A method based on a mechanism model of mass transfer in CO2 huff and puff numerical simulation

Abstract

The present application relates to the field of oil and gas fields, disclose a kind of mechanism model of mass transfer based on numerical simulation CO2 huff and puff, solve the problem of current CO2 huff and puff cross-scale mass transfer mechanism complex;The present application fully considers the mass transfer between fracture and matrix in the process of CO2 huff and puff, a double medium model considering component change is established, and a layer of fracture surface is set on the outer surface of the model to simulate the large fracture formed after volume fracturing;The present application is first to use CMG numerical simulation software GEM component model to establish shale oil horizontal well volume fracturing CO2 huff and puff double medium geological model, and to establish mechanism model according to similarity criterion;Second, according to the fracture characteristics after horizontal well fracturing reconstruction, fracture surface is set on the surface of mechanism model;Third, change model parameters, analyze the energy-increasing effect of CO2 huff and puff under different influence factors, design shale oil CO2 huff and puff optimization scheme.The present application is used for mass transfer research in oil and gas fields.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas field development, and particularly relates to a method of mass transfer mechanism model based on numerical simulation CO2 huff and puff. BACKGROUND

[0002] In recent years, with a large number of conventional oil and gas reservoirs in the world entering the late stage of water injection development, the effect of enhanced oil recovery

[0003] is poor, and the effective development of unconventional reservoirs has become the focus of improving oil production. Shale oil, as an important part of unconventional oil and gas reservoirs, has huge reserves worldwide. According to the statistics of resource evaluation data, the total shale oil resources in the world are 67840x10 8 barrels, and the technically recoverable reserves are 3362x10 8 barrels, with an average recovery rate of 4.96%. In 2009, the US crude oil production broke through the original downward trend and achieved the first increase under the driving of effective development of shale oil reservoirs. Since the 21st century, the United States has successfully developed shale gas, and has realized the large-scale development of Bakken, Eagle Ford and Wolf Camp shale oil in succession. Compared with shale oil in the United States, shale oil reservoirs in China are mainly formed in terrestrial sedimentary environment, which is very difficult to develop.

[0004] China's shale oil recoverable reserves rank third in the world, and the proportion of terrestrial origin is high. Resource evaluation shows that

[0005] China's shale oil geological resources are about 200x10 8 t, and the technically recoverable resources are 20x10 8 ~25x10 8 t, but it is estimated that China's oil dependence on foreign countries will exceed 70% in 2020. The severe oil security problem makes it urgent for China to develop shale oil. There are three types of shale oil in China: terrestrial shale sandstone oil reservoirs, lacustrine carbonate oil reservoirs, and marl fracture oil reservoirs.

[0006] The application of huff and puff to enhance oil recovery technology in China is relatively late due to the lack of CO2 gas source. The earliest CO2 huff and puff field test in China was carried out by East China Petroleum Administration in 1989. The CO2 huff and puff test was carried out on Well Su88 in Zhangjiacao oilfield. The oil layer is 7.1m thick, the porosity is between 12.2% and 20.1%, the reservoir burial depth is 2933.6m, the oil saturation is between 20% and 70%, the viscosity is 20mP·s, the reservoir temperature is 104℃, and the relative density of crude oil is 0.8641. A total of 116t of CO2 was injected in 22.3h, the well was opened for production after 10d of shut-in, and the average oil production was increased from 0.46t / d to 0.52t / d.

[0007] From June to November 2010, CO2 huff and puff test was conducted on four wells in the Mabei block of Santanghu oilfield. The crude oil density of the block is 0.884 g / cm 3 , the crude oil viscosity is 104-1122 mP·s, the reservoir burial depth is 760-1580 m, the effective thickness is 7-13 m, the oil saturation is 55%-60%, and the reservoir temperature is 35℃-46℃. By the end of November 2010, the cumulative oil increment of the four huff and puff wells reached 144 m 3 . Among them, the average daily oil production of Hu220 well and Mabei 103 well increased to 1 m 3 / d after the implementation of huff and puff, and the production effect was relatively obvious.

[0008] By the end of October 2013, 10 wells were selected for field test in the super heavy oil block of Shuguang oil production plant in Liaohe oilfield. At present, the measure wells are all in the cycle oil production peak, and the average daily oil production of single well is 6 t, and the average daily oil production of Du well 84-37-39 is as high as 10 t.

[0009] Shale oil is the oil stored in shale sandstone, shale carbonate rock, etc. with matrix permeability ≤0.1×10 -3 μm 2 , and generally needs certain technical means to obtain industrial production capacity. At present, shale oil has successfully obtained exploration and development breakthrough in six basins such as Sichuan and Qaidam. The shale oil accumulation conditions are generally divided into source-in accumulation and near-source accumulation. In the process of accumulation, part of the crude oil will migrate along the natural fractures to form near-source accumulation, so that the development will be effectively improved by fully utilizing the natural fractures. In order to obtain the industrial oil flow of shale oil reservoir, fracturing reconstruction must be carried out, and the basic concept is to increase the contact area with the reservoir to improve the seepage condition. Volume fracturing is an important means to transform shale oil reservoir, which forms a certain volume of fracturing reconstruction area by making the natural fractures and artificial fractures interlaced and connected, and improves the reservoir permeability. However, the volume fracturing forms a complex fracture network at the same time, which leads to strong multi-scale medium and heterogeneity effect in the reservoir. The conventional reservoir engineering theory and numerical simulation method cannot effectively characterize the fracture characteristics of the shale oil reservoir after fracturing, so it is urgent to find a fracture model with advantages to adapt to the complex fracture network of fractured shale oil reservoir. At present, the commonly used fracture models for numerical simulation in the academic circle are single heavy medium model, double heavy medium model and discrete fracture model, etc. However, these models have poor simulation effect on shale oil reservoir after large volume fracturing, and their own modeling applicability is relatively poor. SUMMARY

[0010] The application comprehensively considers the mass transfer between the fractures and the matrix in the CO2 huff and puff process, and provides a method of a mechanism model based on numerical simulation of CO2 huff and puff mass transfer, and solves the problem of complex CO2 huff and puff cross-scale mass transfer mechanism. A double medium model considering component change is established, and a fracture surface is arranged on the outer surface of the model to simulate the large fracture formed after volume fracturing. The specific process includes the following steps: 1. a shale oil horizontal well volume fracturing CO2 huff and puff double medium geological model is established by using the GEM (component model) module of the CMG numerical simulation software, and a mechanism model is established according to the similarity criterion; 2. according to the fracture characteristics after the horizontal well fracturing, a fracture surface is arranged on the surface of the mechanism model; 3. the model parameters are changed, the CO2 huff and puff energy increasing effect under different influencing factors is analyzed, and the shale oil CO2 huff and puff optimization scheme is designed.

