Fractured buried hill thickened oil bottom water reservoir oil-water front edge prediction method and system

Abstract

The invention discloses a fractured buried hill thick oil bottom water reservoir oil-water front prediction method and system, and belongs to the technical field of oil-gas field development engineering. The method comprises the following steps: quantitatively identifying a water invasion generation area through core testing, respectively constructing a water saturation three-dimensional distribution model and an embedded discrete fracture, and then combining the water saturation three-dimensional distribution model and the embedded discrete fracture to establish a bottom water heavy oil reservoir water invasion front dynamic evolution agent model. A deep learning algorithm is applied to realize quantitative prediction of the water invasion front of the oil reservoir; wherein the deep learning algorithm is applied, the distribution characteristics of the oil-water front of the whole reservoir can be quickly predicted based on the data set, the time required by numerical simulation is further shortened, and the prediction and recognition precision of the oil-water front of the heavy oil bottom water reservoir is greatly improved.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas field development engineering, and particularly relates to an oil-water front prediction method and system for a fractured buried hill heavy oil bottom water reservoir. Background Art

[0002] Fractured buried-hill heavy oil reservoirs with bottom water are a unique type of reservoir. Due to factors such as the development of fractures, high crude oil viscosity, strong reservoir heterogeneity, and diverse flow patterns, production wells are prone to waterlogging after commissioning, resulting in a short plateau period and high decline rates, a major constraint to stable oilfield production. The development of these reservoirs presents four specific challenges: 1. The main factors influencing bottom water intrusion are unclear, making it difficult to provide early warning of water breakthrough and prevent sudden waterlogging. 2. It is difficult to determine the optimal production capacity and production pressure differential for wells at different production stages. 3. Controlling the rate of water cut rise and slowing decline in wells with water breakthrough is challenging. 4. The complex connectivity of fractures within the reservoir and the highly irregular water invasion front make it difficult to predict the location of the front. Therefore, identifying the main factors influencing the evolution of the bottom water front in fractured buried-hill heavy oil reservoirs with bottom water, and quantitatively predicting the location and dynamic evolution of the oil-water front at different times to guide the development of an appropriate production well operation system are key issues for the efficient development of these reservoirs.

[0003] Oil-water front prediction in bottom-water reservoir development aims to accurately predict the location and trend of the oil-water interface within the well, thereby optimizing oilfield production and recovery engineering. Existing technologies for predicting the oil-water front in bottom-water reservoirs primarily include mathematical modeling and simulation, physical measurement and sensor technology, geological models and 4D seismic technology, statistical analysis and historical trends, and laboratory and field testing. Typically, a combination of methods is used in bottom-water reservoir development to ensure accurate prediction and monitoring of the oil-water front. The choice of these methods depends on the reservoir characteristics, available data, and technical resources. The combined use of these methods can improve well production efficiency, reduce waste fluid production, lower production costs, and facilitate more sustainable development of bottom-water reservoirs.

[0004] Patent Publication No. CN104183018A proposes a six-stage modeling method for characterizing gas-water distribution in water-bearing carbonate gas reservoirs. This method assumes a uniform gas-water interface height and modifies the gas-water front using the gas-water relationship of a single well. While this method is highly adaptable to simple gas-water fronts, it is not suitable for identifying gas-water fronts in complex water-bearing gas reservoirs due to geological factors and production systems, and it also fails to capture spatiotemporal evolution. Patent Publication No. CN108222919A proposes a gas-water interface monitoring method for salt-cavern gas storage during the injection and brine removal phases. This method is suitable for identifying gas-water interfaces in cavernous gas reservoirs, but is not suitable for identifying interfaces in fractured buried-hill heavy oil reservoirs with bottom water. Furthermore, the corrosive nature of subsurface fluids makes optical fiber unsuitable for long-term monitoring. Patent Publication No. CN105005074A proposes a method for identifying gas-bearing reservoirs using frequency-dependent seismic reflection coefficients. This method uses seismic frequency to implement static gas-water front identification in water-bearing gas reservoirs. However, none of the above patents mentions a method for predicting the oil-water front in a tight sandstone water-bearing gas reservoir for a heavy oil reservoir. Summary of the Invention

[0005] In view of the complex geological structure of fractured buried-hill heavy oil reservoirs in the prior art, which develop multiple groups and multiple stages of fracture systems, it is difficult to accurately determine the degree of fracture development, occurrence and characteristic parameters. At the same time, due to the high density and viscosity of heavy oil, poor fluidity and large oil-water mobility ratio, the water invasion front is extremely irregular, so the existing front prediction methods are difficult to apply to such reservoirs. In order to solve or partially solve the above problems, in the first aspect, the present invention proposes a method for predicting the oil-water front of fractured buried-hill heavy oil bottom water reservoirs, comprising the following steps:

[0006] Based on the capillary force curve and relative permeability curve of the target core, the fracture aperture limit of bottom water invasion corresponding to the oil-water co-flow area is determined;

[0007] Quantitatively interpret the fracture characteristics and water saturation of bottom water invasion in a single well of the target reservoir, and establish an original three-dimensional water saturation distribution model based on the quantitative interpretation results;

[0008] Establishing a theoretical diagram for dividing the driving stages of bottom-water reservoir oil wells based on the historical production data of the target reservoir, and determining the main influencing factors of the well's waterless production period based on the theoretical diagram for dividing the driving stages of the oil wells;

[0009] Based on the main influencing factors, an embedded discrete fracture model for bottom water heavy oil reservoir is constructed;

[0010] The original three-dimensional water saturation distribution model and the embedded discrete fracture model are combined to establish a dynamic evolution proxy model of the water invasion front of the bottom water heavy oil reservoir, and the oil-water front of the fractured buried hill heavy oil bottom water reservoir is predicted using the dynamic evolution proxy model of the water invasion front of the bottom water heavy oil reservoir.

[0011] Furthermore, the determination of the fracture aperture limit of bottom water intrusion corresponding to the oil-water co-flow region includes the following steps:

[0012] The capillary force curve and relative permeability curve of the natural core of the target reservoir were measured by mercury injection method and steady-state method respectively.

[0013] determining irreducible water and residual oil saturation of the target reservoir according to the capillary force curve and the relative permeability curve;

[0014] The capillary force formula is used to calculate the fracture aperture limit corresponding to the oil-water co-flow region.

[0015] Furthermore, the quantitative interpretation of the fracture characteristics and water saturation of bottom water invasion in a single well of the target oil reservoir and the establishment of an original three-dimensional water saturation distribution model based on the quantitative interpretation results include the following steps:

[0016] Apply seismic inversion, well logging and fracture interpretation data to quantitatively interpret the fracture density, fracture aperture, fracture occurrence and water saturation of individual wells in the target reservoir;

[0017] Determine the initial oil-water contact depth of the target reservoir based on the quantitative interpretation results;

[0018] According to the initial oil-water interface depth of the target reservoir, the original three-dimensional water saturation distribution model is established using the sequential Gaussian modeling method.

[0019] Furthermore, the method of establishing a theoretical diagram for dividing the driving stages of bottom water reservoir oil wells based on the historical production data of the target reservoir includes the following steps:

[0020] Collect historical production data of oil wells that have seen water in the target oil reservoir, including oil well pressure, liquid production, and water-free production period;

[0021] A theoretical diagram for dividing the oil well drive stages in bottom water reservoirs is established based on the historical production data.

[0022] Furthermore, the theoretical diagram of oil well driving stage division divides the breakthrough of the bottom water front to the oil production well into three typical driving stages: elastic drive, elastic drive plus bottom water drive, and complete bottom water drive;

[0023] The duration of each driving stage is the percentage of the liquid production in each driving stage to the water-free liquid production, or the percentage of the oil production in each driving stage to the water-free oil production.

[0024] Furthermore, the method of determining the main influencing factors of the waterless production period of a well based on the theoretical diagram of oil well driving stage division includes the following steps:

[0025] The duration of each driving phase and the waterless production period were used as dependent variables, and the actual field geological development parameters were used as independent variables. These geological development parameters included porosity, permeability, well-controlled reserves, water body multiple, fracture density, fracture dip, fracture aperture, water avoidance height, crude oil viscosity, production pressure difference, wellhead oil pressure, and casing pressure data.

[0026] The partial correlation analysis method is used to screen the geological development parameters and determine the main influencing factors of the waterless production period of the oil well.

[0027] Furthermore, based on the main influencing factors, the embedded discrete fracture model of the bottom water heavy oil reservoir is constructed as follows:

[0028] The embedded discrete fracture model of the bottom water heavy oil reservoir includes multiple main influencing factors, and each influencing factor considers multiple levels;

[0029] The main influencing factors include bottom water reservoir matrix porosity, permeability, fracture density, fracture aperture, fracture inclination, crude oil viscosity, water body multiple, water avoidance height and production pressure difference.

[0030] Furthermore, establishing a proxy model for the dynamic evolution of the water invasion front of a bottom-water heavy oil reservoir by combining the original three-dimensional water saturation distribution model and the embedded discrete fracture model specifically includes the following steps:

[0031] The embedded discrete fracture model is used to obtain prediction data, and the prediction data is used as output parameters of the output layer; the prediction data includes the duration of each driving stage of the oil well, the height of the bottom water front, and the water-free production period;

[0032] The main influencing factors of the screening are used as input parameters of the input layer;

[0033] Combining the prediction data of the embedded discrete fracture model and the actual oil well production data of the original three-dimensional water saturation distribution model to construct a training sample, and establishing a preliminary proxy model for the dynamic evolution of the water invasion front of the bottom water heavy oil reservoir;

[0034] The preliminary dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir is trained and adjusted to form a dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir.

