Method and system for determining well spacing of ultra-low permeability reservoir in CO2 flooding technology

Abstract

The invention discloses a method and system for determining the well spacing of an ultra-low permeability reservoir in a CO2 flooding technology, and relates to the technical field of reservoir development. Comprising the following steps: acquiring heterogeneity data of various ultra-low permeability reservoirs, recording well spacing data of the ultra-low permeability reservoirs, and establishing mapping with the heterogeneity data to generate a training sample data set; and based on the data set, constructing a neural network model, and performing model training to obtain a well spacing initial value prediction model. And obtaining heterogeneity data of an oil reservoir with a to-be-determined well spacing, inputting the heterogeneity data into the model to obtain an initial predicted value of the well spacing, collecting related environmental parameters, oil reservoir fluid properties and rock geological data, calculating an external influence coefficient, and determining the solubility of CO2. And dynamically correcting the well spacing initial prediction value based on the CO2 solubility accurate value and the external influence coefficient to obtain the self-adaptive well spacing, so that a basis is provided for oil well position setting, and the recovery ratio and the development efficiency of oil and gas are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil reservoir development, and is particularly concerned with a method for preparing a CO 2 A method and system for determining well spacing in flooding technology. Background Art

[0002] Ultra-low permeability reservoirs refer to oil and gas reservoirs with extremely low permeability. They are difficult to exploit and traditional oil recovery technologies often fail to achieve efficient development. Due to the poor fluidity of ultra-low permeability reservoirs, the recovery rate of oil and gas is generally low, resulting in waste of resources and reduced economic benefits. In order to solve this problem, researchers and engineers continue to explore improved oil recovery technologies, including CO 2 The driving technology has attracted wide attention due to its good prospects in improving oil and gas recovery and reducing environmental impact. 2 Injecting it into the oil reservoir can effectively improve the fluidity of the fluid in the reservoir and increase the recovery rate of oil and gas.

[0003] However, the heterogeneity of ultra-low permeability reservoirs makes the reasonable setting of well spacing a major challenge in technical implementation. The setting of well spacing not only affects CO 2 The effective injection and flow path of oil and gas are also directly related to economic benefits and effective utilization of resources. For oil reservoirs with high heterogeneity, traditional empirical rules often cannot meet actual needs, resulting in improper well spacing settings, which may cause resource waste or poor injection effects. Therefore, how to scientifically and reasonably determine the well spacing of ultra-low permeability oil reservoirs has become a technical problem that needs to be solved urgently in the current oil and gas production field.

[0004] In the prior art, the authorization announcement number CN114139464B discloses a CO 2 A method for determining the limit well spacing of flooding technology, the method comprising: establishing a CO 2 The numerical simulation component model is used to calculate the pressure and crude oil viscosity of each grid point between the injection and production wells after n days using the established numerical simulation component model; the driving pressure gradient of each grid point is calculated, and the driving pressure curve between the injection and production wells is plotted; the starting pressure gradient of each grid point is calculated, and the starting pressure gradient curve is plotted; the relationship between the driving pressure gradient curve and the starting pressure gradient curve is determined until the two curves are tangent, and the injection and production well spacing is determined to be the technical limit well spacing at this time. However, ultra-low permeability reservoirs usually have obvious heterogeneity, and only relying on the calculation of the overall driving pressure and the starting pressure gradient may not fully consider the influence of local geological characteristics (such as fractures, faults, etc.) on fluid flow, thereby reducing the accuracy and effectiveness of the well spacing determination.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a CO 2 A method and system for determining well spacing of flooding technology are provided to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: A CO2-free ultra-low permeability reservoir 2 The method for determining the well spacing of flooding technology includes the following specific steps: Taking several ultra-low permeability reservoirs with reasonable well spacing as samples, obtaining heterogeneity data of the samples and corresponding well spacing data, preprocessing the well spacing data to obtain sample well spacing data, mapping the sample well spacing data with the sample heterogeneity data one by one, and generating a training sample data set; Based on the data in the training sample data set, a neural network model is established, the heterogeneity data of the ultra-low permeability reservoir samples in the training sample data set is used as the input of the neural network model, and the sample well spacing data in the training sample data set is used as a label to train the neural network model to obtain a well spacing initial value prediction model; Obtaining heterogeneity data corresponding to the ultra-low permeability reservoir with the to-be-determined well spacing, inputting the obtained corresponding heterogeneity data into the trained well spacing initial value prediction model, obtaining the initial prediction value of the well spacing of the ultra-low permeability reservoir with the to-be-determined well spacing, and collecting environmental parameters at the formation thickness where the ultra-low permeability reservoir is located; The fluid properties and rock geological data of the ultra-low permeability reservoir with the to-be-determined well spacing are collected, and the external influence coefficient is calculated based on the fluid properties and rock geological data of the reservoir. The CO in the ultra-low permeability reservoir is determined by experiments. 2 The solubility of CO is also determined based on the environmental parameters at the thickness of the ultra-low permeability reservoir. 2 The solubility of CO is dynamically corrected to obtain 2 Solubility accurate value; Based on the obtained CO 2 The precise solubility value and external influence coefficient are used to dynamically correct the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, and the adaptive well spacing of the ultra-low permeability reservoir to be determined is obtained. The specific location of the oil well is set according to the adaptive well spacing, and the CO prediction of the ultra-low permeability reservoir is completed. 2 Determination of well spacing for flooding technology.