[0011] The application provides a method of a mechanism model based on numerical simulation of CO2 huff and puff mass transfer, which comprises the following steps:

[0012] Step 1: a fluid model is established by fitting fluid states by using typical shale oil block crude oil PVT experimental data

[0013] A reservoir numerical model is established, the array of the crude oil fluid characteristics is fitted into a fluid state to form a fluid digital model, a component model is established by high-pressure physical property PVT fitting experiment; the data of the shale oil block crude oil PVT experiment is obtained;

[0014] Step 2: state equation is calculated by using crude oil PVT physical property data

[0015] After the data of the crude oil PVT experiment is fitted into the PVT physical property data of the crude oil, the state equation is calculated by using the PVT physical property data of the crude oil; the fluid PVT attribute is determined; the fluid PVT attribute includes: crude oil density, crude oil saturation pressure, crude oil viscosity, crude oil relative volume and minimum miscibility pressure;

[0016] Step 3: a shale oil corner point network geological model is established

[0017] The actually measured basic parameters of the shale oil geological model are input into the Petrel RE module of the three-dimensional geological model Petrel software, and a shale oil corner point grid geological model is established according to the grid scale; a three-layer porosity distribution field map is generated by adopting a layer porosity heterogeneity processing method and a three-layer network porosity generation method, and the generated three-layer porosity distribution field map is integrated to be a typical porosity model;

[0018] Step 4: a shale oil reservoir horizontal well volume fracturing double medium geological model and a shale oil CO2 huff and puff mechanism model are established

[0019] The shale oil corner point network geological model is imported into the CMG component module to generate a shale oil reservoir horizontal well volume fracturing dual medium geological model, based on the shale oil reservoir horizontal well production mode, the dual medium geological model is subjected to CO2 huff and puff production history data numerical simulation, fitting oil production, water production and formation pressure, to obtain simulated values of production dynamic parameters; according to the real oil production, water production and formation pressure change of the reservoir, by adjusting the porosity, permeability and oil-water two-phase permeability curve, the fitted dual medium geological model is obtained, the characteristic unit body with a square edge smaller than the minimum edge length of the dual medium geological model is extracted from the fitted dual medium geological model, and the fracture surface is set on the surface of the characteristic unit body, to form a shale oil CO2 huff and puff mechanism model.

[0020] Step 5: Optimizing the shale oil CO2 huff and puff scheme based on the shale CO2 huff and puff mechanism model

[0021] According to the shale oil CO2 huff and puff mechanism model, the boundary conditions, soak time and injection-production system are selected as the influencing factors of CO2 huff and puff; in the boundary condition research, the mechanism model fracture surface is set to open on one side, two sides, four sides and six sides; and on the basis of the boundary condition research, three-side open fracture surface models in different conditions are set; the recovery rate under different boundary conditions is analyzed; in the soak time research, different soak times are selected, and the recovery rate under different soak times is analyzed; in the injection-production system research, the CO2 injection pressure is set to constant pressure injection, stepwise pressure increase and stepwise pressure decrease; the recovery rate under different injection-production systems is analyzed; the recovery rate parameters of CO2 huff and puff mass transfer under different influencing factors are obtained, and the boundary conditions, soak time and injection-production system corresponding to the maximum recovery rate are selected.

[0022] Preferably, the UNCONG state equation is used in step 2 to calculate the state equation based on the fitted fluid PVT data; the UNCONG state equation is a three-parameter cubic state equation, and has a general expression as formula (1):

[0023] (1)

[0024] In formula (1), Z represents the compressibility factor of the fluid, s , q , r respectively represent the three parameters of the cubic state equation; the three parameters are determined by formula (2):

[0025] (2)

[0026] In formula 2, w and u are general coefficients; A and B are coefficients related to the mixed state, which are determined by formula (3):

[0027] (3)

[0028] Since the three-parameter cubic equation of state requires the calculation of the compressibility factor of the mixture, a 、 b which needs to be determined by mixing rules, and the linear mixing rule is usually used, and a and b under the linear mixing rule can be determined by equation 4; for a mixture system containing N c components, the mole fractions of the i-th and j-th components in the mixture are x i , x j , and the interaction coefficient between the i-th and j-th components is i 、 j k i,j；

[0029] (4)

[0030] wherein, x i , x j are the mole fractions of the i-th and j-th components in the mixture, respectively, k i,j is the interaction coefficient between the i-th and j-th components. c i 、 j

[0031] Preferably, the step 3 of establishing the shale oil corner point network geology model further comprises the following steps:

[0032] Step 301 collects the data required for modeling, including stratification data, logging data and core test data;

[0033] Step 302 uses the Petrel RE module in the Petrel software to establish a shale oil corner point grid model;

[0034] Step 303 combines the experience of establishing the shale oil corner point network geology model, and the step length of the grid model in the X direction and the Y direction is selected as 20 m.

[0035] Preferably, the basic parameters of the shale oil geology model in step 3 include reservoir size, reservoir depth, reservoir pressure, matrix permeability average, matrix porosity average, main fracture permeability, main fracture porosity, main fracture aperture, and main fracture half-length; the method for processing interlayer porosity heterogeneity is to use three layers of surfaces with inconsistent properties, and the porosity field generation method of the corresponding three-layer grid is to apply different random seed numbers for Gaussian dispersion and adjust the offset; after integrating the generated three-layer porosity distribution field map, a typical porosity model is obtained.

[0036] ​​​Preferably, step 4 of establishing the CO2 injection mechanism model of shale oil specifically comprises the following steps:

[0037] Step 401 imports the geological model established by the Petrel software into the CMG numerical simulation component model to establish a shale oil reservoir horizontal well volume fracturing dual medium geological model; the percolation equation on which it is based is formula (5), and the oil and gas phase material balance equation on which it is based is formula (6);

[0038] (5)

[0039] Wherein, k is the permeability, unit: mD; p w , p o , p g , respectively, the water phase density, oil phase density and gas phase density, unit g / cm 3 ; m w , m o , m g , respectively, the water phase viscosity, oil phase viscosity and gas phase viscosity, unit: mP·s; x i The mole fraction of component i, unit: mol; f The phase field variable is dimensionless, F The interfacial energy is J / m 3 ; The porosity is q The flow rate is m 3 / s; t is the time, unit: s;

[0040] (6)

[0041] Wherein, z i The total mole fraction, unit: mol; L is the liquid phase mole fraction, unit: mol; V is the gas phase mole fraction, unit: mol;

[0042] Step 402 of fitting oil production, water production and formation pressure is based on the shale oil reservoir horizontal well production mode, and the CO2 huff and puff production history data of the geological model is numerically simulated to fit the oil production, water production and formation pressure, and the production dynamic parameter simulation value is obtained; according to the real oil production, water production and formation pressure change of the reservoir, the daily oil production, daily water production and formation pressure are fitted by adjusting the porosity, permeability and oil-water two-phase permeability curve;

[0043] Step 403 extracts the characteristic unit cell and establishes a shale oil CO2 huff and puff mechanism model; the cubic characteristic unit cell extracted in the dual medium geological model is encrypted, and the unit cell grid is encrypted, and the fracture surface is set on the six surfaces of the unit cell. When the fracture surface permeability is set to a threshold value of 100 mD, it is considered that the fracture surface is opened, and when the fracture surface permeability is set to 0.05 mD, it is considered that the fracture surface is closed.

