Random forest algorithm-based discontinuous combination method for multiple oil displacement systems
Through the discontinuous combination method of multiple oil flooding systems based on random forest algorithm, the problem of difficult to predict oil flooding effect and optimize injection parameters in the existing technology is solved, and the effect of improving crude oil recovery is achieved, providing technology for oil field development support.
Patent Information
- Application Number
- CN202411557281.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-04
- Publication Date
- 2025-02-18
AI Technical Summary
The prior art is difficult to accurately predict the discontinuous combination of oil-fighting systems, optimize the injection parameters, and improve the crude oil recovery rate.
A variety of discontinuous combination methods of oil flooding systems based on random forest algorithm are adopted, including physical simulation optimization, numerical simulation optimization and random forest prediction model establishment, optimize injection parameters, and predict the effect of improving recovery.
It can accurately reflect the contribution of the discontinuous combination injection methods of multiple oil flooding systems to the oil increase effect, improve crude oil recovery, and provide technical support for the discontinuous combination application of multiple oil flooding systems.
Smart Images
Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] The invention relates to a discontinuous combination method of multiple oil displacement systems based on a random forest algorithm, and belongs to the technical field of oil reservoir development. Background Art
[0002] Most of my country's oil fields are continental sedimentary oil fields. Unlike marine sedimentary oil fields overseas, their reservoirs are severely heterogeneous, the underground crude oil has high viscosity, and most of them have entered the high / ultra-high water content period. Such oil reservoirs are generally not suitable for thermal recovery and miscible flooding, but are suitable for "chemical-dependent" chemical flooding. Chemical agents are mainly surfactants, polyacrylamide and its derivatives. After more than half a century of development, chemical flooding has achieved great results and benefits in mine applications. However, with the increase in the depth and breadth of oil field development, the increase in complex oil reservoirs and the difficulty of tapping potential, the development of chemical agents for chemical flooding faces new situations and new challenges: on the one hand, after chemical flooding, there is still about 50% or even more oil in the reservoir that has not been exploited, facing the major technical problem of how to further improve the recovery rate (EOR); on the other hand, oil reservoirs that have not yet implemented chemical flooding also have huge demand for and potential for improving recovery, but currently no suitable chemical agents can be screened. Therefore, according to indoor research and oilfield practical experience, several oil recovery technologies need to be applied in combination to achieve broad and subtle results, so as to solve the contradictions of microscopic, macroscopic and intra-layer heterogeneity. When multiple oil recovery systems are applied in a non-continuous combination, a large number of physical simulation experiments are needed to optimize the injection parameters and analyze the influence of the parameters, and at the same time, quantitatively predict the extent of oil increase and enhanced oil recovery. At present, there are not many research papers on related issues, and it is difficult to provide technical support for the application of the non-continuous combination of multiple oil recovery systems in the mine to enhance the recovery rate. Summary of the invention
[0003] The purpose of the present invention is to provide a method for discontinuous combination of multiple oil recovery systems based on a random forest algorithm, so as to overcome the problems that the existing research methods in the background technology are difficult to accurately predict the oil recovery effect of the discontinuous combination of multiple oil recovery systems and optimize the injection parameters. It can provide important technical support for the application of discontinuous combination of multiple oil recovery systems to improve crude oil recovery, and is an analysis method with a simple process and very obvious technical and economic effects.
[0004] The present invention provides a method for discontinuously combining multiple oil displacement systems based on a random forest algorithm, comprising the following steps: S1. Physical simulation optimization of discontinuous combined injection parameters of oil displacement system; S2, numerical simulation optimization of discontinuous combined injection parameters of oil displacement system; S3. Random forest prediction model is established.
[0005] Preferably, in step S1, the physical simulation optimization comprises the following steps: Under the condition of the same total cost of reagents, discontinuous combined flooding experiments of various flooding systems with different permeability differences were carried out for the target oil fields. The pressure during each experiment was recorded, and the water cut and recovery factor were calculated.