[0008] Furthermore, the heterogeneous data includes reservoir volume, reservoir rock type, fluid composition ratio in the reservoir, and the number of fractures and faults in the reservoir. The well spacing data corresponding to the ultra-low permeability reservoir is preprocessed specifically as normalization preprocessing, wherein the normalization preprocessing calculation is specifically based on the formula: ; In the formula, is the normalized data of the well spacing data of the ith sample, is the well spacing data of the ith sample, and Respectively represent the minimum well spacing data and the maximum well spacing data of all samples, where i is the index of the sample, ,in is the total number of samples; The method for generating the training sample data set is: mapping the sample well spacing data with the heterogeneity data of the ultra-low permeability reservoir one by one to form a corresponding grid, and recording the formed grid as the training sample data set.

[0009] Furthermore, based on the data of the training sample data set, a neural network model is established, wherein the neural network model is specifically established through a long short-term memory network model LSTM model, and the long short-term memory network model LSTM model selects an activation function and an optimization algorithm, wherein the Tanh function is selected as the activation function, and Adam is selected as the optimization algorithm of the LSTM model; the formula of the Tanh function is: ; In the formula, Represents the Tanh function, independent variable Represents the weighted sum of the neuron's input, that is, the result of weighted summation of the input received by the neuron from the previous layer; At the same time, the hyperparameters of the LSTM model are set, and the hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing number, and the number of hidden layer neurons; The number of network layers is set to 4 layers, the number of iterations is set to 200, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, the batch size is set to 256, and the number of hidden layer neurons is 32; The input of the trained initial well spacing prediction model is the heterogeneity data of the ultra-low permeability reservoir, including the reservoir volume, reservoir rock type, fluid composition ratio in the reservoir, and the number of fractures and faults in the reservoir, and the output is the corresponding well spacing data prediction value.

[0010] Furthermore, the fluid properties and rock geological data of the oil reservoir in the ultra-low permeability oil reservoir with the to-be-determined well spacing are collected, wherein the fluid properties include the viscosity and density of the crude oil, and the rock geological data include the average particle size and average density of the rock particles at the thickness of the formation where the oil reservoir is located. The external influence coefficient is calculated based on the fluid properties and rock geological data of the oil reservoir, wherein the formula for calculating the external influence coefficient is: ; In the formula, is the external influence coefficient, is the density of crude oil, is the average density of rock particles, is the average particle size of rock particles, is the permeability of crude oil, where the permeability of crude oil Based on the viscosity of crude oil, the permeability of crude oil is calculated experimentally. The calculation is based on the formula: ; In the formula, is the cross-sectional area of ​​the flow in the experiment, is the pressure difference at both ends of the flow, is the crude oil flow rate, is the length of the flow.

[0011] Furthermore, the environmental parameters of the ultra-low permeability reservoir formation thickness were used to analyze the CO 2 The solubility of CO is dynamically corrected to obtain 2 The solubility is accurate, where the environmental parameters at the formation thickness include the average ambient temperature and average pressure, where CO 2 The specific formula on which the exact value of solubility is calculated is: ; In the formula, For CO 2 Solubility exact value, CO in the test environment 2 Solubility, and are the average temperature and average pressure at the thickness of the ultra-low permeability reservoir, is the reference temperature in the test environment, is the reference pressure under the test environment, is the temperature correction coefficient, is the pressure correction factor.

[0012] Furthermore, based on the obtained CO 2The solubility precise value and external influence coefficient are used to dynamically correct the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, and the adaptive well spacing of the ultra-low permeability reservoir to be determined is obtained. The specific calculation formula of the adaptive well spacing is: ; In the formula, The adaptive well spacing of the ultra-low permeability reservoir to be determined is: is the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, is the weight coefficient of the external influence coefficient, For CO 2 The weighting factor for the exact value of solubility is , and and are all greater than 0, among which is the reference external influence coefficient.

[0013] The present invention also provides a CO 2 A system for determining well spacing in flooding technology, a CO 2 The well spacing determination system of flooding technology is used to implement the above-mentioned CO2 displacement in an ultra-low permeability reservoir. 2 The method for determining the well spacing of flooding technology includes: The training data processing module is used to take several ultra-low permeability reservoirs with reasonable well spacing as samples, obtain heterogeneity data of the samples and corresponding well spacing data, pre-process the well spacing data to obtain sample well spacing data, map the sample well spacing data with the heterogeneity data of the samples one by one, and generate a training sample data set; The prediction model training module is used to establish a neural network model based on the data in the training sample data set, take the heterogeneity data of the ultra-low permeability reservoir samples in the training sample data set as the input of the neural network model, and take the sample well spacing data in the training sample data set as the label to train the neural network model to obtain the well spacing initial value prediction model; The environmental parameter analysis module is used to obtain the heterogeneity data corresponding to the ultra-low permeability reservoir with the to-be-determined well spacing, input the obtained corresponding heterogeneity data into the well spacing initial value prediction model that has been trained, obtain the initial prediction value of the well spacing of the ultra-low permeability reservoir with the to-be-determined well spacing, and collect the environmental parameters at the formation thickness where the ultra-low permeability reservoir is located, wherein the environmental parameters at the formation thickness include the average environmental temperature and average pressure; The external influence correction module is used to collect the fluid properties and rock geological data of the ultra-low permeability reservoir in the well spacing to be determined, calculate the external influence coefficient based on the fluid properties and rock geological data of the reservoir, and determine the CO in the ultra-low permeability reservoir through experiments. 2The solubility of CO is also determined based on the environmental parameters at the thickness of the ultra-low permeability reservoir. 2 The solubility of CO is dynamically corrected to obtain 2 The precise value of solubility, the fluid properties include the viscosity and density of crude oil, and the rock geological data include the average particle size and average density of rock particles at the thickness of the formation where the oil reservoir is located; The well spacing correction determination module is used to determine the well spacing based on the obtained CO 2 The precise solubility value and external influence coefficient are used to dynamically correct the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, and the adaptive well spacing of the ultra-low permeability reservoir to be determined is obtained. The specific location of the oil well is set according to the adaptive well spacing, and the CO prediction of the ultra-low permeability reservoir is completed. 2 Determination of well spacing for flooding technology.