[0044] Preferably, the parameter adjustment principle determined by fitting the daily oil production, daily water production and formation pressure fitting standard is as follows:

[0045] (1) Porosity

[0046] According to the basic characteristics of the reservoir, it is known that the physical properties of the shale oil reservoir are mainly low porosity and low permeability, the average porosity is 11.8%, and the porosity has a certain range of variation, and the allowable variation range is between ±5%;

[0047] (2) Permeability

[0048] The permeability of the shale oil reservoir is 0.1~0.01 mD, considering the measurement error and the uncertain factors of interwell distribution, the adjustment range of permeability is enlarged or reduced by 2-5 times;

[0049] (3) Rock fluid compressibility

[0050] The fluid compressibility is determined by laboratory measurement and is considered as a determined parameter;

[0051] (4) Initial fluid saturation and initial pressure

[0052] The parameters determined during the reservoir geological modeling are usually considered as determined parameters, and if necessary, small range modification is allowed;

[0053] (5) Relative permeability curve

[0054] Because the grid of the reservoir simulation model is fine, there is serious heterogeneity inside the grid, so the relative permeability curve is considered as an uncertain parameter, and in the fitting process, the permeability curve is adjusted according to experience;

[0055] (6) PVT high pressure property

[0056] The PVT high pressure property parameters of oil and water are measured by indoor test and have been fitted by numerical simulation phase simulator, and are considered as determined parameters.

[0057] Preferably, the recovery parameters of CO2 huff and puff mass transfer under different influencing factors obtained in step 5 are obtained, preferably the soak time and injection-production scheme refer to the following three aspects to obtain the maximum recovery, the specific steps are as follows:

[0058] Step 501 studies the recovery rate under different boundary conditions

[0059] A fracture surface is arranged on the surface of the mechanism model, and the model is divided into four forms of single-face opening, double-face opening, four-face opening and six-face opening; the mechanism model can intuitively reflect the communication relationship between fractures and between fractures and matrix, different opening faces can simulate the fracture development degree of the reservoir after volume fracturing, and the mechanism model has shorter operation time; the mass transfer law of CO2 huff and puff under different boundary conditions is analyzed, and the boundary condition corresponding to the maximum recovery rate of the oil well is regarded as the best boundary condition;

[0060] Step 502 studies the recovery rate under different soaking times

[0061] In combination with the actual mine field and the research experience of predecessors, the soaking time is preferably simulated as 5 days, 10 days, 15 days and 20 days; the mass transfer law of CO2 huff and puff under different soaking times is analyzed, and the soaking time corresponding to the maximum recovery rate of the oil well is regarded as the best soaking time;

[0062] Step 503 studies the recovery rate under different injection-production conditions

[0063] The injection-production conditions are divided into constant pressure injection, gradient pressure injection and gradient pressure injection; wherein, the constant pressure injection is that the injection pressure is increased to 1.25 times of the formation pressure each time, the gradient pressure injection is 1.0, 1.1, 1.2, 1.3, 1.4 and 1.5 times of the formation pressure respectively; the gradient pressure injection is 1.5, 1.4, 1.3, 1.2, 1.1 and 1.0 times of the formation pressure respectively; the mass transfer law of CO2 huff and puff under different injection-production systems is analyzed, and the injection-production scheme corresponding to the maximum recovery rate of the oil well is regarded as the best injection-production scheme.

[0064] The present application provides a mechanism model method based on numerical simulation of CO2 huff and puff mass transfer, which solves the problem of complex CO2 huff and puff mass transfer mechanism across scales; the present application fully considers the mass transfer effect between fractures and matrix during CO2 huff and puff process, establishes a dual medium model considering component change, and sets a fracture surface on the outer surface of the model to simulate large fractures formed after volume fracturing. The specific process includes: first, a shale oil horizontal well volume fracturing CO2 huff and puff dual medium geological model is established by using the GEM (component model) module of the CMG numerical simulation software, and a mechanism model is established according to the similarity criterion; second, a fracture surface is arranged on the surface of the mechanism model according to the fracture characteristics after the horizontal well fracturing; third, the model parameters are changed, the CO2 huff and puff energy increasing effect under different influencing factors is analyzed, and an optimization scheme of shale oil CO2 huff and puff is designed. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 It is a mechanism model method flowchart based on numerical simulation of CO2 huff and puff mass transfer;

[0066] Figure 2 It is a method embodiment technical route diagram based on the mechanism model of numerical simulation of CO2 huff and puff mass transfer;

[0067] Figure 3 It is a method three open different situation model diagram based on the mechanism model of numerical simulation of CO2 huff and puff mass transfer;

[0068] Figure 4 It is a method recovery curve diagram under different boundary conditions based on the mechanism model of numerical simulation of CO2 huff and puff mass transfer;

[0069] Figure 5 It is a method crude oil recovery diagram under different three open situations based on the mechanism model of numerical simulation of CO2 huff and puff mass transfer;

[0070] Figure 6 It is a method crude oil recovery curve diagram under different soak time based on the mechanism model of numerical simulation of CO2 huff and puff mass transfer;

[0071] Figure 7 It is a method crude oil recovery curve diagram under different injection-production system based on the mechanism model of numerical simulation of CO2 huff and puff mass transfer. DETAILED DESCRIPTION

[0072] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, the same numbers refer to the same or similar elements throughout the drawings. The implementation described in the following exemplary embodiments is not meant to represent all implementations consistent with the present invention.

[0073] The method provided by the embodiment of the present application is a mechanism model method based on numerical simulation of CO2 huff and puff mass transfer, as shown in the figure, the method comprises the following steps: Figure 1

[0074] Step 1: Establish a fluid model by fitting fluid state with typical shale oil block crude oil PVT experimental data

[0075] Establish a numerical model of the reservoir, fit the array of crude oil fluid characteristics into a fluid state to form a fluid digital model, and establish a component model through high-pressure physical property PVT fitting experiment; obtain the data of the shale oil block crude oil PVT experiment;

[0076] Step 2: Calculate the equation of state using crude oil PVT physical property data

[0077] After fitting the data of the crude oil PVT experiment into the PVT physical property data of the crude oil, the equation of state is calculated using the PVT physical property data of the crude oil; the PVT properties of the fluid are determined; the PVT properties of the fluid include: crude oil density, crude oil saturation pressure, crude oil viscosity, crude oil relative volume and minimum miscibility pressure;​

[0078] Step 3: Establishing a shale oil corner point network geology model

[0079] The actually measured shale oil geology model basic parameters are input into the Petrel RE module of the three-dimensional geology model Petrel software, a shale oil corner point network geology model is established according to the grid scale, a three-layer porosity distribution field map is generated by using the interlayer porosity heterogeneity processing method and the three-layer porosity generation method, and the generated three-layer porosity distribution field map is integrated to obtain a typical porosity model.

[0080] Step 4: Establishing a shale oil reservoir horizontal well volume fracturing dual medium geology model and a shale oil CO2 huff and puff mechanism model

[0081] The shale oil corner point network geology model is imported into the CMG component module to generate a shale oil reservoir horizontal well volume fracturing dual medium geology model, CO2 huff and puff production history data numerical simulation is carried out on the dual medium geology model based on the shale oil reservoir horizontal well production mode, oil production, water production and formation pressure are fitted to obtain simulated production dynamic parameter values, the fitted dual medium geology model is obtained by adjusting the porosity, permeability and oil-water two-phase permeability curves according to the actual oil production, water production and formation pressure changes, a characteristic unit body with a side length of 1.04 m is extracted from the fitted dual medium geology model, and a fracture surface is set on the surface of the characteristic unit body to form a shale oil CO2 huff and puff mechanism model.

