequence. Based on the real-time dissolved gas-oil ratio obtained through iteration, and combined with the original differences in dissolved gas endowment and pore volume conservation, the real-time free gas saturation of the reservoir after pressure drop and dissolved gas precipitation is calculated: ； In the formula: This represents the real-time gas saturation of the formation under the current formation pressure. This represents the cumulative volume of free gas released per cubic meter of crude oil. S54. Based on the real-time dissolved gas-oil ratio decay law obtained from S53, and combined with the degassing viscosity characteristics of low-permeability reservoirs, a dynamic correction model for crude oil viscosity is established to replace the initial constant viscosity parameter and achieve dynamic updating of physical properties. The dynamic correction expression for crude oil viscosity is as follows: ； In the formula: This represents the dynamic crude oil viscosity under the current formation pressure. The base viscosity of crude oil under saturated pressure conditions; The viscosity sensitivity coefficient for degassing in low-permeability reservoirs; S55. Based on the real-time formation gas saturation generated by degassing, dynamically update the relative permeability of the oil and gas phases: ； In the formula: Real-time dynamic relative permeability of the oil phase; Real-time dynamic relative permeability of the gas phase; The relative permeability of the original oil phase during the initial single-phase seepage stage; The initial relative permeability of the gas phase when the initial free gas has not yet developed; Real-time formation gas saturation obtained for the degassing evolution of S53; The maximum gas saturation level confined within the reservoir; , are the permeability variation indices of the oil phase and gas phase, respectively, and the fitting constants for the physical properties of low-permeability reservoirs; S56. Based on the multi-slit coupled continuous production dynamic sequence output by S4, a production prediction model is constructed in stages: In the pure elastic drive stage, the formation pressure is higher than the saturation pressure, and there is no degassing or two-phase flow effect, so the production dynamic sequence of S4 is directly used; In the dissolved gas drive stage, the formation pressure is lower than the saturation pressure, and the dynamic correction coefficient of crude oil viscosity from S54 and the real-time dynamic relative permeability of the oil phase from S55 are substituted, and the production capacity decay rate is corrected by coupling the dissolved gas expansion and energy replenishment characteristics, and the final full-cycle production capacity is calculated: ； in, ； ； In the formula: To adjust the dynamic capacity for full-cycle flexible development; This is the continuous dynamic sequence of production capacity output by S4; This is the dynamic correction factor for crude oil viscosity. This is a correction factor for the two-phase flow capacity; The dissolved gas drive energy supply coefficient; This represents the viscosity of the benchmark crude oil under saturated pressure.

7. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 6, characterized in that: Step S6 specifically includes the following steps: S61, a single-phase seepage model framework for elliptical horizontal fractures constructed based on S2, retaining the geometry of the elliptical fractures, reservoir anisotropy, initiation pressure gradient, and the core mechanism of horizontal-vertical coupled seepage; retrieving the injection-production well pattern type and injection-production well spacing collected in S1. Correspondence between injection and production layers, and coefficient of variation of permeability. Penetration rate differences and injection / production parameters; S62. Based on the S2 elliptical horizontal fracture single-phase seepage model framework and combined with the actual injection-production correspondence, the seepage evolution law of the three stages of water injection development is divided, and an injection-production coupled dynamic seepage model adapted to fracturing low-permeability reservoirs is established: In the early stage of development without water seepage: the injected water has not yet affected the artificial fractures, and the reservoir fluid only undergoes vertical linear seepage from the matrix to the elliptical horizontal fractures and radial seepage from the fractures. The water cut of the oil well is 0, and the production capacity is maintained by the formation elastic energy and the initial water injection energy. Mid-stage water drive phase: Water injection energy continuously replenishes formation pressure, the water drive front gradually advances towards the elliptical fracture, and the reservoir simultaneously exhibits vertical matrix flow, radial planar flow, and water drive propulsion flow; Late-stage water channeling: Due to reservoir heterogeneity and high-pressure water injection, elliptical fractures and high-permeability strips form dominant water flow channels, resulting in fracture water channeling. S63. The classic C-type water drive curve of the oilfield is used to characterize the water drive law of the whole process of water injection development. The curve intercept and slope parameters are obtained by fitting production dynamic data. The classic C-type water drive curve expression for oilfields is as follows: ； In the formula: This represents the cumulative liquid production of the reservoir. The intercept for fitting the classic Type C waterflooding curve in the oilfield; The slope of the fitting curve for the classic Type C waterflooding curve in the oilfield; S64. Based on the curve intercept and slope parameters obtained from the S63 fitting, and combined with the formation crude oil volume factor and geological reserves, a quantitative calculation formula for water cut and volumetric sweep efficiency is established: ； In the formula: Moisture content The reservoir volumetric sweep efficiency; S65. Combining the reservoir heterogeneity parameters and water injection engineering parameters output from S1, and considering the deviations in productivity and water cut caused