[0006] The physical simulation optimization also includes the following steps: The experimental results under each permeability difference were compared respectively, and the optimal alternating cycle, the optimal oil displacement system slug and concentration combination with the largest recovery enhancement were screened out.
[0007] Preferably, in step S2, the numerical simulation optimization comprises the following steps: Numerical simulation inversion is performed based on the physical simulation optimization results obtained in step S1, and the numerical simulation results at the core scale are effectively enlarged to the reservoir scale.
[0008] The numerical simulation optimization also includes the following steps: According to the orthogonal design method, the recovery rate can be improved by discontinuous combination of various oil displacement systems compared with continuous polymer injection under different experimental conditions.
[0009] Preferably, in step S3, the random forest prediction model establishment comprises the following steps: First, multiple sub-samples are randomly extracted from the original data set with replacement, and then a decision tree is trained for each sub-sample; when training each decision tree, a part of the features are randomly selected for node splitting; after each tree is trained independently, the prediction results of each tree are finally integrated by voting or averaging to output the final prediction value.
[0010] Preferably, the random forest prediction model establishment comprises the following steps: The single-factor sensitivity analysis method was used to study the relationship between different factors and the enhanced oil recovery. The training set data was learned using the random forest model. The important features in the data were filtered and extracted through the neural network layer. The production sample data was trained and fitted to establish a model for predicting the discontinuous combination displacement effect of various oil recovery systems. The important characteristics refer to injection parameters such as permeability difference, injection timing, heterogeneous system concentration, injection PV number, alternating cycle, etc.; Preferably, according to the model obtained in the previous step, the correlation between enhanced oil recovery and different influencing factors is analyzed, the importance of different factors to enhanced oil recovery is predicted, and various factors are comprehensively considered to form an enhanced oil recovery comprehensive efficiency index chart.
[0011] Preferably, in step S1, when the permeability difference is greater than 4, the gel system is first used to adjust the liquid absorption profile and then multiple oil displacement systems are discontinuously combined for injection. The present invention has the following beneficial effects: 1. Compared with the continuous polymer injection method, it can accurately reflect the contribution of the discontinuous combination injection method of multiple oil displacement systems to the oil increase effect. During the displacement process, the discontinuous combination of multiple oil displacement systems can slow down the reversal process of the liquid absorption profile, increase the liquid absorption of the low permeability layer, achieve balanced displacement, and thus increase the oil recovery rate.
[0012] 2. The random forest algorithm is easy to parallelize, has strong resistance to overfitting, is insensitive to outliers and noise, and can process high-dimensional data without the need for feature scaling. In addition, its built-in feature importance evaluation function gives it the advantages of fast training speed, high model robustness, and good feature interpretability.
[0013] 3. It can accurately analyze the correlation between enhanced oil recovery and different influencing factors, predict the importance of different factors to enhanced oil recovery, and form an enhanced oil recovery comprehensive efficiency index chart after integrating various factors. It can more intuitively and accurately determine the pros and cons of the proposed oil recovery scheme, and provide technical support for formulating non-continuous combination implementation plans for various oil recovery systems, conducting economic evaluation and risk analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 is a random forest prediction model.
[0015] Figure 2 It is the moisture content of the numerical model and physical model of the fitting inversion.
[0016] Figure 3 It is the degree of extraction of numerical and physical models of fitting inversion.
[0017] Figure 4 It is the relationship diagram of the determination coefficient of random forest prediction for improving oil recovery.
[0018] Figure 5 This is a comparison chart between the random forest prediction and the true value of the improved oil recovery.
[0019] Figure 6 This is a graph of the importance of different influencing factors predicted by random forest on improving oil recovery.
[0020] Figure 7 It is a chart of comprehensive efficiency index for improving oil recovery. DETAILED DESCRIPTION
[0021] Unless otherwise specified, the experimental methods used in the following examples are conventional methods.
[0022] Unless otherwise specified, the materials and reagents used in the following examples can be obtained from commercial sources.