[0082] Step 5: Optimizing a shale oil CO2 huff and puff scheme based on the shale CO2 huff and puff mechanism model

[0083] According to the shale oil CO2 huff and puff mechanism model, the boundary conditions, soak time and injection-production system are selected as the influencing factors of CO2 huff and puff; in the boundary condition research, the mechanism model fracture surface is set to be opened on one side, two sides, four sides and six sides; and on the basis of the boundary condition research, three-side opening fracture surface models in different conditions are set; the recovery rate under different boundary conditions is analyzed; in the soak time research, different soak times (5, 10, 15 and 20 days) are selected, and the recovery rate under different soak times is analyzed; in the injection-production system research, the CO2 injection pressure is set to be injected at a constant pressure, in a gradient pressure and in a gradient pressure; (wherein, the constant pressure injection is that the injection pressure is increased to 1.25 times of the formation pressure each time, the gradient pressure is 1.0, 1.1, 1.2, 1.3, 1.4, 1.5 times of the formation pressure, respectively; the gradient pressure is 1.5, 1.4, 1.3, 1.2, 1.1, 1.0 times of the formation pressure, respectively) the recovery rate under different injection-production systems is analyzed; the recovery rate parameters of CO2 huff and puff mass transfer under different influencing factors are obtained, and the boundary conditions, soak time and injection-production system corresponding to the maximum recovery rate are selected.

[0084] In one embodiment, the UNCONG state equation is a three-parameter cubic state equation, expressed in general form as equation (1):

[0085] (1)

[0086] In the formula: Z represents the compressibility factor of the fluid. s , q , r These are represented by the three parameters of the cubic state equation; the three parameters are determined by equation (2):

[0087] (2)

[0088] Formula 2 w u and are general coefficients; A and B are coefficients related to the mixing state, determined by equation (3):

[0089] (3)

[0090] Since the three-parameter cubic equation of state requires the calculation of the compressibility factor of the mixture, therefore a , b It needs to be determined through mixing rules, usually linear mixing rules are used, and a and b under linear mixing rules can be determined by Equation 4; for a mixture system, if it contains N c The mole fractions of components i and j in the mixture are respectively x i , x j Components i , j The interaction coefficient between them is k i,j；

[0091] (4)

[0092] in, x i , x j , respectively, are the mole fractions of the i-th and j-th components in the mixture. k i,j For N c Components i , j The interaction coefficient between them.

[0093] In one embodiment, step 3 of establishing the shale oil corner network geological model further includes the following steps:

[0094] Step 301: Collect the data required for modeling, including stratification data, well logging data, and core test data;

[0095] Step 302 uses the Petrel RE module in the Petrel software to establish a shale oil corner point grid model;

[0096] Step 303 combines the experience of establishing the shale oil corner point network geological model to select the step length of the grid model in the X direction and the Y direction as 20 m.

[0097] Preferably, the basic parameters of the shale oil geological model in step 3 include reservoir size, reservoir depth, reservoir pressure, matrix permeability average, matrix porosity average, main fracture permeability, main fracture porosity, main fracture opening, and main fracture half length; the processing method of interlayer porosity heterogeneity is to use three layers with inconsistent attribute distribution, and the corresponding porosity field generation method of the three-layer grid is to apply different random seed numbers for Gaussian dispersion and adjust the offset; after integrating the generated three-layer porosity distribution field map, the typical porosity model is obtained.

[0098] In one embodiment, step 4 of establishing a shale oil CO2 injection mechanism model specifically includes the following steps:

[0099] Step 401 imports the geological model established by the Petrel software into the CMG numerical simulation component model to establish a shale oil reservoir horizontal well volume fracturing dual medium geological model; the percolation equation it relies on is formula (5), and the oil and gas phase material balance equation it relies on is formula (6);

[0100] (5)

[0101] Wherein, k is the permeability, unit: mD; p w 、 p o 、 p g , respectively, the water phase density, oil phase density and gas phase density, unit g / cm 3 ; m w 、 m o 、 m g , respectively, the water phase viscosity, oil phase viscosity and gas phase viscosity, unit: mP·s; x i is the mole fraction of component i, unit: mol; f is the phase field variable, dimensionless, F is the interfacial energy, unit: J / m 3 ; is the porosity, q is the flow rate, unit: m 3 / s; t is the time, unit: s;

[0102] (6)

[0103] Among them, z i The total mole fraction is expressed in mol; L is the liquid phase mole fraction in mol; V is the gas phase mole fraction in mol.

[0104] Step 402, fitting oil production, water production, and formation pressure, is based on the horizontal well production method of the shale oil reservoir. It involves numerically simulating historical CO2 huff and puff production data from the geological model to fit oil production, water production, and formation pressure, obtaining simulated values ​​for dynamic production parameters. Based on the actual changes in oil production, water production, and formation pressure in the reservoir, daily oil production, daily water production, and formation pressure are fitted by adjusting porosity, permeability, and oil-water two-phase permeability curves.

[0105] Step 403: Extract feature units and establish a shale oil CO2 injection mechanism model; extract cube feature units with side lengths smaller than the minimum side length of the dual-medium geological model from the dual-medium geological model, refine the unit grid, and set fracture surfaces on the six faces of the unit. When the permeability of the fracture surface is set to the threshold of 100mD, it is considered that the fracture surface is open; when the permeability of the fracture surface is set to 0.05mD, it is considered that the fracture surface is closed.

[0106] In one embodiment, the parameter adjustment principles for the fitting standard determined by fitting daily oil production, daily water production, and formation pressure are as follows:

[0107] (7) Porosity

[0108] Based on the basic reservoir characteristics, it is known that the physical properties of shale oil reservoirs are mainly low porosity and low permeability, with an average porosity of 11.8%. The porosity has a certain range of variation, with an allowable range of ±5%.

[0109] (8) Penetration

[0110] Shale oil reservoirs have permeability ranging from 0.1 to 0.01 mD. Considering measurement errors and uncertainties in inter-well distribution, the adjustment range for permeability can be increased or decreased by 2 to 5 times.

[0111] (9) Rock fluid compressibility

[0112] The fluid compressibility coefficient is determined in the laboratory and is considered a fixed parameter;

[0113] (10) Initial fluid saturation and initial pressure

[0114] The parameters determined during reservoir geological modeling are generally considered to be fixed, but minor modifications are allowed if necessary.

[0115] (11) Relative permeability curve

[0116] Because the reservoir simulation model has a fine mesh and severe heterogeneity within the mesh, the relative permeability curve is considered an uncertain parameter. During the fitting process, the permeability curve is adjusted based on experience.

[0117] (12) High-voltage physical properties of PVT

[0118] The PVT high-pressure physical properties of oil and water were obtained through indoor tests and have been fitted by a numerical simulation phase simulator, and are considered as determined parameters.

[0119] In one embodiment, obtaining the recovery parameters of CO2 huff-and-puff mass transfer under different influencing factors in step 5, and optimizing the well-keeping time and injection-production scheme, refers to obtaining the maximum recovery rate by studying the following three aspects, and the specific steps are as follows:

[0120] The boundary conditions corresponding to the maximum recovery rate of the oil well are considered as the optimal boundary conditions.

[0121] Step 502: Study the recovery rate under different well-drying times.

[0122] Based on the actual conditions of the mine and the research experience of predecessors, the well shut-in time was selected as 5 days, 10 days, 15 days and 20 days for simulation; the CO2 huff and puff mass transfer law under different well shut-in time was analyzed, and the well shut-in time corresponding to the maximum oil recovery rate of the oil well was regarded as the optimal well shut-in time.