by non-uniform water drive and fracture water channeling, correct the dynamic productivity and water cut: ； ； in, ； ； ； In the formula: The dynamic production capacity during the water injection stage after multi-factor correction; This refers to the uncorrected basic water injection capacity. This is the corrected moisture content; Predict water cut based on classic type C water drive curves for oilfields; The dynamic volumetric sweep efficiency is calculated after correction for reservoir heterogeneity. The water injection sensitivity coefficient for low-permeability elliptical fractured reservoirs; This refers to the water cross-contamination correction factor for injection and production parameters; Inject pressure to average the formation; The steady-state injection-production ratio at the site; This is the correction coefficient for injection-production layer matching; This is a layer connectivity correction constant; The effective vertical overlap thickness of the injection and production layers; S66. Based on the phased seepage characteristics of water injection development, establish a segmented exponentially decreasing production capacity model: Early stage of rapid capacity reduction: ; Production capacity in the mid-to-late stage of slow decline: ; In the formula: These represent the daily oil production of oil wells in the early and middle-to-late stages, respectively. These represent the initial production capacity for the two production phases. These represent the production decline rates in the early and middle / late stages, respectively. For water injection development and production time; By combining S67, S66 segmented decreasing basic capacity, S64 time-varying water-drive volumetric sweep efficiency, and S65 multi-dimensional water channeling heterogeneity correction coefficient, explicit prediction formulas for dynamic capacity and water cut throughout the entire water injection development cycle are established. The final outputs are the dynamic sequence of capacity and dynamic change sequence of water cut throughout the entire water injection development cycle. ； ； in, ； In the formula: for Final production capacity at any given moment; To reduce basic production capacity in stages; for Dynamic volumetric sweep efficiency after reservoir heterogeneity correction at all times; for Moisture content at any given time; for Predicted water cut based on classic C-type water drive curve in Shike Oilfield.

8. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 7, characterized in that: Step S7 specifically includes the following steps: S71. Based on the actual development method of the reservoir, select the calculation results of the previous model as the fitting basis: in the elastic development stage, retrieve the full-cycle production capacity prediction sequence output by S5, and in the water injection development stage, retrieve the dynamic prediction sequence of production capacity and water cut output by S6; simultaneously collect the time-series production data measured on-site in the target reservoir and construct a historical fitting sample dataset. S72. Establish a multi-parameter coupled error objective function with the goal of minimizing the deviation between model predictions and actual field values: ； In the formula: This is to account for the overall fitting error; The weighting coefficients are used to fit the output. These are the weighting coefficients for the moisture content fitting. For the first Time-based models predict daily oil production; For the first Real-time daily oil production of the reservoir; For the first Moisture content is predicted by a time-based model; For the first Real-time reservoir water cut during production; This represents the total number of time-series samples in the historical fitted data. S73. Combining the seepage control mechanism of low-permeability elliptical horizontal fracture reservoirs and the dynamic response characteristics of water injection development, three types of sensitive parameters to be corrected—those with the highest sensitivity to dynamic changes in well productivity and water cut, the greatest on-site uncertainty, and the most significant contribution to modeling error—were selected as the sole iterative correction variables: fracture conductivity. Effective reservoir permeability and permeability variation coefficient ; All other reservoir physical parameters, fracture geometric parameters, fluid high-pressure physical parameters, and static geological parameters are fixed to the measured values ​​of S1 and S2 and do not participate in the inversion iteration to ensure the uniqueness of parameter inversion solution, physical and geological rationality, and model convergence stability. S74. Using the error objective function minimization iterative algorithm, iteratively optimize the three types of sensitive parameters to be corrected in S73 until the desired result is achieved. , For the first Round iteration comprehensive error; For the first Round iteration comprehensive error; This is a preset fitting accuracy threshold; After iterative convergence, the corrected optimal parameter set is output: corrected fracture conductivity, corrected reservoir effective permeability, and corrected permeability variation coefficient. S75. Based on the reservoir geological characteristics and the range of fracturing parameters, the optimal parameter set obtained by inversion in S74 is checked for boundary constraints: the parameter values ​​fall within the reservoir physical property range, the fracture modification characteristic range, and the heterogeneous distribution range. S76. Substitute the optimal parameters that meet the boundary constraints back into the single-phase seepage physical model of the elliptical horizontal fracture in S2, the calculation model of flow field coupling and pressure disturbance of multiple horizontal fractures in S4, the dissolved gas drive material balance model in S5, and S6. Replace the initial empirical parameters, complete the calibration of the full-link model parameters, and obtain the final production capacity prediction model that is adapted to the geological characteristics, fracturing stimulation characteristics, and production dynamics of the target reservoir.