[0023] Example 1. Experimental conditions 1. The heterogeneous system is a flexible microgel particle water dispersion system with an effective content of 100%; the polymer is partially hydrolyzed polyacrylamide powder (HPAM) with a relative molecular mass of 20-22 million and a solid content of 88%, and a phenolic crosslinker. The above agents are all provided by CNOOC Research Institute Co., Ltd. The experimental oil is a mixture of crude oil and kerosene in a certain proportion from the D area of S oil field, with a viscosity of 45mPa·s at 60°C. The experimental water uses simulated water prepared according to the mineralization of S oil field, and the water quality analysis is shown in Table 1.
[0024] Table 1 Water quality analysis
[0025] When conducting indoor experiments, the experimental temperature should simulate the actual temperature of the oil reservoir in the target block of the offshore oil field, which is 60°C.
[0026] 2. The experimental core is a man-made homogeneous core bonded with quartz sand and epoxy resin. The experimental model is made up of two cores connected in parallel. The external geometric dimensions are length × width × height = 30cm × 4.5cm × 4.5cm. The gas permeability is K g =200, 400, 600, 800, 1000, 1200×10 -3 μm 2 The core is sealed by epoxy resin casting.
[0027] 3. Solution preparation (1) Polymer solution In order to simulate the shearing effect on the polymer solution during preparation and injection in the mine, it was pre-sheared before the experiment so that its viscosity retention rate was 60%.
[0028] (2) Heterogeneous system 1) Before preparing the heterogeneous system solution, shake the sample bottle vigorously manually or stir with a glass rod to make the heterogeneous system evenly dispersed.
[0029] 2) Extract the stock solution of the heterogeneous system according to the designed concentration, mix it with a certain amount of simulated water, and then place it on a magnetic stirrer and stir for 15 minutes.
[0030] 3) Transfer the heterogeneous system solution into a piston container with a stirring device and start the stirrer to avoid particle aggregation and sedimentation and thus adverse effects on the experimental results.
[0031] 4. The core displacement experimental device mainly includes a horizontal flow pump, a pressure sensor and an intermediate container. Except for the horizontal flow pump, the other parts are placed in a 60℃ constant temperature box. The experimental steps are as follows: (1) Pump out the saturated water from the core at room temperature and calculate the porosity; (2) The core is saturated with oil at reservoir temperature and the oil saturation is calculated; (3) The core is composed of a parallel model, and water is driven to 98%, and the water drive recovery factor is calculated; (4) Inject the chemical solution with designed PV number into the core and calculate the water content and recovery factor; (5) The core is subsequently water-flooded to a water content of 98% and the water content and recovery factor are calculated.
[0032] In the experiment of the present invention, the injection rate of the core flooding experiment was 0.6 mL / min.
[0033] Example 2: Discontinuous combination of multiple oil recovery systems based on random forest algorithm 1. Physical simulation of non-continuous combined injection parameter optimization for multiple oil recovery systems The physical simulation experimental scheme for optimizing the discontinuous combination injection parameters of various oil displacement systems is shown in Table 2.
[0034] Table 2 Scheme and process conditions
[0035] (Note: When the permeability difference is greater than 4, it is necessary to use the gel system to adjust the liquid absorption profile before performing discontinuous combined injection of multiple oil displacement systems.) Experiments were carried out under the conditions of permeability differences of 2, 3, 4, 5, and 6. The pressure during each experiment was recorded, and the water cut and recovery factor were calculated. The experimental phenomena and principles of discontinuous combination of various oil recovery systems were analyzed.
[0036] 2. Numerical simulation of optimization of injection parameters for discontinuous combination of multiple oil displacement systems According to the above physical simulation experimental results, the STARS module of CMG software was used to conduct core experimental inversion research and establish a numerical inversion model, in which the numerical model parameters were as follows: number of grids: 30×1×2; grid size: 1cm×4.5cm×4.5cm; adsorption parameters, polymer concentration changes, residual resistance coefficient and other polymer flooding characteristic parameters were set according to the indoor experimental results. The parameters were adjusted comprehensively for fitting, and the physical simulation experimental results were analyzed with the numerical simulation results.