[0014] Compared with the prior art, the present invention has the following beneficial effects: First, a neural network model was established using heterogeneous data of ultra-low permeability reservoirs to accurately determine the CO 2 Well spacing under flooding technology. The effectiveness of this method stems from its comprehensive consideration of multi-dimensional heterogeneous characteristics, including reservoir volume, rock type, fluid composition, and the number of fractures and faults. The introduction of these heterogeneous data enables the model to capture the complexity of the reservoir more comprehensively, thereby achieving more scientific decisions on well spacing settings. Compared with traditional empirical rules, this scheme significantly improves the accuracy of well spacing prediction, reduces resource waste caused by misjudgment, and at the same time improves oil and gas recovery and increases the economic benefits of the oil field.

[0015] In addition, the dynamic correction mechanism of this scheme provides a guarantee for the accuracy of well spacing prediction. By collecting information such as environmental parameters and fluid properties in actual operation, the CO 2 Accurate dynamic correction of solubility. This process allows the calculation of well spacing to be closer to the actual situation under different operating conditions and adapt to the changes in fluid properties in the reservoir. This flexibility can not only effectively respond to the dynamic changes of the reservoir, but also optimize the well spacing setting in real time during the long-term development process, thereby maximizing the recovery rate and development efficiency of oil and gas, and further ensuring the sustainable use of energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the overall method flow of the present invention; Figure 2 It is a schematic diagram of the overall system structure of the present invention. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0018] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0019] Example: See also Figure 1 , the present invention provides a technical solution: A CO2-free ultra-low permeability reservoir 2 The method for determining the well spacing of flooding technology includes the following specific steps: Step 1: Take several ultra-low permeability reservoirs with reasonable well spacing as samples, obtain the heterogeneity data of the samples and the corresponding well spacing data, pre-process the well spacing data to obtain the sample well spacing data, map the sample well spacing data with the sample heterogeneity data one by one, and generate a training sample data set.

[0020] The plurality of ultra-low permeability oil reservoirs with reasonable well spacing specifically refer to ultra-low permeability oil reservoirs with well spacing determined to accomplish target acquisition tasks.

[0021] The heterogeneous data include reservoir volume, reservoir rock type, fluid composition ratio in the reservoir, and the number of fractures and faults in the reservoir. The well spacing data corresponding to the ultra-low permeability reservoir is preprocessed, specifically normalized preprocessing, wherein the normalized preprocessing calculation is specifically based on the formula: ; In the formula, is the normalized data of the well spacing data of the ith sample, is the well spacing data of the ith sample, and Respectively represent the minimum well spacing data and the maximum well spacing data of all samples, where i is the index of the sample, ,in is the total number of samples; The method for generating the training sample data set is: mapping the sample well spacing data with the heterogeneity data of the ultra-low permeability reservoir one by one to form a corresponding grid, and recording the formed grid as the training sample data set.

[0022] Reservoir volume refers to the total amount of recoverable reserves in the reservoir, which directly affects the setting of well spacing. A larger reservoir volume usually means higher production potential, and a larger well spacing can be selected to better distribute the well locations, thereby improving the overall recovery rate. However, if the reservoir volume is small, setting too large a well spacing may result in the inability to effectively extract fluids in certain areas, resulting in a waste of resources. Therefore, when formulating well spacing, it is necessary to fully consider the reservoir volume to ensure that each well location can evenly cover the reservoir and improve the utilization efficiency of resources.

[0023] The type of reservoir rock affects the mobility of fluids in the pores of the rock. Different rock types have different permeabilities and porosities. For example, sandstone generally has a higher permeability, while mudstone or dense rock has a lower permeability. This means that in reservoirs with higher permeability, a relatively large well spacing can be selected because the fluid can move more easily in the rock. In reservoirs with lower permeability, the well spacing needs to be reduced accordingly to ensure that the fluid flow and production can be effectively promoted. In addition, the distribution and bedding structure of different rock types will also affect the distribution and flow path of the fluid, so the influence of rock type needs to be comprehensively considered when setting the well spacing.