[0123] Step 503: Study the recovery rate under different injection and production conditions.

[0124] The injection and production conditions were divided into constant pressure injection, stepped pressure-increasing injection, and stepped pressure-decreasing injection. Constant pressure injection involved each injection pressure increasing to 1.25 times the formation pressure, with stepped pressure increases of 1.0, 1.1, 1.2, 1.3, 1.4, and 1.5 times the formation pressure, respectively. Stepped pressure decreases were 1.5, 1.4, 1.3, 1.2, 1.1, and 1.0 times the formation pressure, respectively. The CO2 huff and puff mass transfer patterns under different injection and production regimes were analyzed, and the injection and production scheme corresponding to the maximum oil recovery rate of the well was considered the optimal injection and production scheme.

[0125] This invention proposes a method for modeling the mechanism of CO2 huff and puff mass transfer based on numerical simulation, solving the problem of complex cross-scale mass transfer mechanisms in current CO2 huff and puff systems. This invention fully considers the mass transfer between fractures and the matrix during CO2 huff and puff, establishing a dual-medium model that considers compositional variations, and setting a fracture surface on the outer surface of the model to simulate large fractures formed after volumetric fracturing. The specific process includes: first, using the GEM (Composition Model) module of the CMG numerical simulation software to establish a dual-medium geological model of CO2 huff and puff for volumetric fracturing in shale oil horizontal wells, and establishing a mechanism model based on similarity criteria; second, setting a fracture surface on the surface of the mechanism model based on the fracture characteristics after horizontal well fracturing; and third, changing the model parameters, analyzing the energy enhancement effect of CO2 huff and puff under different influencing factors, and designing an optimized CO2 huff and puff scheme for shale oil.

[0126] Example 1

[0127] The technical solution of the present invention will be further described in detail below with reference to embodiments, but the scope of protection of the present invention is not limited thereto. Figure 2 As shown;

[0128] In one embodiment, the basic data for this model were selected from a typical shale oil reservoir. Based on the dual-medium model, a fracture surface was added to simulate the fracture network after volumetric fracturing.

[0129] Based on the understanding of the fluid characteristics, geological properties, and fluid-structure interaction of shale reservoirs, a numerical model of CO2 huff and puff components in horizontal wells of shale reservoirs is established.

[0130] (i) Using PVT experimental data of crude oil from typical shale oil blocks to fit the fluid state and establish a fluid model; (PVT fitting of shale oil reservoir fluids)

[0131] Establishing a reservoir numerical model first requires establishing a fluid numerical model. Accurate numerical representation of fluid characteristics requires fitting the fluid state. Unlike the black oil model, the core method for simulating and reproducing fluid characteristics in the component model is the equation of state. The prerequisite for realizing the equation of state is that the high-pressure physical property PVT fitting experiment must match the physical model experiment. Therefore, the PVT numerical simulation fitting experiment is an essential prerequisite step for establishing the component model. The three-parameter cubic equation of state upon which the model is based is shown in formula (1).

[0132] The physical properties of the model crude oil were obtained from PVT experiments of crude oil from a shale oil block in a domestic oilfield. In order to restore the fluid state of the formation crude oil, the experimental crude oil sample was obtained by blending surface degassed crude oil and flash vapor with a dissolved gas-oil ratio of 20. The detailed data of the simulated components are shown in Table 1.

[0133] Table 1. Model crude oil pseudo-component physical property parameters

[0134]

[0135] As can be seen from the data in Table 1, this oil sample has a relatively high content of heavy components and a relatively low content of methane compared to C9 and above. Therefore, this oil sample belongs to medium-high viscosity crude oil. The viscosity of the crude oil under the laboratory formation conditions (formation temperature 89.28℃, formation pressure 35.25MPa) is approximately 15-20 mPa∙s. CO2 immersion experiments show that this oil sample has a strong viscosity-reducing effect and good miscibility with CO2.

[0136] (ii) Calculate the equation of state using crude oil PVT property data.

[0137] After fitting the PVT physical property data of crude oil, the formation state of the fluid is restored to a high degree, and it can express the high pressure physical properties and seepage state of crude oil fluid to a certain extent. Among them, oil reservoir fluid refers to oil, natural gas and formation water stored underground.

[0138] High-pressure properties: The physical properties of fluids in oil and gas reservoirs under high pressure and high temperature underground;

[0139] Seepage refers to the flow of fluid in a porous medium. A porous medium is a substance composed of granular or fragmented materials containing numerous pores or fissures. Seepage in soil or rock strata below the Earth's surface is typically called groundwater movement and is the most common seepage phenomenon in nature. Seepage has wide applications in fields such as water conservancy, geology, mining, petroleum, environmental protection, chemical engineering, biology, and medicine.

[0140] The equation of state is calculated using the fitted fluid PVT data. The UNCONG equation of state is solved using the PR equation or the SRK equation, both of which are three-parameter cubic equations of state with a general expression as shown in equation (1).

[0141] (1)

[0142] In the formula, Z The compressibility factor of the fluid; s , q , r —The three parameters are determined by the following formula:

[0143] (2)

[0144] In equation 2, A and B are coefficients related to the mixing state, determined by equation (3):

[0145] (3)

[0146] Since the three-parameter cubic equation of state requires the calculation of the compressibility factor of the mixture, thereforea , b It needs to be determined through mixing rules. There are many types of mixing rules, but the linear mixing rule is commonly used. Under the linear mixing rule, a and b can be determined by Equation 4. For a mixture system, if it contains N... c The mole fractions of components i and j in the mixture are respectively x i , x j Components i , j The interaction coefficient between them is k i,j ,but:

[0147] (4)

[0148] The fluid PVT properties calculated by the fitted equation of state can be regarded as the PVT properties of the actual fluid flowing in the formation. This is used to establish a fluid model for application on the carrier of the geological model. Once the fluid PVT properties are determined, they can generally be applied to most geological models. However, the more mismatched the fluid model is with the geological model, the shorter the convergence time will be, and the lower the degree of agreement between the model calculation results and reality will be in the later stages of production.

[0149] (III) Establishing a shale oil corner network geological model using Petrel software

[0150] Natural fracture development is a typical geological feature of shale oil reservoir development, and its impact on permeability anisotropy and local seepage conditions is particularly important, manifesting as heterogeneity in porosity and permeability fields across the entire reservoir. First, the necessary data for modeling (layer data, well logging data, core test data, etc.) were collected, and the basic model parameters are shown in Table 2. Then, a corner grid model with a grid size of 51×100×1.5km was established using the Petrel RE module in Petrel software. While ensuring the accuracy of the geological conditions description, and based on experience in geological model building, the step size in both the X and Y directions of the grid model was chosen to be 20m. The heterogeneity of interlayer porosity was addressed by using three layers with inconsistent attribute distributions. The porosity field generation method for the corresponding three-layer grid involved Gaussian discretization using different random seed numbers and adjusting the offset. The generated three-layer porosity distribution field maps were then integrated to form a typical porosity model.

[0151] (iv) Import the model into CMG numerical simulation software and use the CEM module of CMG numerical simulation software to establish a dual-medium geological model of horizontal well volumetric fracturing in shale oil reservoirs;

[0152] A shale oil corner network geological model was imported into CMG numerical simulation software. Using the component module of CMG, a dual-medium geological model of horizontal well volumetric fracturing in shale oil reservoirs was established. The CMG component model is a type of reservoir simulation software used to evaluate reservoir development and production effects. This model has multiple sub-models and can simulate the generation, movement, and discharge of various fluid components in the reservoir, as well as the interaction and impact between fluids and rocks, making it a relatively comprehensive reservoir model.