[0037] The numerical simulation results at the core scale are effectively enlarged to the reservoir scale. According to the characteristics of typical blocks in the S oilfield, a three-layer heterogeneous injection-production mechanism model is established. It is divided into three small layers: upper, middle and lower. The number of grids in the X, Y and Z directions is 100×18×3, the grid step is 3×10×5 m, the total thickness is 15m, the top surface depth is 800m, the horizontal permeabilities of each layer are 700mD, 1400mD and 3500mD respectively, the vertical permeability is 1 / 10 of the horizontal permeability, the porosity is 30%, and the well pattern adopts one injection and one production.
[0038] The values of reservoir and fluid composition parameters are shown in Table 3.
[0039] Table 3 Reservoir and fluid composition parameters
[0040] The optimization process of discontinuous combination injection parameters of multiple oil recovery systems involves the evaluation of five factors, as shown in Table 4. In combination with the characteristics of orthogonal experimental design, L16 (5 factors and 4 levels) orthogonal design was adopted. According to the orthogonal design results, the STARS module of CMG software was used to complete the calculation, and the recovery rate of discontinuous combination of multiple oil recovery systems was obtained under different experimental conditions compared with continuous polymer injection.
[0041] Table 4 Injection parameters and values
[0042] 3. Establish a random forest prediction model Random Forest (RF) is a method that uses a random method to build a forest containing multiple decision trees (see Figure 1 ), the calculation formula of random forest prediction value is shown in formula (1).
[0043] (1) in, Indicates A regression tree, Indicates the number of samples.
[0044] Random forest needs to be Iterations, including The specific steps of the random forest of trees are: (1) For samples equal to The data set is resampled and bootstrapped to obtain the bootstrap sample .
[0045] (2) Based on bootstrap samples Build a regression tree or classification tree The decision tree grows from the root node as follows until the pre-pruning parameters of the tree are met: 1) From Randomly select from The input variables form a random subset of the input variables , or 。
[0046] 2) From Determine the "best" grouping variable, group and generate two child nodes.
[0047] (3) Output includes Tree Random Forest.
[0048] (4) The regression prediction result or classification prediction result is the class with the highest number of votes.
[0049] (5) When analyzing the factors that affect enhanced oil recovery, the data should be normalized and the dimensions of each data should be normalized to make the results comparable.
[0050] (6) Based on the analysis of the influence of various factors on enhanced oil recovery through machine learning, in order to more clearly reflect the relationship between enhanced oil recovery and various factors, a comprehensive efficiency index is defined to judge the oil increase effect under specific working conditions. The specific calculation model is as follows: (2) In the formula, -Comprehensive oil displacement index, dimensionless - Number of key influencing factors -No. Key parameter weight coefficient -No. The key parameters are standardized and the value parameters are obtained By integrating various factors, a comprehensive efficiency index chart for enhanced oil recovery is formed, which can more intuitively and accurately determine the pros and cons of the proposed oil recovery plan.
[0051] Example 3, Experimental Results Analysis 1. Physical simulation experimental results of non-continuous combined injection parameter optimization for various oil recovery systems The results of physical simulation experiments on optimization of discontinuous combined injection parameters for various oil displacement systems are shown in Table 5.
[0052] Table 5 Experimental results
[0053] It can be seen from Table 5 that, under the conditions of reservoir permeability difference of 2-6 and the same reagent cost, compared with continuous polymer injection, the periodic injection of polymer solution and heterogeneous system can increase the recovery rate by 2.8%-4.2%. This is because the discontinuous combination injection method of multiple oil displacement systems can produce a better liquid flow diversion effect, slow down the process of liquid absorption profile reversal, and expand the swept volume, thereby increasing the oil recovery rate.