[0024] The composition of the fluid in the reservoir, such as the ratio of oil, gas, and water, directly affects the physical properties of the fluid, including viscosity, density, and phase behavior. For example, if the proportion of oil in the reservoir is high and the fluidity is good, the well spacing can be appropriately increased, because the oil will have better fluidity and can flow effectively within a larger well spacing. However, if the proportion of water or gas is high, it may affect the efficiency of the flow, especially in the case of multiphase flow, and it is necessary to improve the fluid recovery efficiency by reducing the well spacing. In addition, changes in fluid composition may also cause changes in phase behavior, affecting CO 2 The effect of injection, so this is also an important factor to be considered when setting the well spacing.

[0025] Fractures and faults in rock formations are important factors that affect fluid flow. Fractures can significantly increase the permeability of the reservoir and promote fluid flow, while faults may cause fluid separation or flow barriers. Therefore, in reservoirs with more fractures, a larger well spacing can usually be selected, because fractures provide more unobstructed flow channels, making the flow of fluids between wells more efficient. In reservoirs with fewer or unevenly distributed fractures and faults, the well spacing needs to be reduced to ensure that every part of the reservoir is effectively covered and developed. In addition, the complexity and directionality of the fracture network may also affect the movement path of the fluid, which also needs to be considered in the well spacing setting.

[0026] Step 2: Based on the data in the training sample data set, a neural network model is established. The heterogeneity data of the ultra-low permeability reservoir samples in the training sample data set is used as the input of the neural network model. The sample well spacing data in the training sample data set is used as a label to train the neural network model to obtain a well spacing initial value prediction model.

[0027] Based on the data of the training sample data set, a neural network model is established, wherein the neural network model is specifically established through a long short-term memory network model LSTM model. The long short-term memory network model LSTM model selects an activation function and an optimization algorithm, wherein the Tanh function is selected as the activation function, and Adam is selected as the optimization algorithm of the LSTM model; the formula of the Tanh function is: ; In the formula, Represents the Tanh function, independent variable Represents the weighted sum of the neuron's input, that is, the result of weighted summation of the input received by the neuron from the previous layer; At the same time, the hyperparameters of the LSTM model are set, and the hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing number, and the number of hidden layer neurons; The number of network layers is set to 4 layers, the number of iterations is set to 200, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, the batch size is set to 256, and the number of hidden layer neurons is 32; The input of the trained initial well spacing prediction model is the heterogeneity data of the ultra-low permeability reservoir, including the reservoir volume, reservoir rock type, fluid composition ratio in the reservoir, and the number of fractures and faults in the reservoir, and the output is the corresponding well spacing data prediction value.

[0028] Among them, LSTM is particularly suitable for processing and predicting time series data because it can remember past information and use this information to make more accurate predictions. In reservoir development, heterogeneous data such as reservoir volume, fluid composition, number of fractures, etc., often change over time, and LSTM can model this time series feature well and capture the dynamic process of fluid flow and reservoir changes.

[0029] The relationship between the heterogeneous characteristics of ultra-low permeability reservoirs and well spacing is often nonlinear. The multi-layer structure and activation function of the LSTM model enable it to learn these complex nonlinear relationships, thereby improving the prediction accuracy of the model. The LSTM network can adjust its parameters through continuous training to adapt to the characteristics of different reservoirs. With the introduction of more data, LSTM can gradually optimize its prediction ability. Therefore, when facing different types of reservoirs, the flexibility and adaptability of the LSTM model can improve the overall well spacing prediction effect.

[0030] Step 3: Obtain the heterogeneity data corresponding to the ultra-low permeability reservoir with the to-be-determined well spacing, input the corresponding heterogeneity data obtained into the trained well spacing initial value prediction model, obtain the initial prediction value of the well spacing of the ultra-low permeability reservoir with the to-be-determined well spacing, and collect the environmental parameters at the formation thickness where the ultra-low permeability reservoir is located.

[0031] The following method is used to obtain the heterogeneity data corresponding to the ultra-low permeability reservoir with the well spacing to be determined; Among them, geological exploration technology is used to conduct detailed geological surveys of oil reservoirs, collect geological profile data, build three-dimensional geological models, and then estimate the volume of oil reservoirs. These data include information such as formation thickness, area, and the shape of the oil reservoir. Through electrical logging, sonic logging, density logging and other technologies, the porosity and saturation information of the oil reservoir is obtained. These logging data can help estimate the recoverable reserves of the oil reservoir.

[0032] During the drilling process, core samples are collected for laboratory analysis. Through core observation and testing such as petrological analysis and mineral composition analysis, the rock type and its physical properties can be determined, and logging data such as density, porosity, and sound waves can be used to infer the type of rock. Specific logging curves can be used to identify different rock types and layered structures.

[0033] Laboratory analysis is performed on core or extracted fluid samples, including gas chromatography (GC), mass spectrometry and other techniques to determine the composition of the fluid, such as the ratio of crude oil, natural gas and water. During the production process of the oil field, the outflowing fluid is regularly analyzed to help obtain the current fluid composition ratio data.

[0034] By analyzing geological maps and profiles, we can understand the structural characteristics of the region, including the distribution of fractures and faults, and use the reflection and refraction characteristics of seismic waves to analyze seismic data. This can help identify and map the fractures and faults in the ground, especially over a large area.