[0153] The seepage equation on which the model is based is shown in Formula (5), and the oil and gas phase material balance equation on which it is based is shown in Formula (6); the basic parameters of the shale oil corner network geological model are shown in Table 2 below.

[0154] (5)

[0155] Where k is the permeability, mD, p w , p o , p g These are the densities of the aqueous phase, oil phase, and gas phase, respectively, in g / cm³. 3 , m w , m o , m g These represent the viscosity of the aqueous phase, the viscosity of the oil phase, and the viscosity of the gas phase, in mP·s. x i Let i be the mole fraction of component i, in mol. f These are phase field variables, dimensionless. F For interface energy, J / m 3 , Porosity q For flow rate, m 3 / s, where t is time, in seconds.

[0156] (6)

[0157] Among them, z i denoted as total mole fraction (mol), L as liquid phase mole fraction (mol), and V as gas phase mole fraction (mol).

[0158] Table 2 Basic parameters of shale oil geological model

[0159]

[0160] Table 3 Basic parameters of shale oil mechanism model

[0161]

[0162] (v) Fitting of oil production, water production and formation pressure

[0163] Based on the horizontal well production method of the shale oil reservoir, numerical simulation of historical CO2 huff and puff production data was performed on the geological model to fit oil production, water production, and formation pressure; simulated values ​​of dynamic production parameters were obtained; and porosity, permeability, and oil-water two-phase permeability curves were adjusted according to the actual changes in oil production, water production, and formation pressure. Daily oil production, daily water production, and formation pressure were then fitted. Historical data fitting is the process of inversely calculating and correcting reservoir physical properties based on observed actual dynamic parameters. This inversion process often has multiple solutions, meaning that combinations of multiple physical properties can yield the same result. Therefore, it is necessary to determine fitting standards to standardize the solution process of the inversion calculation. Currently, there is no universally applicable method for historical data fitting. Based on the understanding and summary of shale oil reservoirs and geological laws, the parameter adjustment principles for the fitting standards are determined as follows:

[0164] (13) Porosity

[0165] Based on the basic characteristics of reservoirs described in the reservoir description, it can be known that the physical properties of shale oil reservoirs are generally poor, mainly characterized by low porosity and low permeability, with an average porosity of 11.8%. Therefore, the porosity has a certain range of variation, and the allowable range of change is within ±5%.

[0166] (14) Penetration

[0167] Shale oil reservoirs generally have very low permeability and a wide range of permeability variations. Considering uncertain factors such as measurement errors and inter-well distribution, the adjustment range for permeability is relatively large, and can be magnified or reduced by 2-5 times.

[0168] (15) Rock fluid compressibility coefficient

[0169] The compressibility of liquids is determined in the laboratory and varies within a very small range, so it should be considered a fixed parameter. However, the compressibility of rocks is affected by the internal saturated liquid and stress state, and has a certain range of variation. Moreover, the non-effective part connected to the effective thickness also has a certain amount of porosity, which plays a certain elastic role during the rising process. Taking this part of the influence into account, the compressibility of rocks can be increased by 1-1.5 times.

[0170] (16) Initial fluid saturation and initial pressure

[0171] These parameters are determined during reservoir geological modeling and should generally be considered fixed parameters, with minor modifications allowed if necessary.

[0172] (17) Relative permeability curve

[0173] Because the reservoir simulation model has a fine mesh and there is severe heterogeneity within the mesh, the impact of which cannot be ignored, the relative permeability curve should be regarded as an uncertain parameter. During the fitting process, appropriate adjustments to the relative permeability curve are allowed.

[0174] (18) High-voltage physical properties of PVT

[0175] The PVT high-pressure physical properties of oil and water were obtained through indoor tests and have been fitted by a numerical simulation phase simulator. The variation range is very small and should be regarded as definite parameters.

[0176] (vi) Establishing a CO2 injection mechanism model for shale oil

[0177] Feature cells were extracted from the fitted geological model, and a mechanistic model was established based on similarity criteria. Cubic feature cells of 1.04 × 1.04 × 1.04 m were extracted from the geological model, and the cell mesh was refined. Fracture surfaces with a thickness of 0.02 m were set on the six faces of the cells. A fracture surface permeability of 100 mD was considered open, and a permeability of 0.05 mD was considered closed. After the model was established, a constant pressure production method was adopted, with the formation fluid pressure set at 35 MPa. The CO2 injection pressure was 1.25 times the formation pressure. The initial mechanistic model settings were: six-sided fracture opening, six injection cycles, constant pressure injection, and a 10-day well shut-in period.

[0178] (vii) Establish models with different boundary conditions

[0179] Because the fractures after horizontal well fracturing are characterized by their length and width, high conductivity, and complex, grid-like interweaving between fractures and the matrix, as well as between fractures themselves, fracture surfaces are incorporated into the mechanistic model. This results in four types of models: single-sided, double-sided, four-sided, and six-sided opening. This mechanistic model visually represents the communication relationships between fractures and between fractures and the matrix. Different opening surfaces can simulate the degree of fracture development in the reservoir after large-volume fracturing. Furthermore, the mechanistic model is computationally simpler, significantly reducing computation time. Analysis of CO2 huff and puff mass transfer under different boundary conditions shows that the recovery rate increases with the increase of the opening surface. Figure 4 (Table 4).

[0180] Table 4 Recovery rate data under different boundary conditions

[0181]

[0182] This is because increasing the opening surface facilitates the mass transfer of CO2 between the fracture and the matrix, allowing CO2 to come into more complete contact with crude oil, which is more conducive to dissolving crude oil, reducing interfacial tension, and extracting light components from crude oil.

[0183] In the study of the three-sided opening model, due to the different positions of the opening surfaces, such as... Figure 3 As shown, Case 1: Top, bottom, and sides are opened; Case 2: Top and two non-adjacent sides are opened; Case 3: Top and two adjacent sides are opened, and the harvesting rates are also different. Figure 5 (See Table 5), which shows that CO2 drive also exists during CO2 throughput.

[0184] Table 5. Oil recovery rate data under different three-sided opening conditions.

[0185]

[0186] (VIII) CO2 huff and puff models were established for different well-closing times (5 days, 10 days, 15 days, and 20 days) to analyze the mass transfer law of CO2 huff and puff under different well-closing times. Too short a well-closing time is not conducive to sufficient contact between CO2 and crude oil, while too long a well-closing time is not conducive to oil well production efficiency, and excessively long well-closing times may cause CO2 to diffuse into deeper formations, which is not conducive to maintaining reservoir pressure. Based on actual field conditions and previous research experience, well-closing times of 5 days, 10 days, 15 days, and 20 days were selected for simulation. The results show that as the well-closing time increases, the contact time between CO2 and crude oil during the well-closing process is longer, and the recovery rate also increases accordingly. Figure 6 (Table 6).