[0035] Furthermore, training and adjusting the preliminary dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir to form the dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir specifically includes the following steps:

[0036] The prediction data of the embedded discrete fracture model and the actual oil well production data of the three-dimensional water saturation distribution model are used as training samples for the preliminary proxy model of the dynamic evolution of the water invasion front in bottom-water heavy oil reservoirs.

[0037] According to the back propagation of the output error of the actual water breakthrough time of the sample oil well, the connection weights of the input layer, hidden layer and output layer network of the preliminary dynamic evolution proxy model of the water invasion front in the bottom water heavy oil reservoir are dynamically adjusted;

[0038] The adjustment is stopped until the error between the water breakthrough time predicted by the preliminary proxy model of the dynamic evolution of the water invasion front of the bottom water heavy oil reservoir and the actual water breakthrough time is reduced to a set threshold;

[0039] A certain proportion of sample oil wells are reserved to perform error analysis and accuracy test on the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs, until the prediction results of the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs meet the preset accuracy requirements, and then the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs is formed.

[0040] In a second aspect, the present invention provides an oil-water front prediction system for a fractured buried-hill heavy oil bottom-water reservoir, comprising:

[0041] A fracture aperture limit determination unit is used to determine the fracture aperture limit of bottom water intrusion corresponding to the oil-water co-flow region based on the capillary force curve and relative permeability curve of the target core;

[0042] The water saturation 3D distribution model building unit is used to quantitatively interpret the fracture characteristics and water saturation of bottom water invasion in a single well of the target reservoir, and to establish the original water saturation 3D distribution model based on the quantitative interpretation results;

[0043] A main influencing factor determination unit is used to establish a theoretical diagram for dividing the driving stages of bottom water reservoir oil wells based on the historical production data of the target reservoir, and determine the main influencing factors of the well's waterless production period based on the theoretical diagram for dividing the driving stages of the oil wells;

[0044] An embedded discrete fracture model construction unit is used to construct an embedded discrete fracture model for a bottom water heavy oil reservoir based on the main influencing factors;

[0045] The prediction unit is used to establish a dynamic evolution proxy model of the water invasion front of a bottom water heavy oil reservoir by combining the original three-dimensional water saturation distribution model and the embedded discrete fracture model, and to predict the oil-water front of a fractured buried hill heavy oil bottom water reservoir using the dynamic evolution proxy model of the water invasion front of a bottom water heavy oil reservoir.

[0046] In a third aspect, the present invention provides an electronic device comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0047] a memory storing a computer program;

[0048] The processor is configured to implement the oil-water front prediction method for a fractured buried hill heavy oil bottom water reservoir when executing the program stored in the memory.

[0049] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed, the method for predicting the oil-water front in a fractured buried-hill heavy oil bottom-water reservoir is executed.

[0050] Beneficial effects of the present invention:

[0051] The present invention quantitatively identifies water invasion regions through core testing and establishes a water invasion front evolution proxy model using embedded discrete fractures. A deep learning algorithm is applied to quantitatively predict the water invasion front in this type of reservoir. The deep learning algorithm can rapidly predict the distribution characteristics of the oil-water front in the entire reservoir based on a data set, further reducing the time required for numerical simulation and significantly improving the accuracy of oil-water front prediction and identification.

[0052] The prediction method proposed in the present invention can realize the quantitative prediction of the water flooding front of fractured bottom water buried hill heavy oil reservoirs, can effectively guide the optimization of development methods of fractured heavy oil reservoirs, and has broad practical engineering application value.

[0053] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 A flow chart of a method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir proposed in an embodiment of the present invention is shown;

[0056] Figure 2 A theoretical diagram of the division of driving stages of a bottom water reservoir oil well in an embodiment of the present invention is shown;

[0057] Figure 3 A schematic diagram of an embedded discrete fracture model of an oil reservoir in an embodiment of the present invention is shown;

[0058] Figure 4 A schematic diagram of a three-dimensional distribution model of water saturation in an embodiment of the present invention is shown;

[0059] Figure 5 A schematic diagram of a neural network algorithm for a dynamic evolution proxy model of water invasion front in a bottom-water heavy oil reservoir according to an embodiment of the present invention is shown;

[0060] Figure 6 A schematic diagram of an oil-water front prediction system for a fractured bedrock buried hill bottom water reservoir according to an embodiment of the present invention is shown;