[0054] At the same time, through this group of experimental results, the alternating cycle under different permeability differences can be optimized. A reasonable alternating cycle can effectively increase the relative liquid absorption of the low permeability layer. If the alternating rounds are too few, the length of a single plug is large, and the effect of improving the seepage resistance is not good; if the alternating rounds are too frequent, the high-viscosity plug cannot form an effective sealing and shielding effect in the high permeability layer, and the subsequent oil recovery system will advance along the high permeability layer, weakening the liquid flow diversion effect, resulting in a decrease in the recovery rate increase. In this group of experiments, when the reservoir permeability difference is 2~6, the optimal alternating cycle increases with the increase of the difference, but the liquid flow diversion effect gradually deteriorates after the alternating cycle exceeds 4.
[0055] 2. Numerical simulation results of non-continuous combined injection parameter optimization for various oil recovery systems The fitting curve of the core flooding experiment results is as follows Figure 2 and Figure 3 As shown. Figure 2 and Figure 3 It can be seen that the numerical simulation results are highly consistent with the core displacement experiment results, indicating that the parameter settings are relatively reasonable. The established mechanism model can better simulate the non-continuous combination injection experimental process of various oil displacement systems. At the same time, it also verifies the reliability of indoor experiments. The parameters can be used in subsequent numerical simulation studies to provide support for regularity research and parameter optimization.
[0056] Table 6 Summary of orthogonal design and calculation results
[0057] (Note: When the permeability difference is greater than 4, it is necessary to use the gel system to adjust the liquid absorption profile before performing discontinuous combined injection of multiple oil displacement systems.) As can be seen from Table 6, the recovery rate of the discontinuous combination of multiple oil displacement systems obtained by orthogonal test is between 2.6% and 4.6%, which has a high degree of overlap with the experimental results of the discontinuous combination of multiple oil displacement systems. Therefore, the results of physical simulation experiments and numerical simulation calculations can be used for quantitative prediction of EOR of subsequent discontinuous combinations of multiple oil displacement systems.
[0058] 3. Establish a random forest prediction model The present invention is based on CMG simulation of discontinuous combination flooding of various oil recovery systems, and adopts a single factor sensitivity analysis method to study the relationship between factors such as permeability difference, injection PV number, alternating cycle, heterogeneous system concentration, injection timing and other factors on improving oil recovery. 1024 groups of production data samples are obtained, and the training set data are learned by a machine learning model. The important features in the data are filtered and extracted through numerous neural network layers, and the production sample data are trained and fitted. Finally, a prediction model is established to predict the discontinuous combination displacement effects of various oil recovery systems.
[0059] (1) Correlation analysis The correlation analysis between enhanced oil recovery and different influencing factors is shown in Table 7.
[0060] Table 7 Correlation coefficients between parameter variables and enhanced oil recovery
[0061] It can be seen from Table 7 that the enhanced oil recovery is positively correlated with the permeability difference, heterogeneous system concentration, injection PV number and alternating cycle, and negatively correlated with the injection timing.
[0062] (2) Analysis of Random Forest Algorithm 1) Enhanced oil recovery prediction The results of the determination coefficient of random forest for enhanced oil recovery prediction are shown in Figure 4 .
[0063] from Figure 4 It can be seen that the determination coefficient is 0.986, the relative absolute error (RAE) is 0.085, and the determination coefficient of the training set is almost the same as that of the test set, indicating that random forest has a good prediction effect on the recovery enhancement of discontinuous combination flooding of various oil flooding systems.
[0064] Comparison of the prediction results of random forest for enhanced oil recovery with the true value is shown in Figure 5 .
[0065] from Figure 5 It can be seen that most of the data points are very close to the straight line. x = y , indicating that the EOR predicted by random forest is close to the actual EOR value, the model training fitting accuracy is high, and it can accurately predict the discontinuous combination displacement effect of various oil displacement systems.
[0066] 2) Analysis of the importance of factors affecting oil recovery The importance of different factors predicted by random forest for improving oil recovery is shown in Table 8 and Figure 6 .