[0187] Table 6. Oil recovery rate data under different well shut-in times

[0188]

[0189] (ix) CO2 huff and puff models were established under different injection and production regimes (constant pressure injection, stepped pressurization, and stepped depressurization) to analyze the CO2 huff and puff mass transfer characteristics under different regimes. Constant pressure injection involved each injection pressure increasing to 1.25 times the formation pressure; stepped pressurization was 1.0, 1.1, 1.2, 1.3, 1.4, and 1.5 times the formation pressure; stepped depressurization was 1.5, 1.4, 1.3, 1.2, 1.1, and 1.0 times the formation pressure. The results showed that the recovery rate was higher under the stepped pressurization regime. Figure 7 (Table 7) This is because in the middle and later stages of the cycle, the formation pressure is insufficient and more formation pressure needs to be replenished.

[0190] Table 7. Oil recovery rate data under different injection and production systems.

[0191]

[0192] (x) Based on the above research results, optimize the CO2 injection and huffing scheme for shale oil.

[0193] The above studies show that the recovery rate increases with the increase of the opening face; the recovery rate also increases with the increase of the well shut-in time; and the recovery rate is even higher under the step-up pressurization working system. Based on a comprehensive consideration of the actual geological characteristics of the reservoir and reservoir economics, the working system of six-sided opening, 15-day well shut-in, and step-up pressurization was finally selected as the optimal system.

[0194] (X) Study on the mass transfer law of CO2 intake and output in the mechanism model

[0195] Boundary conditions, well-closing time, and injection-production regime were selected as influencing factors for CO2 huff and puff. When studying the mass transfer law of CO2 huff and puff under different boundary conditions, mechanistic models were set up with single-sided, two-sided, four-sided, and six-sided fracture opening. The final recovery rate, CO2 extraction of light components, pressure field changes, CO2 injection rate, and CO2 injection layer were analyzed. The study shows that the more opening surfaces there are, the more thoroughly CO2 mixes with crude oil, the more significant the CO2 extraction of light components, and the higher the recovery rate. Furthermore, in the three-sided opening model, the recovery rate varies slightly due to the different positional relationships between the opening surfaces and the huff and puff wells, indicating that CO2 also plays a role in CO2 displacement during the huff and puff process. When studying the mass transfer law of CO2 huff and puff under different well-closing times, the well-closing time t of the characteristic unit was calculated according to the similarity criterion, corresponding to (5, 10, 15, and 20 days). Studies have shown that the longer the well is shut-in, the more fully CO2 contacts crude oil, the better the extraction of light components from crude oil, and the greater the recovery rate. When studying the CO2 huff and puff mass transfer law under different injection and production regimes, the injection and production regimes are divided into constant pressure injection (each time rising to 1.25 times the formation pressure), step-by-step pressure increase (1.0, 1.1, 1.2, 1.3, 1.4, 1.5) and step-by-step pressure decrease (1.5, 1.4, 1.3, 1.2, 1.1, 1.0).

[0196] Based on the model results, in models with different boundary conditions, such as Figure 4 As shown, the recovery rate increases with the increase of the fracture surface. This is because the more fracture surfaces are opened, the more favorable it is for CO2 diffusion, resulting in more sufficient contact between CO2 and crude oil, thus improving the recovery rate. In the three-sided fracture model, as shown... Figure 5 As shown, the recovery rate varies depending on the location of the opening face, indicating that CO2 flooding also plays a role in the CO2 huff and puff process; in models with different well shut-in times, as shown... Figure 6 As shown, with the increase of well-shutting time, CO2 has a longer contact time with crude oil during the well-shutting process, and the recovery rate also increases accordingly; in models of different injection-production regimes, such as... Figure 7As shown, the recovery rate is higher under the step-up pressurization operating system. This is because the formation pressure is insufficient in the middle and later stages of the cycle, requiring additional formation pressure. Based on a comprehensive consideration of the actual geological characteristics of the reservoir and reservoir economics, the optimal operating system was ultimately selected: six-sided opening, 15-day well shut-in, and step-up pressurization.

[0197] This invention provides a method for modeling the mechanism of CO2 huff and puff mass transfer based on numerical simulation, solving the problem of complex cross-scale mass transfer mechanisms in current CO2 huff and puff technologies. Based on the seepage and geological characteristics of typical shale oilfields, this invention establishes a large-scale geological model of horizontal well volumetric fracturing in the reservoir and performs historical fitting of production data from the oilfield. Based on this, characteristic unit cells are extracted, and the fluid exchange characteristics of the fracture system and matrix system in a dual-medium model within numerical simulation software are used to establish the fracture-matrix mass transfer law of CO2 injection in shale oil, combined with the large fractures formed after horizontal well volumetric fracturing. This method reveals the CO2 diffusion law, mass transfer mechanism, and energy enhancement effect under different conditions, providing theoretical guidance for improving CO2 huff and puff technology strategies, thereby achieving efficient development of CO2 injection in shale oil. Currently, there is no known mechanism model for the cross-scale mass transfer mechanism of CO2 huff and puff in the field of oil and gas field development technology.

Claims

1. A method based on a mechanism model of mass transfer in CO2 huff and puff based on numerical simulation, characterized in that, The method comprises the following steps: Step 1: Fitting fluid state with typical shale oil block crude oil PVT experimental data to establish a fluid model Step 1: Fitting fluid state with typical shale oil block crude oil PVT experimental data to establish a fluid model Step 1: Fitting fluid state with typical shale oil block crude oil PVT experimental data to establish a fluid model Step 2: Calculating state equation using crude oil PVT property data After fitting the PVT experimental data of the crude oil into the PVT property data of the crude oil, the state equation is calculated using the PVT property data of the crude oil; Determine the PVT properties of the fluid; The PVT properties of the fluid include: the density of the crude oil, the saturation pressure of the crude oil, the viscosity of the crude oil, the relative volume of the crude oil and the minimum miscibility pressure; Step 3: Establishing a shale oil corner point network geological model The actual measured shale oil geological model basic parameters are input into the Petrel RE module of the three-dimensional geological model Petrel software, and a shale oil corner point grid geological model is established according to the grid scale; A three-layer porosity distribution field map is generated by using the interlayer porosity heterogeneity processing method and the three-layer porosity generation method, and the generated three-layer porosity distribution field map is integrated into a typical porosity model; Step 4: Establishing a shale oil reservoir horizontal well volume fracturing dual medium geological model and a shale oil CO2 huff and puff mechanism model The shale oil corner point network geological model is imported into the CMG component module to generate a shale oil reservoir horizontal well volume fracturing dual medium geological model, and based on the shale oil reservoir horizontal well production mode, the CO2 huff and puff production history data numerical simulation is carried out on the dual medium geological model to fit the oil production, water production and formation pressure, and the production dynamic parameter simulation value is obtained; According to the real oil production, water production and formation pressure change of the reservoir, the porosity, permeability and oil-water two-phase permeability curve are adjusted to obtain the fitted dual medium geological model, the characteristic unit body with a square edge smaller than the minimum edge length of the dual medium geological model is extracted from the fitted dual medium geological model, and the fracture surface is set on the surface of the characteristic unit body to form a shale oil CO2 huff and puff mechanism model; Step 5: Optimizing the shale oil CO2 huff and puff scheme based on the shale CO2 huff and puff mechanism model According to the shale oil CO2 huff and puff mechanism model, the boundary conditions, soak time and injection-production system are selected as the influencing factors of CO2 huff and puff; In the boundary condition research, the mechanism model fracture surface is set to open on one side, on two sides, on four sides and on six sides; On the basis of the boundary condition research, different three-side open fracture surface models are set; Analyze the recovery rate under different boundary conditions; In the soak time research, different soak times are selected to analyze the recovery rate under different soak times; In the injection-production system research, the CO2 injection pressure is set to constant pressure injection, stepwise pressure increase and stepwise pressure decrease; Analyze the recovery rate under different injection-production systems; Obtain the recovery rate parameters of CO2 huff and puff mass transfer under different influencing factors, and select the boundary conditions, soak time and injection-production system corresponding to the maximum recovery rate.