[0061] Figure 7 A schematic diagram of an electronic device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0063] The technical methods adopted by the present invention are as follows: based on natural core analysis and fluid PVT test results, an oil-water flow equation is established to clarify the movable water boundary; using historical oilfield production data, bottom hole data and reservoir geological information, an embedded discrete fracture model is applied to establish an agent model that considers fracture and fluid properties, and predicts the bottom water front position and evolution law under different fracture, water body and fluid parameter combinations; finally, data mining and machine learning techniques are applied to construct an oil-water front prediction model for this type of reservoir. Figure 1 As shown, the present invention proposes a method for predicting the oil-water front of a fractured buried-hill heavy oil bottom water reservoir, comprising the following steps:

[0064] S1: Determine the fracture aperture limit of bottom water intrusion corresponding to the oil-water co-flow region based on the capillary force curve and relative permeability curve of the target core; specifically, the following steps are included:

[0065] Mercury intrusion method and steady-state method were used to measure the capillary force curve and relative permeability curve of the natural core of the target reservoir respectively.

[0066] Determine the irreducible water and residual oil saturation of the target reservoir based on the capillary force curve and relative permeability curve;

[0067] The capillary force formula is used to calculate the fracture aperture limit corresponding to the oil-water co-flow area, thus providing a basis for quantitatively identifying the area where water invasion may occur.

[0068] The capillary force formula is as follows:

[0069] r = 0.732 / Pc;

[0070] Among them, r is the capillary radius, Pc is the capillary force, and the larger r is, the larger the crack opening and the greater the flow rate.

[0071] S2: Quantitatively interpret the fracture characteristics and water saturation of bottom water invasion in a single well of the target reservoir, and establish an original 3D water saturation distribution model based on the quantitative interpretation results. Fracture characteristics include fracture density, fracture aperture, and fracture occurrence.

[0072] Apply seismic inversion, well logging and fracture interpretation data to quantitatively interpret the fracture density, fracture aperture, fracture occurrence and water saturation of individual wells in the target reservoir;

[0073] Determine the initial oil-water interface depth of the reservoir by combining field resistivity logging interpretation and oil testing data;

[0074] Based on the PETREL commercial software platform and the initial oil-water contact depth of the reservoir, a three-dimensional water saturation distribution model of the target reservoir was established using the sequential Gaussian modeling method.

[0075] S3: establishing a theoretical diagram for dividing the driving stages of bottom-water reservoir oil wells based on the historical production data of the target reservoir, and determining the main influencing factors of the well's waterless production period based on the theoretical diagram for dividing the driving stages of the oil wells;

[0076] The specific steps include:

[0077] S31: Collect and organize historical production data of oil wells that have seen water in the target oil reservoir; the historical production data include but are not limited to oil well pressure, liquid production, water-free production period, etc.;

[0078] S32: Based on the historical production data, a theoretical chart for the division of oil well drive stages in a bottom-water reservoir is established; wherein the abscissa of the chart represents cumulative liquid production, and the ordinate includes flow temperature and flow pressure. Based on reservoir engineering principles and the variation patterns of indicators, the breakthrough of the bottom water front to the production well is divided into three typical drive stages: elastic drive Le, elastic drive Le + bottom water drive Lw, and complete bottom water drive Lw. The duration of each drive stage, LnD, is defined (n = e, w, representing elastic drive and bottom water drive, respectively): the liquid (oil) production in each drive stage is the percentage of the water-free liquid (oil) production.

[0079] In one embodiment of the present invention, the theoretical diagram for dividing the driving stages of bottom water reservoir oil wells is established as follows: Figure 2 As shown in the figure, the horizontal axis is the cumulative liquid production, the vertical axis is the flow (oil) pressure drop and flow temperature, and the circle in the figure represents the flow temperature; the slope of the curve k1 is the flow (oil) pressure drop level of the reservoir per 10,000 tons of liquid (oil) produced, k2 is the slope of the curve when the curve changes for the first time, k n is the slope of the curve when it changes for the n-1th time. L1, L2, and L3 are the duration of each driving stage, respectively. For example, L2 is the duration of each driving stage during the duration of k2, and the duration of each driving stage is the water-free liquid volume / liquid volume of each driving stage.

[0080] S33: Screening the main factors affecting the water-free production period of oil wells and their weights;

[0081] The main steps include:

[0082] The duration of each driving phase and the waterless production period are used as dependent variables, and the actual geological development parameters of the mine are used as independent variables. The geological development parameters include but are not limited to porosity, permeability, well-controlled reserves, water body multiple, fracture density, fracture dip, fracture aperture, water avoidance height, crude oil viscosity, production pressure difference, wellhead oil pressure and casing pressure data;

[0083] The partial correlation analysis method was used to screen the main influencing factors and their weights of the waterless production period of oil wells; the main influencing factors included bottom water reservoir matrix porosity, permeability, fracture density, fracture aperture, fracture dip, crude oil viscosity, water body multiple, water avoidance height and production pressure difference.