[0035] The environmental parameters at the thickness of the formation where the ultra-low permeability reservoir is located include the average ambient temperature and average pressure. During the drilling process, temperature logging tools such as thermocouples or optical fiber temperature sensors can be used to directly measure the temperature in the formation. This method can provide high-precision temperature data.

[0036] The static and dynamic pressure data of the well are obtained through the pressure logging tool. These measurements can directly reflect the fluid pressure in the formation. By establishing a geological model and combining the known downhole pressure and fluid physical property data, the average pressure at a specific depth can be calculated. The relevant geological data and literature can also be consulted to obtain the pressure data of similar formations or adjacent oil fields for calculation and comparison.

[0037] Step 4: Collect the fluid properties and rock geological data of the ultra-low permeability reservoir in the to-be-determined well spacing, calculate the external influence coefficient based on the fluid properties and rock geological data of the reservoir, and determine the CO 2 The solubility of CO is also determined based on the environmental parameters at the thickness of the ultra-low permeability reservoir. 2 The solubility of CO is dynamically corrected to obtain 2 Solubility exact value.

[0038] The fluid properties and rock geological data of the ultra-low permeability reservoir in the to-be-determined well spacing are collected, wherein the fluid properties include the viscosity and density of the crude oil, and the rock geological data include the average particle size and average density of the rock particles at the thickness of the formation where the reservoir is located. The external influence coefficient is calculated based on the fluid properties and rock geological data of the reservoir, wherein the formula for calculating the external influence coefficient is: ; In the formula, is the external influence coefficient, is the density of crude oil, is the average density of rock particles, is the average particle size of rock particles, is the permeability of crude oil.

[0039] It should be noted that the external influence coefficient It is used to characterize the influence of reservoir external characteristic parameters on crude oil fluidity, where the external influence coefficient The smaller the value, the better the fluidity of the crude oil in the reservoir, and the well spacing can be appropriately increased to reduce construction resources.

[0040] The density of crude oil directly affects its fluidity and recoverability in the reservoir. Higher density usually means slower fluid flow, which may affect the transmission efficiency of oil and gas from the reservoir to the wellhead. In the same reservoir, crude oil with higher density may require a smaller well spacing to ensure that the fluid can flow smoothly and be effectively recovered. Therefore, the density of crude oil External influence coefficient Directly proportional.

[0041] The rock particle density represents the pore structure and complexity of the rock. More particle density means higher porosity, which may lead to increased complexity of the flow path. When the particle density increases, the fluid flow path becomes more complicated, which may lead to increased flow resistance. Therefore, the well spacing may need to be reduced accordingly to ensure the smooth output of oil and gas. Therefore, the average density of rock particles is External influence coefficient Proportional, take the natural logarithm In order to smooth and linearize the particle density Since the density of rock particles may vary greatly between different strata, directly using the number as a parameter may lead to uneven effects. By taking the logarithm, the suppression effect of large numbers on the calculation results can be reduced, so that the effect of increased particle density on ECI presents a more stable change, which helps to understand its impact on fluid flow.

[0042] The average particle size of the particles affects the pore size of the rock and the flow channel of the fluid. Smaller particle size usually means higher specific surface area and higher flow resistance. If the particle size is small, the well spacing may need to be adjusted to adapt to the change in flow, because fine particles may lead to higher seepage resistance, thus requiring more frequent well layout. Therefore, the average particle size of rock particles External influence coefficient Inversely proportional, square root operation It is generally used to normalize measurements, especially when dealing with parameters related to flow and porosity. The particle size of rock particles directly affects the flow paths of fluids in the reservoir, and larger particle sizes generally mean larger pores. By taking the square root, the effect of particle size on flow conditions can be better reflected, especially when considering flow resistance.

[0043] Permeability is the ability of fluid to pass through rock and is a core parameter that affects fluid flow. High permeability means that fluid can flow more easily, while low permeability indicates that flow is restricted. Lower permeability may lead to a decrease in flow rate, so a smaller well spacing is required to ensure that the fluid can be effectively extracted. Therefore, the permeability of crude oil is External influence coefficient Inversely proportional, permeability is a key parameter that describes the ability of fluid to flow in rocks. This is to reflect its nonlinear effect on fluid flow. An increase in permeability usually significantly increases the flow capacity of a fluid, so using an exponential function can better demonstrate the strong effect of permeability on flow behavior.

[0044] The permeability of crude oil Based on the viscosity of crude oil, the permeability of crude oil is calculated experimentally. The calculation is based on the formula: ; In the formula, is the cross-sectional area of ​​the flow in the experiment, is the pressure difference at both ends of the flow, is the crude oil flow rate, is the length of the flow.

[0045] The cross-sectional area of ​​the flow in the experiment is , the pressure difference between the two ends of the flow , Crude oil flow and the length of the flow Mean the data recorded during the experiment.

[0046] Based on the environmental parameters collected at the formation thickness of the ultra-low permeability reservoir, the solubility of CO2 is dynamically corrected to obtain the precise value of CO2 solubility, wherein the environmental parameters at the formation thickness include the average ambient temperature and average pressure, and the specific formula for calculating the precise value of CO2 solubility is: ; In the formula, For CO 2 Solubility exact value, CO in the test environment 2 Solubility, and are the average temperature and average pressure at the thickness of the ultra-low permeability reservoir, is the reference temperature in the test environment, is the reference pressure under the test environment, is the temperature correction coefficient, is the pressure correction factor.