[0067] Table 8 Random forest prediction parameter variables importance for improving oil recovery
[0068] From Table 8 and Figure 6 It can be seen that the importance of various factors on improving oil recovery is ranked as follows: number of injected PVs > concentration of injected heterogeneous system > injection timing > permeability difference > alternating cycle, among which the number of injected PVs is the most important influencing factor, followed by the concentration of injected heterogeneous system.
[0069] After combining various factors, the comprehensive efficiency index chart of enhanced oil recovery is shown in Figure 7 , the advantages and disadvantages of the proposed oil recovery plan can be determined more intuitively and accurately.
[0070] from Figure 7 It can be seen that there is a good correlation between the enhanced oil recovery and the efficiency index in the two intervals of less than 0.75 and greater than 0.75. In practical applications, the permeability difference, injection PV number, alternating cycle, heterogeneous system concentration, and injection timing can be used to calculate the response comprehensive efficiency index, and the effect of the scheme can be judged in combination with the chart.
Claims
1. A method for discontinuous combination of multiple oil displacement systems based on random forest algorithm, comprising the following steps: S1. Physical simulation optimization of discontinuous combined injection parameters of oil displacement system; S2, numerical simulation optimization of discontinuous combined injection parameters of oil displacement system; S3. Random forest prediction model is established.
2. The method for discontinuously combining multiple oil displacement systems according to claim 1, characterized in that: In step S1, the physical simulation optimization includes the following steps: Under the condition of the same total cost of reagents, discontinuous combined flooding experiments of various flooding systems with different permeability differences were carried out for the target oil fields. The pressure during each experiment was recorded, and the water cut and recovery factor were calculated.
3. The method for discontinuously combining multiple oil displacement systems according to claim 2, characterized in that: The physical simulation optimization also includes the following steps: The experimental results under each permeability difference were compared respectively, and the optimal alternating cycle, the optimal oil displacement system slug and concentration combination with the largest recovery factor were screened out.
4. The method for discontinuously combining multiple oil displacement systems according to any one of claims 1 to 3, characterized in that: In step S2, the numerical simulation optimization includes the following steps: Numerical simulation inversion is performed based on the physical simulation optimization results obtained in step S1, and the numerical simulation results at the core scale are effectively enlarged to the reservoir scale.
5. The method for discontinuously combining multiple oil displacement systems according to claim 4, characterized in that: The numerical simulation optimization also includes the following steps: According to the orthogonal design method, the recovery rate can be improved by discontinuous combination of various oil displacement systems compared with continuous polymer injection under different experimental conditions.
6. The method for discontinuously combining multiple oil displacement systems according to any one of claims 1 to 5, characterized in that: In step S3, the random forest prediction model is established including the following steps: First, multiple sub-samples are randomly extracted from the original data set with replacement, and then a decision tree is trained for each sub-sample; when training each decision tree, a part of the features are randomly selected for node splitting; after each tree is trained independently, the prediction results of each tree are finally integrated by voting or averaging to output the final prediction value.
7. The method for discontinuously combining multiple oil displacement systems according to claim 6, characterized in that: Preferably, the random forest prediction model establishment comprises the following steps: The single-factor sensitivity analysis method was used to study the impact of different factors on enhanced oil recovery. The random forest model was used to learn the training set data. The important features in the data were filtered and extracted through the neural network layer. The production sample data was trained and fitted, and a model for predicting the discontinuous combination displacement effect of various oil recovery systems was established.
8. The method for discontinuously combining multiple oil displacement systems according to claim 7, characterized in that: According to the model obtained in claim 7, the correlation between enhanced oil recovery and different influencing factors is analyzed, the importance of different factors to enhanced oil recovery is predicted, and various factors are comprehensively considered to form an enhanced oil recovery comprehensive efficiency index chart.
9. The method for discontinuously combining multiple oil displacement systems according to any one of claims 1 to 8, characterized in that: In step S1, when the permeability difference is greater than 4, the gel system is first used to adjust the liquid absorption profile and then a plurality of oil displacement systems are injected in a discontinuous combination.