[0084] The geological and development factors that affect the water breakthrough time of oil wells are standardized to overcome the problem of large differences in the dimensions and absolute values ​​of dynamic and static factors.

[0085] Taking water body multiples as an example, the standardized processing formula used is as follows:

[0086] V=(V-Vmin) / (Vmax-Vmin)

[0087] Among them, Vmax is the maximum value of the water body multiple, Vmin is the minimum value of the water body multiple, and V represents the water body multiple after standardization.

[0088] S4: Constructing an embedded discrete fracture model for bottom water heavy oil reservoir based on the main influencing factors; in one embodiment of the present invention, the established embedded discrete fracture model is specifically as follows Figure 3 shown.

[0089] The embedded discrete fracture model for bottom-water heavy oil reservoirs reflects different geological, fracture, water body, and development parameters. It includes eight key influencing factors: matrix porosity Φ, permeability k; fracture density ρ, fracture aperture w, fracture inclination θ; crude oil viscosity μo; water body multiple V, water avoidance height h, and production pressure differential Δp. Each key influencing factor is considered at five levels, resulting in a total of N sets of embedded discrete fracture models. Based on the embedded discrete fracture model, a simulation method for embedded discrete fracture models in bottom-water heavy oil reservoirs is applied to predict key parameters such as the duration of each driving stage LnD, the bottom water front uplift height H, and the water-free production period t for each of the N embedded discrete fracture models.

[0090] S5: Combine the original three-dimensional water saturation distribution model and the embedded discrete fracture model to establish a dynamic evolution proxy model of the water invasion front of the bottom water heavy oil reservoir, and use the dynamic evolution proxy model of the water invasion front of the bottom water heavy oil reservoir to predict the oil-water front of the fractured buried hill heavy oil bottom water reservoir; In one embodiment of the present invention, the established dynamic evolution proxy model of the water invasion front of the bottom water heavy oil reservoir is specifically as follows Figure 5 shown.

[0091] The specific steps include:

[0092] The embedded discrete fracture model is used to obtain prediction data, and the prediction data is used as output parameters of the output layer; the prediction data includes the duration of each driving stage of the oil well, the height of the bottom water front, and the waterless production period;

[0093] The main influencing factors of the screening are used as input parameters of the input layer;

[0094] Combining the prediction data of the embedded discrete fracture model and the actual oil well production data of the original three-dimensional water saturation distribution model to construct a training sample, and establishing a preliminary proxy model for the dynamic evolution of the water invasion front of the bottom water heavy oil reservoir;

[0095] The preliminary dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir is trained and adjusted to form a dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir.

[0096] Specifically, training and adjusting the preliminary dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir includes the following steps:

[0097] The prediction data from the embedded discrete fracture model and the actual oil well production data from the three-dimensional water saturation distribution model were used as training samples for a proxy model of the dynamic evolution of the water invasion front in bottom-water heavy oil reservoirs. The BP neural network algorithm was used to predict the duration of each driving stage, the height of the bottom water front, and the water-free production period.

[0098] According to the back propagation of the output error of the actual water breakthrough time of the sample oil well, the connection weights of the input layer, hidden layer and output layer network of the preliminary dynamic evolution proxy model of the water invasion front in the bottom water heavy oil reservoir are dynamically adjusted;

[0099] The adjustment is stopped until the error between the water breakthrough time predicted by the preliminary bottom water heavy oil reservoir water invasion front dynamic evolution proxy model and the actual water breakthrough time is reduced to a set threshold;

[0100] A certain proportion of sample oil wells are reserved to perform error analysis and accuracy test on the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs, until the prediction results of the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs meet the preset accuracy requirements, and then the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs is formed.

[0101] In conjunction with the specific target reservoir CBII, the prediction method is described in detail, which specifically includes the following steps:

[0102] The target reservoir, CBII, is a typical fractured buried-hill heavy oil reservoir with bottom water. The relative permeability and capillary force curves of the natural core of the target interval were measured. Based on the endpoints of the relative permeability curve, the irreducible water saturation was determined to be 0.25, and the residual oil saturation was 0.32. The capillary force curve determined the capillary force at irreducible water saturation (i.e., Swc = 0.25) to be 0.013 MPa. The surface tension of mercury was assumed to be 480 mN / m, and the wetting angle of mercury to the rock solid was assumed to be 140°. The fracture aperture limit for movable water flow in the target reservoir, CBII, was calculated using the following formula:

[0103] r≈(0.735) / Pc=(0.735) / (0.013)=56.5μm

[0104] That is, only cracks with an opening greater than 56.5 μm can become water invasion channels, while in cracks with an opening less than or equal to this, formation water is bound and cannot invade.