[0047] It should be noted that CO 2 Solubility Exact Value The larger the value, the higher the CO 2 The higher the solubility in crude oil, the better the fluidity of crude oil. As the fluidity increases, the well spacing can be appropriately increased.

[0048] Temperature has a significant effect on gas solubility. Generally speaking, an increase in temperature leads to a decrease in gas solubility, and vice versa. Through this linear correction term, CO 2 The solubility of the oil reservoir is used to reflect the actual situation under different temperature conditions. Therefore, the average temperature at the thickness of the ultra-low permeability reservoir is With CO 2 Solubility Exact Value Inversely proportional.

[0049] An increase in pressure generally increases the solubility of a gas, which has been confirmed in many gas solubility studies. By expressing the change in pressure as a relative change, the effect of pressure change on CO can be effectively reflected. 2 The influence of solubility and therefore the average pressure at the thickness of the formation in ultra-low permeability reservoirs With CO 2 Solubility Exact Value Directly proportional.

[0050] Among them, the temperature correction factor and pressure correction factor The solubility test is carried out at different temperatures and pressures, and its value is , .

[0051] The viscosity of crude oil at different temperatures can be directly measured using a rotational viscometer, such as a Brookfield viscometer. The device rotates at a specific speed and measures the torque required, which allows the viscosity of the crude oil to be calculated.

[0052] The density of crude oil can be directly measured using a digital density meter such as an Archimedes density meter. This device uses the principle of buoyancy to measure the buoyancy of an object in a fluid, thereby obtaining the density value of the crude oil.

[0053] During the drilling process, core samples are collected to obtain actual rock data. Cores are solid samples extracted from oil reservoirs, which can represent the actual situation of the formation. The core samples are graded through sieves with different apertures to obtain the particle size distribution of rock particles. According to the screening results, the average particle size can be calculated; using image processing technology, the core samples are photographed and analyzed, and the number and size of particles can be automatically counted to provide more accurate data. Consult the regional geological survey report or literature to obtain the rock characteristics of the formation where the oil reservoir is located, including information such as the average particle size and number of particles. These data are often counted and analyzed based on multiple wells, which can provide a reference for the acquisition of parameters. The specific method for obtaining the average density of rock particles is as follows: select several sub-areas in the oil reservoir area for exploration, record the number of rock particles in each sub-area, and obtain the rock particle density of the sub-area based on the number of rock particles in each sub-area divided by the area of ​​the corresponding sub-area. The average density of rock particles is calculated by the rock particle density of each sub-area.

[0054] Step 5: Based on the obtained CO 2 The precise solubility value and external influence coefficient are used to dynamically correct the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, and the adaptive well spacing of the ultra-low permeability reservoir to be determined is obtained. The specific location of the oil well is set according to the adaptive well spacing, and the CO prediction of the ultra-low permeability reservoir is completed. 2 Determination of well spacing for flooding technology.

[0055] Based on the obtained CO 2 The solubility precise value and external influence coefficient are used to dynamically correct the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, and the adaptive well spacing of the ultra-low permeability reservoir to be determined is obtained. The specific calculation formula of the adaptive well spacing is: ; In the formula, The adaptive well spacing of the ultra-low permeability reservoir to be determined is: is the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, is the weight coefficient of the external influence coefficient, For CO 2 The weighting factor for the exact value of solubility is , and and are all greater than 0, among which is the reference external influence coefficient.

[0056] The reference external influence coefficient According to the experience of experts, the parameters used in the above calculation process are determined. The common values ​​of the parameters in the general ultra-low permeability reservoirs are used. According to the determined common values, the reference external influence coefficient is calculated by the calculation formula of the external influence coefficient. .

[0057] Among them, since the external influence coefficient has been explained above and CO 2 Solubility Exact Value The influence of well spacing setting is not elaborated here, and the square root is used This means that when the external influence coefficient increases, its influence on the adaptive well spacing is gradual. The increase in well spacing will not cause too much reduction in the well spacing, thus avoiding resource waste or development risks caused by sudden changes.

[0058] Using the Logarithmic Function Indicates CO 2 The effect of solubility is gradually decreasing. When the solubility is low, the impact on well spacing is large, but as the solubility increases, its marginal impact gradually decreases, which is consistent with the actual situation because under high solubility conditions, the impact of further increases on fluid behavior will become relatively small.

[0059] External factors usually have a greater impact on well spacing than CO 2 This is because the optimization of well spacing is not only affected by gas solubility, but also by the combined effects of multiple factors such as the physical properties of the formation, injection and production methods, etc. , and and Both are greater than 0.