2. The method of claim 1, wherein, The UNCONG equation is a three-parameter cubic equation, and has a general expression as shown in formula (1): （1） wherein: Z is the compressibility factor of the fluid, s , q , r respectively represent three parameters of a cubic equation of state; the three parameters are determined by equation (2): （2） of formula (2) w and u is a coefficient; A, B are coefficients related to the mixing state, determined by formula (3): （3） Since the three-parameter cubic equation of state needs to calculate the compressibility factor of the mixture, a and b need to be determined by mixing rules, and the linear mixing rule is usually used. The a and b under the linear mixing rule can be determined by formula (4); for the mixture system, if it contains N c components, the mole fraction of the i-th and j-th components in the mixture is x i , x j , the interaction coefficient between the i-th and j-th components is i 、 j k i,j ;​ （4） wherein, x i , x j are the mole fractions of the i, j components in the mixture, respectively, k i,j is N c components i , j is the interaction coefficient between the components.

3. The method of claim 1, wherein, The step 3 of establishing the shale oil corner point network geological model further includes the following steps. The step 301 collects the data required for modeling, and the required data includes stratified data, logging data and core test data; The step 302 establishes the shale oil corner point grid model by using the Petrel RE module in the Petrel software; The step 303 combines the experience of establishing the shale oil corner point network geological model, and the step length of the grid model in the X direction and the Y direction is selected as 20 m.

4. The method of claim 1, wherein, The basic parameters of the shale oil geological model in the step 3 include reservoir size, reservoir depth, reservoir pressure, matrix permeability average, matrix porosity average, main fracture permeability, main fracture porosity, main fracture opening, and main fracture half length; The processing method of the interlayer porosity heterogeneity is to use three layers of surfaces with inconsistent attribute distribution, and the porosity field generation method of the three layers of grids is to apply different random seed numbers for Gaussian dispersion and adjust the offset; and the generated three-layer porosity distribution field diagram is integrated into a typical porosity model.

5. The method of claim 1, wherein, The step 4 of establishing the shale oil CO2 injection mechanism model specifically includes the following steps. The step 401 imports the geological model established by the Petrel software into the CMG numerical simulation component model to establish a shale oil reservoir horizontal well volume fracturing dual medium geological model; the percolation equation thereof is as shown in formula (5), and the oil and gas phase material balance equation thereof is as shown in formula (6); （5） where k is permeability, unit: mD; ρ w , ρ o , ρ g are water phase density, oil phase density and gas phase density, unit: g / cm 3 ; μ w , μ o , μ g are water phase viscosity, oil phase viscosity and gas phase viscosity, unit: mP·s; x i is the mole fraction of component i, unit: mol; Φ is the phase field variable, dimensionless, F is the interfacial energy, unit: J / m 3 ; is the porosity, q is the flow rate, unit: m 3 / s; t is time, unit: s; （6） wherein z i is the total mole fraction, unit: mol; L is the liquid mole fraction, unit: mol; V is the vapor mole fraction, unit: mol; The step 402 of oil, water and formation pressure fitting is based on the shale oil reservoir horizontal well production mode, and the CO2 huff and puff production history data of the geological model are numerically simulated to fit the oil, water and formation pressure, and the production dynamic parameter simulation value is obtained; according to the real oil production, water production and formation pressure change of the reservoir, the porosity, permeability and oil-water two-phase permeability curve are adjusted to fit the daily oil production, daily water production and formation pressure; The step 403 extracts a characteristic unit body to establish a shale oil CO2 huff and puff mechanism model; the characteristic unit body with a square size less than the minimum size of the dual medium geological model is extracted in the dual medium geological model, the unit body grid is encrypted, and the fracture surface is set on the six surfaces of the unit body; when the fracture surface permeability is set to a threshold value of 100 mD, the fracture surface is considered to be opened; and when the fracture surface permeability is set to 0.05 mD, the fracture surface is considered to be closed.

6. The method of claim 5, wherein, The parameter adjustment principle of the fitting standard determined by fitting the daily oil production, daily water production and formation pressure is as follows: (1) Porosity According to the basic characteristics of the reservoir, it is known that the shale oil reservoir has low porosity and low permeability, the average porosity is 11.8%, and the porosity has a certain range of variation, and the allowable variation range is between ±5%; (2) Permeability The permeability of the shale oil reservoir is 0.1~0.01 mD, and the adjustment range of the permeability is enlarged or reduced by 2-5 times; (3) Rock fluid compressibility The fluid compressibility is determined by laboratory measurement, and is considered as a determined parameter; (4) Initial fluid saturations and initial pressure Parameters determined during the geological modeling of the reservoir are generally considered as deterministic parameters; (5) Relative permeability curves Due to the fine grid of the reservoir simulation model, the grid is severely heterogeneous, and therefore the relative permeability curves are considered as uncertain parameters, which are adjusted according to experience during the fitting process; (6) PVT high pressure properties The PVT high pressure properties of oil and water are measured by laboratory tests and have been fitted by a numerical simulation phase simulator, and are considered as deterministic parameters.

7. The method of claim 1, wherein, In step 5, the recovery parameters of CO2 huff and puff mass transfer under different influencing factors are obtained, and the boundary conditions, soak time and injection-production system corresponding to the maximum recovery of the oil well are selected, that is, the maximum recovery is obtained by studying the following three aspects, and the specific steps are as follows: Step 501 study the recovery under different boundary conditions The fracture surface is set on the surface of the mechanism model, and the model is divided into four forms of single-face opening, double-face opening, four-face opening and six-face opening; the mechanism model can intuitively reflect the communication relationship between fractures and fractures, and between fractures and matrix; different opening faces can simulate the fracture development degree of the reservoir after volume fracturing; and the mechanism model has shorter operation time; analyze the mass transfer law of CO2 huff and puff under different boundary conditions, and consider the boundary condition corresponding to the maximum recovery of the oil well as the best boundary condition; Step 502 study the recovery under different soak times According to the actual situation and previous research experience, the soak time is selected as 5 days, 10 days, 15 days and 20 days for simulation; analyze the mass transfer law of CO2 huff and puff under different soak times, and consider the soak time corresponding to the maximum recovery of the oil well as the best soak time; Step 503 study the recovery under different injection-production conditions The injection-production conditions are divided into constant pressure injection, gradient pressure injection and gradient pressure injection; among them, the constant pressure injection is that the injection pressure is increased to 1.25 times of the formation pressure each time, the gradient pressure is 1.0, 1.1, 1.2, 1.3, 1.4, 1.5 times of the formation pressure; the gradient pressure is 1.5, 1.4, 1.3, 1.2, 1.1, 1.0 times of the formation pressure; analyze the mass transfer law of CO2 huff and puff under different injection-production systems, and consider the injection-production scheme corresponding to the maximum recovery of the oil well as the best injection-production scheme.

Images