[0105] For the CBII fractured buried-hill bottom-water heavy oil reservoir, the single-well fracture density, fracture aperture, fracture occurrence and single-well water saturation are quantitatively interpreted based on conventional logging and imaging logging data. With seismic coherence and edge detection as constraints, the original three-dimensional distribution model of water saturation of the CBII bottom-water reservoir is established. Figure 4 shown.

[0106] Based on historical production data from wells in the CBII reservoir, a theoretical diagram for the phase division of water-invasion wells in water-invasion reservoirs was developed. Taking Well A as an example, the duration of each phase during elastic flooding (Le) was 0.23, the elastic flooding (Le) combined with bottom water flooding (Lw) was 0.72, and the complete bottom water flooding (Lw) was 0.04, resulting in a waterless production period of 345 days. Similarly, the above analysis was repeated for all wells in the CBII reservoir, resulting in a water invasion phase division diagram and a waterless production period data table for each well.

[0107] Analysis of well logging, core observations, perforation data, and reservoir engineering methods revealed that the fracture density of individual wells in the CBII reservoir ranges from 1.2 to 2.1 fractures per meter, the controlled reserves per well range from 125,000 to 563,000 tons, the water-mass multiples range from 8.6 to 35, the fracture aperture ranges from 100 to 236 μm, the water-avoidance height ranges from 12 m to 30.2 m, the crude oil viscosity ranges from 82 to 134 mPa·s, the production differential pressure is 4.6 MPa, and the wellhead oil pressure and casing pressure are 12.4 MPa and 10.6 MPa, respectively. Using partial correlation analysis, factors influencing the waterless production period were identified as fracture density, well-controlled reserves, production differential pressure, water-mass multiples, water-avoidance height, and crude oil viscosity.

[0108] The ranking of factors affecting the oil-water front position in this embodiment is shown in Table 1:

[0109] Table 1

[0110] Factors affecting water breakthrough Partial correlation coefficient Significance Crack density 0.711 0 Well-controlled reserves 0.586 0 Production pressure difference -0.562 0 Water body multiples -0.152 0.361 Water avoidance height -0.105 0.53 Crude oil viscosity -1.02 0.43

[0111] Based on the analysis results of the above-mentioned main influencing factors, 32 embedded discrete fracture models were established for bottom-water buried-hill reservoirs, taking into account fracture density, well-controlled reserves, production pressure difference, water body multiple, water avoidance height, and crude oil viscosity. Based on this discrete model, an oil-water two-phase embedded discrete fracture numerical simulation model was applied to predict sample data of key parameters such as the duration of each driving stage LnD, the bottom water front uplift height H, and the water-free production period t for the 32 embedded discrete fracture models.

[0112] The duration of each driving stage LnD of the oil well and the elevation height of the bottom water front H are used as the output parameters of the output layer, and the main influencing factors of the water-free production period of the oil well are selected as the input parameters of the input layer. The actual oil well production data of the constructed embedded discrete fracture model and the three-dimensional water saturation distribution model are jointly used to construct training samples. Then, a proxy model of the dynamic evolution of the water invasion front in bottom water heavy oil reservoirs is established based on the duration of each driving stage LnD of the oil well, the elevation height H of the bottom water front, and the water-free production period t.

[0113] The proxy model for the dynamic evolution of the water invasion front in the bottom water heavy oil reservoir used in this embodiment is as follows: Figure 5As shown in the figure, the neural network model uses prediction data from an embedded discrete fracture model and actual oil well production data from a three-dimensional water saturation distribution model as training samples. A BP neural network algorithm is used to predict the duration of each drive phase, LnD, the bottom water front elevation, H, and the water-free production period, t. By backpropagating the output error with the actual water breakthrough time of the sample wells, the connection weights of the input, hidden, and output layers are dynamically adjusted until the error between the predicted and actual water breakthrough times is reduced to 5%. A certain percentage of sample wells are reserved for error analysis and accuracy testing of the BP neural network model, thus developing a method for predicting the water flooding front in fractured bottom-water buried-hill heavy oil reservoirs. The deep learning algorithm uses 80% of the data set as training samples and 20% as validation samples.

[0114] Based on the same concept of the present invention, the present invention proposes an oil-water front prediction system for fractured buried hill heavy oil bottom water reservoirs, such as Figure 6 As shown, including:

[0115] The fracture aperture limit determining unit 601 is used to determine the fracture aperture limit of bottom water invasion corresponding to the oil-water co-flow region based on the capillary force curve and relative permeability curve of the target core;

[0116] The water saturation three-dimensional distribution model building unit 602 is used to quantitatively interpret the fracture characteristics and water saturation of the bottom water invasion of a single well in the target oil reservoir, and to build an original water saturation three-dimensional distribution model based on the quantitative interpretation results;

[0117] The main influencing factor determination unit 603 is used to establish a theoretical diagram for dividing the driving stages of the bottom water reservoir oil well based on the historical production data of the target reservoir, and determine the main influencing factors of the well's waterless production period based on the theoretical diagram for dividing the driving stages of the oil well;

[0118] An embedded discrete fracture model construction unit 604 is configured to construct an embedded discrete fracture model for a bottom water heavy oil reservoir based on the main influencing factors;

[0119] The prediction unit 605 is configured to establish a dynamic evolution proxy model of the water invasion front of a bottom-water heavy oil reservoir by combining the original three-dimensional water saturation distribution model with the embedded discrete fracture model, and to predict the oil-water front of a fractured buried-hill heavy oil bottom-water reservoir using the dynamic evolution proxy model of the water invasion front of a bottom-water heavy oil reservoir.