[0060] See also Figure 2 The present invention also provides a system for determining the well spacing of CO2 flooding technology in ultra-low permeability oil reservoirs. 2 The well spacing determination system of flooding technology is used to implement the above-mentioned CO2 displacement in an ultra-low permeability reservoir. 2 The method for determining the well spacing of flooding technology includes: The training data processing module is used to take several ultra-low permeability reservoirs with reasonable well spacing as samples, obtain heterogeneity data of the samples and corresponding well spacing data, pre-process the well spacing data to obtain sample well spacing data, map the sample well spacing data with the heterogeneity data of the samples one by one, and generate a training sample data set; The prediction model training module is used to establish a neural network model based on the data in the training sample data set, take the heterogeneity data of the ultra-low permeability reservoir samples in the training sample data set as the input of the neural network model, and take the sample well spacing data in the training sample data set as the label to train the neural network model to obtain the well spacing initial value prediction model; The environmental parameter analysis module is used to obtain the heterogeneity data corresponding to the ultra-low permeability reservoir with the to-be-determined well spacing, input the obtained corresponding heterogeneity data into the well spacing initial value prediction model that has been trained, obtain the initial prediction value of the well spacing of the ultra-low permeability reservoir with the to-be-determined well spacing, and collect the environmental parameters at the formation thickness where the ultra-low permeability reservoir is located, wherein the environmental parameters at the formation thickness include the average environmental temperature and average pressure; The external influence correction module is used to collect the fluid properties and rock geological data of the ultra-low permeability reservoir in the well spacing to be determined, calculate the external influence coefficient based on the fluid properties and rock geological data of the reservoir, and determine the CO in the ultra-low permeability reservoir through experiments. 2 At the same time, the solubility of CO2 is dynamically corrected based on the environmental parameters at the thickness of the ultra-low permeability reservoir, and the CO 2 The precise value of solubility, the fluid properties include the viscosity and density of crude oil, and the rock geological data include the average particle size and average density of rock particles at the thickness of the formation where the oil reservoir is located; The well spacing correction determination module is used to determine the well spacing based on the obtained CO 2 The precise solubility value and external influence coefficient are used to dynamically correct the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, and the adaptive well spacing of the ultra-low permeability reservoir to be determined is obtained. The specific location of the oil well is set according to the adaptive well spacing, and the CO prediction of the ultra-low permeability reservoir is completed. 2 Determination of well spacing for flooding technology.

[0061] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0062] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0063] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0064] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for determining the well spacing of CO2 flooding technology in ultra-low permeability reservoirs, characterized in that: The specific steps include: Taking several ultra-low permeability reservoirs with reasonable well spacing as samples, obtaining heterogeneity data of the samples and corresponding well spacing data, preprocessing the well spacing data to obtain sample well spacing data, mapping the sample well spacing data with the sample heterogeneity data one by one, and generating a training sample data set; Based on the data in the training sample data set, a neural network model is established, the heterogeneity data of the ultra-low permeability reservoir samples in the training sample data set is used as the input of the neural network model, and the sample well spacing data in the training sample data set is used as a label to train the neural network model to obtain a well spacing initial value prediction model; Obtaining heterogeneity data corresponding to the ultra-low permeability reservoir with the to-be-determined well spacing, inputting the obtained corresponding heterogeneity data into the trained well spacing initial value prediction model, obtaining the initial prediction value of the well spacing of the ultra-low permeability reservoir with the to-be-determined well spacing, and collecting environmental parameters at the formation thickness where the ultra-low permeability reservoir is located; Collect the fluid properties and rock geological data of the ultra-low permeability reservoir in the to-be-determined well spacing, calculate the external influence coefficient based on the fluid properties and rock geological data of the reservoir, and determine the solubility of CO2 in the ultra-low permeability reservoir through experiments. At the same time, dynamically correct the solubility of CO2 based on the environmental parameters at the formation thickness of the ultra-low permeability reservoir collected to obtain the accurate value of CO2 solubility; Based on the obtained precise value of CO2 solubility and external influence coefficient, the initial predicted value of well spacing for the ultra-low permeability reservoir to be determined is dynamically corrected to obtain the adaptive well spacing for the ultra-low permeability reservoir to be determined. The specific location of the oil well is set according to the adaptive well spacing to complete the determination of the well spacing of CO2 flooding technology for ultra-low permeability reservoirs.

2. The method for determining the well spacing of CO2 flooding technology in ultra-low permeability reservoirs according to claim 1, characterized in that: The heterogeneous data include reservoir volume, reservoir rock type, fluid composition ratio in the reservoir, and the number of fractures and faults in the reservoir. The well spacing data corresponding to the ultra-low permeability reservoir is preprocessed, specifically normalized preprocessing, wherein the normalized preprocessing calculation is specifically based on the formula: ； In the formula, is the normalized data of the well spacing data of the ith sample, is the well spacing data of the ith sample, and Respectively represent the minimum well spacing data and the maximum well spacing data of all samples, where i is the index of the sample, ,in is the total number of samples; The method for generating the training sample data set is: mapping the sample well spacing data with the heterogeneity data of the ultra-low permeability reservoir one by one to form a corresponding grid, and recording the formed grid as the training sample data set.

3. The method for determining the well spacing of CO2 flooding technology in ultra-low permeability reservoirs according to claim 2, characterized in that: Based on the data of the training sample data set, a neural network model is established, wherein the neural network model is specifically established through a long short-term memory network model LSTM model. The long short-term memory network model LSTM model selects an activation function and an optimization algorithm, wherein the Tanh function is selected as the activation function, and Adam is selected as the optimization algorithm of the LSTM model; the formula of the Tanh function is: ； In the formula, Represents the Tanh function, independent variable Represents the weighted sum of the neuron's input, that is, the result of weighted summation of the input received by the neuron from the previous layer; At the same time, the hyperparameters of the LSTM model are set, and the hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing number, and the number of hidden layer neurons; The number of network layers is set to 4 layers, the number of iterations is set to 200, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, the batch size is set to 256, and the number of hidden layer neurons is 32; The input of the trained initial well spacing prediction model is the heterogeneity data of the ultra-low permeability reservoir, including the reservoir volume, reservoir rock type, fluid composition ratio in the reservoir, and the number of fractures and faults in the reservoir, and the output is the corresponding well spacing data prediction value.