9. The method for predicting the productivity of low-permeability reservoirs based on elliptical horizontal fractures according to claim 8, characterized in that: Step S8 specifically includes the following steps: S81. Using the final production capacity prediction model after historical fitting correction in S7 as the basis for numerical simulation, and combining the controllable parameters of fracturing construction, reservoir dynamic parameters, and water injection development engineering parameters, the following sensitivity analysis parameters are determined: fracture half-length, fracture shape factor, vertical to horizontal permeability ratio, initiation pressure gradient, injection-production ratio, and injection pressure. S82. Based on the measured basic parameters of S1, and combined with the feasible range of on-site fracturing construction and water injection development project, upper and lower limit fluctuation ranges are set for each analysis parameter, and multiple sets of gradient values ​​are evenly set within the upper and lower limit fluctuation ranges. S83. Keep the other parameters constant, only change the gradient value of the sensitivity analysis parameter, substitute it into the final production capacity prediction model of S7, and calculate the steady-state production capacity, cumulative oil production, production capacity deceleration rate and water cut increase rate of the reservoir under different sensitivity analysis parameter values. Obtain the dynamic response law of production capacity corresponding to the independent change of each sensitivity analysis parameter, and obtain the positive and negative influence trend and change range of the single parameter on production capacity. S84. Calculate the parameter sensitivity coefficient: ； In the formula: For the first The capacity sensitivity coefficient of the parameter; This represents the change in capacity forecast indicators. The initial production capacity is based on the baseline parameters. For the first The change in the values ​​of the sensitivity analysis parameters; For the first The baseline initial values ​​for the sensitivity analysis parameters; S85. Based on the absolute values ​​of the production capacity sensitivity coefficients of each parameter calculated in S84, the influence weights of fracture half-length, fracture shape factor, vertical to horizontal permeability ratio, start-up pressure gradient, injection-production ratio, and injection pressure are ranked in order of level. And distinguish between promoting and inhibiting parameters: the sensitivity coefficient is positive; the sensitivity coefficient is negative. S86. Based on the sensitivity analysis results, and combining the objectives of maximizing production capacity, minimizing production capacity decline, and maximizing the stable production period, a comprehensive optimization objective function for fracturing parameters is constructed: ； In the formula: To optimize the target values ​​for fracturing parameters; , , These are the weighting coefficients for capacity level, capacity deceleration rate, and length of stable production period, respectively. This represents the average effective production capacity corresponding to the current combination of fracturing parameters; The baseline production capacity corresponds to the initial baseline fracturing parameters; The initial rate of decline in production capacity under the baseline fracturing parameters; This represents the overall capacity reduction rate under the current fracturing parameters; This represents the effective stable production period length corresponding to the current fracturing parameters; The baseline stable production period length corresponds to the baseline fracturing parameters; Iterate through all gradient values ​​of fracture half-length, fracture shape factor, and fracturing stimulation strength parameters, and calculate each group sequentially. and take The largest interval is taken as the optimal interval for fracturing parameters: ； In the formula: and These represent the lower and upper limits of the optimal range of values ​​for a single type of fracturing parameter, respectively. The optimal parameter selection threshold is preset; S87. Combining the suppression of water channeling in fractures, reduction of the negative impact of heterogeneity, and ensuring the stable termination of water drive replacement energy, a comprehensive optimization objective function for water injection parameters is constructed: ； Constraints: ； In the formula: The target values ​​for comprehensive optimization of water injection parameters; , , These are the weighting coefficients for production capacity maintenance, water control effectiveness, and resistance to water channeling, respectively. This represents the average stable production capacity corresponding to the current water injection parameters; Base production capacity under benchmark water injection parameters; The average moisture content over the production cycle; The minimum allowable water channeling resistance threshold; and These are the upper and lower limits of the injection pressure engineering, respectively; Average injection pressure; and These are the lower and upper limits of the reasonable range for the injection-production ratio, respectively. For steady-state injection-production ratio; By iterating through the values ​​of injection pressure and injection-production ratio gradient, the parameter range corresponding to the maximum value of the objective function is taken as the optimal range of water injection development parameters. S88. Summarize the sensitivity weight ranking results of each parameter, the influence law of single parameter, and the coupling and adaptation relationship of multiple parameters, and finally output two sets of core optimal parameter ranges: one is the optimal range of fracturing parameters that adapts to the reservoir seepage characteristics, and the other is the optimal range of water injection development parameters that adapts to water injection to stabilize oil and control water.

Images