[0120] Based on the same inventive concept, another exemplary embodiment of the present invention provides an electronic device. Figure 7As shown, the electronic device includes at least one processor 701, at least one communication interface 702, at least one memory 703 and at least one communication bus 704; wherein the processor 701, the communication interface 702 and the memory 703 communicate with each other via the communication bus 704;

[0121] Memory 703, storing computer programs;

[0122] The processor 701 is configured to implement the oil-water front prediction method for a fractured buried-hill heavy oil reservoir with bottom water when executing the program stored in the memory 703 .

[0123] Optionally, the communication interface may be an interface of a communication module, such as an interface of a GSM module; the processor may be a CPU, or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. The memory may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk storage. The memory stores a program, and the processor calls the program stored in the memory to execute some or all of the above-mentioned method embodiments.

[0124] Based on the same inventive concept, an embodiment of the present application further provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed, some or all of the above-mentioned method embodiments are implemented. Optionally, the storage medium may be a non-transitory computer-readable storage medium, for example, a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0125] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the oil-water front in a fractured buried-hill heavy oil reservoir with bottom water, characterized in that: The following steps are involved: Based on the capillary force curve and relative permeability curve of the target core, the fracture aperture limit of bottom water invasion corresponding to the oil-water co-flow area is determined; Quantitatively interpret the fracture characteristics and water saturation of bottom water invasion in a single well of the target reservoir, and establish an original three-dimensional water saturation distribution model based on the quantitative interpretation results; Establishing a theoretical diagram for dividing the driving stages of bottom-water reservoir oil wells based on the historical production data of the target reservoir, and determining the main influencing factors of the well's waterless production period based on the theoretical diagram for dividing the driving stages of the oil wells; Based on the main influencing factors, an embedded discrete fracture model for bottom water heavy oil reservoir is constructed; The original three-dimensional water saturation distribution model and the embedded discrete fracture model are combined to establish a dynamic evolution proxy model of the water invasion front of the bottom water heavy oil reservoir, and the oil-water front of the fractured buried hill heavy oil bottom water reservoir is predicted using the dynamic evolution proxy model of the water invasion front of the bottom water heavy oil reservoir.

2. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 1, wherein: Determining the fracture aperture limit of bottom water intrusion corresponding to the oil-water two-phase co-flow region comprises the following steps: The capillary force curve and relative permeability curve of the natural core of the target reservoir were measured by mercury injection method and steady-state method respectively. determining irreducible water and residual oil saturation of the target reservoir according to the capillary force curve and the relative permeability curve; The capillary force formula is used to calculate the fracture aperture limit corresponding to the oil-water co-flow region.

3. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 1, wherein: The quantitative interpretation of the fracture characteristics and water saturation of the bottom water invasion of a single well in the target oil reservoir and the establishment of an original three-dimensional water saturation distribution model based on the quantitative interpretation results include the following steps: Apply seismic inversion, well logging and fracture interpretation data to quantitatively interpret the fracture density, fracture aperture, fracture occurrence and water saturation of individual wells in the target reservoir; Determine the initial oil-water contact depth of the target reservoir based on the quantitative interpretation results; According to the initial oil-water interface depth of the target reservoir, the original three-dimensional water saturation distribution model is established using the sequential Gaussian modeling method.

4. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 1, wherein: The method of establishing a theoretical diagram for dividing the driving stages of bottom water reservoir oil wells based on the historical production data of the target reservoir includes the following steps: Collect historical production data of oil wells that have seen water in the target oil reservoir, including oil well pressure, liquid production, and water-free production period; A theoretical diagram for dividing the oil well drive stages in bottom water reservoirs is established based on the historical production data.

5. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 4, characterized in that: The theoretical diagram of oil well driving stage division divides the breakthrough of the bottom water front to the oil production well into three typical driving stages: elastic drive, elastic drive plus bottom water drive, and complete bottom water drive. The duration of each driving stage is the percentage of the liquid production in each driving stage to the water-free liquid production, or the percentage of the oil production in each driving stage to the water-free oil production.

6. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 5, characterized in that: Determining the main influencing factors of the waterless production period of a well based on the theoretical diagram of oil well driving stage division includes the following steps: The duration of each driving phase and the waterless production period were used as dependent variables, and the actual field geological development parameters were used as independent variables. These geological development parameters included porosity, permeability, well-controlled reserves, water body multiple, fracture density, fracture dip, fracture aperture, water avoidance height, crude oil viscosity, production pressure difference, wellhead oil pressure, and casing pressure data. The partial correlation analysis method is used to screen the geological development parameters and determine the main influencing factors of the waterless production period of the oil well.

7. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 6, characterized in that: The embedded discrete fracture model of bottom water heavy oil reservoir is constructed based on the main influencing factors as follows: The embedded discrete fracture model of the bottom water heavy oil reservoir includes multiple main influencing factors, and each influencing factor considers multiple levels; The main influencing factors include bottom water reservoir matrix porosity, permeability, fracture density, fracture aperture, fracture inclination, crude oil viscosity, water body multiple, water avoidance height and production pressure difference.

8. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 5, characterized in that: Combining the original three-dimensional water saturation distribution model and the embedded discrete fracture model to establish a proxy model for the dynamic evolution of the water invasion front in a bottom-water heavy oil reservoir specifically includes the following steps: The embedded discrete fracture model is used to obtain prediction data, and the prediction data is used as output parameters of the output layer; the prediction data includes the duration of each driving stage of the oil well, the height of the bottom water front, and the water-free production period; The main influencing factors of the screening are used as input parameters of the input layer; Combining the prediction data of the embedded discrete fracture model and the actual oil well production data of the original three-dimensional water saturation distribution model to construct a training sample, and establishing a preliminary proxy model for the dynamic evolution of the water invasion front of the bottom water heavy oil reservoir; The preliminary dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir is trained and adjusted to form a dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir.

9. The method for predicting the oil-water front of a fractured buried hill heavy oil bottom water reservoir according to claim 8, characterized in that: Training and adjusting the preliminary dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir to form the dynamic evolution proxy model of the water invasion front in a bottom-water heavy oil reservoir specifically includes the following steps: The prediction data of the embedded discrete fracture model and the actual oil well production data of the three-dimensional water saturation distribution model are used as training samples for the preliminary proxy model of the dynamic evolution of the water invasion front in bottom-water heavy oil reservoirs. According to the back propagation of the output error of the actual water breakthrough time of the sample oil well, the connection weights of the input layer, hidden layer and output layer network of the preliminary dynamic evolution proxy model of the water invasion front in the bottom water heavy oil reservoir are dynamically adjusted; The adjustment is stopped until the error between the water breakthrough time predicted by the preliminary bottom water heavy oil reservoir water invasion front dynamic evolution proxy model and the actual water breakthrough time is reduced to a set threshold; A certain proportion of sample oil wells are reserved to perform error analysis and accuracy test on the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs, until the prediction results of the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs meet the preset accuracy requirements, and then the dynamic evolution proxy model of the water invasion front in bottom water heavy oil reservoirs is formed.

10. A system for predicting the oil-water front in a fractured buried-hill heavy oil reservoir with bottom water, characterized in that: include: A fracture aperture limit determination unit is used to determine the fracture aperture limit of bottom water intrusion corresponding to the oil-water co-flow region based on the capillary force curve and relative permeability curve of the target core; The water saturation 3D distribution model building unit is used to quantitatively interpret the fracture characteristics and water saturation of bottom water invasion in a single well of the target reservoir, and to establish the original water saturation 3D distribution model based on the quantitative interpretation results; A main influencing factor determination unit is used to establish a theoretical diagram for dividing the driving stages of bottom water reservoir oil wells based on the historical production data of the target reservoir, and determine the main influencing factors of the well's waterless production period based on the theoretical diagram for dividing the driving stages of the oil wells; An embedded discrete fracture model construction unit is used to construct an embedded discrete fracture model for a bottom water heavy oil reservoir based on the main influencing factors; The prediction unit is used to establish a dynamic evolution proxy model of the water invasion front of a bottom water heavy oil reservoir by combining the original three-dimensional water saturation distribution model and the embedded discrete fracture model, and to predict the oil-water front of a fractured buried hill heavy oil bottom water reservoir using the dynamic evolution proxy model of the water invasion front of a bottom water heavy oil reservoir.

11. An electronic device, characterized in that: The processor, the communication interface, the memory and the communication bus are connected to each other via the communication bus. a memory storing a computer program; The processor is configured to implement the oil-water front prediction method for a fractured buried hill heavy oil bottom water reservoir according to any one of claims 1 to 9 when executing the program stored in the memory.

12. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed, the method for predicting the oil-water front of a fractured buried-hill heavy oil bottom-water reservoir according to any one of claims 1 to 9 is executed.

Images