4. The method for determining the well spacing of CO2 flooding technology in ultra-low permeability reservoirs according to claim 1, characterized in that: The fluid properties and rock geological data of the ultra-low permeability reservoir in the to-be-determined well spacing are collected, wherein the fluid properties include the viscosity and density of the crude oil, and the rock geological data include the average particle size and average density of the rock particles at the thickness of the formation where the reservoir is located. The external influence coefficient is calculated based on the fluid properties and rock geological data of the reservoir, wherein the formula for calculating the external influence coefficient is: ； In the formula, is the external influence coefficient, is the density of crude oil, is the average density of rock particles, is the average particle size of rock particles, is the permeability of crude oil, where the permeability of crude oil Based on the viscosity of crude oil, the permeability of crude oil is calculated experimentally. The calculation is based on the formula: ； In the formula, is the cross-sectional area of ​​the flow in the experiment, is the pressure difference at both ends of the flow, is the crude oil flow rate, is the length of the flow.

5. The method for determining the well spacing of CO2 flooding technology in ultra-low permeability reservoirs according to claim 4, characterized in that: Based on the environmental parameters collected at the formation thickness of the ultra-low permeability reservoir, the solubility of CO2 is dynamically corrected to obtain the precise value of CO2 solubility, wherein the environmental parameters at the formation thickness include the average ambient temperature and average pressure, and the specific formula for calculating the precise value of CO2 solubility is: ； In the formula, is the precise value of CO2 solubility, is the CO2 solubility under the test environment, and are the average temperature and average pressure at the thickness of the ultra-low permeability reservoir, is the reference temperature in the test environment, is the reference pressure in the test environment, is the temperature correction coefficient, is the pressure correction factor.

6. The method for determining the well spacing of CO2 flooding technology in ultra-low permeability reservoirs according to claim 5, characterized in that: Based on the obtained CO2 solubility accurate value and external influence coefficient, the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined is dynamically corrected to obtain the adaptive well spacing of the ultra-low permeability reservoir to be determined, where the specific calculation formula of the adaptive well spacing is: ； In the formula, The adaptive well spacing of the ultra-low permeability reservoir to be determined is: is the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined, is the weight coefficient of the external influence coefficient, is the weight coefficient of the exact value of CO2 solubility, where , and and are all greater than 0, among which is the reference external influence coefficient.

7. A system for determining well spacing of CO2 flooding technology in ultra-low permeability reservoirs, characterized by: The system for determining the well spacing of CO2 flooding technology in ultra-low permeability oil reservoirs is used to execute the method for determining the well spacing of CO2 flooding technology in ultra-low permeability oil reservoirs according to any one of claims 1 to 6, specifically comprising: The training data processing module is used to take several ultra-low permeability reservoirs with reasonable well spacing as samples, obtain heterogeneity data of the samples and corresponding well spacing data, pre-process the well spacing data to obtain sample well spacing data, map the sample well spacing data with the heterogeneity data of the samples one by one, and generate a training sample data set; The prediction model training module is used to establish a neural network model based on the data in the training sample data set, take the heterogeneity data of the ultra-low permeability reservoir samples in the training sample data set as the input of the neural network model, and take the sample well spacing data in the training sample data set as the label to train the neural network model to obtain the well spacing initial value prediction model; The environmental parameter analysis module is used to obtain the heterogeneity data corresponding to the ultra-low permeability reservoir with the to-be-determined well spacing, input the obtained corresponding heterogeneity data into the well spacing initial value prediction model that has been trained, obtain the initial prediction value of the well spacing of the ultra-low permeability reservoir with the to-be-determined well spacing, and collect the environmental parameters at the formation thickness where the ultra-low permeability reservoir is located, wherein the environmental parameters at the formation thickness include the average environmental temperature and average pressure; An external influence correction module is used to collect fluid properties and rock geological data of the ultra-low permeability reservoir in the to-be-determined well spacing, calculate the external influence coefficient based on the fluid properties and rock geological data of the reservoir, and determine the solubility of CO2 in the ultra-low permeability reservoir through experiments. At the same time, the solubility of CO2 is dynamically corrected based on the environmental parameters at the formation thickness of the ultra-low permeability reservoir collected to obtain the accurate value of CO2 solubility. The fluid properties include the viscosity and density of crude oil, and the rock geological data include the average particle size and average density of rock particles at the formation thickness of the reservoir; The well spacing correction determination module is used to dynamically correct the initial predicted value of the well spacing of the ultra-low permeability reservoir to be determined based on the obtained CO2 solubility accurate value and the external influence coefficient, obtain the adaptive well spacing of the ultra-low permeability reservoir to be determined, set the specific position of the oil well according to the adaptive well spacing, and complete the determination of the well spacing of the CO2 flooding technology for the ultra-low permeability reservoir.

Images