Method and system for predicting minimum miscible pressure of CO2 miscible oil displacement
Through adaptive fuzzy neural network and ant colony evolution algorithm, the cumbersome and accuracy problems of prediction of minimum mixed pressure of CO2 and formation crude oil are solved, and higher accuracy prediction is achieved, which improves the effectiveness of CO2 displacement method and CCUS-EOR Effect.
Patent Information
- Application Number
- CN202310607495.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-26
- Publication Date
- 2024-11-26
AI Technical Summary
In the prior art, the prediction process of the minimum mixed pressure between CO2 and formation crude oil is cumbersome and the accuracy is poor, resulting in improper CO2 flooding crude oil, resulting in a decrease in crude oil recovery rate and CO2 geological burial stock or an increase in mining costs, affecting the CCUS-EOR effect.
Adaptive fuzzy neural network and ant colony evolution algorithm are used to obtain the CO2 mixed phase oil flooding minimum mixed phase pressure prediction model based on reservoir crude oil data training and learning. Accurate prediction is achieved by fuzzing the input variables, calculating the trigger intensity and optimizing membership.
The calculation accuracy and data point range of the minimum mixed pressure prediction of CO2 mixed oil dispersed oil is improved, providing a wider applicability, providing a reliable basis for the CCUS development plan, and improving the accuracy and effectiveness of the prediction.
Smart Images

Figure CN119026443A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of carbon dioxide capture, storage and enhanced oil recovery, and in particular to a method for predicting CO 2 A method and system for minimum miscible pressure in miscible oil recovery. Background Art
[0002] CCUS-EOR (carbon dioxide capture storage and enhanced oil recovery) technology can simultaneously reduce CO emissions 2 It is widely used in major oil fields for the purpose of improving the recovery rate of crude oil in oil reservoirs and is one of the effective means for the country to achieve the strategic goal of "carbon peak and carbon neutrality". 2 There are three main oil recovery modes: miscible recovery, near-miscible recovery and immiscible recovery. To achieve miscible recovery, the formation pressure must be greater than the injected CO 2 The minimum miscibility pressure for achieving miscibility with formation crude oil. 2 The minimum miscible pressure between the CO and the formation crude oil is particularly important. If the predicted value is less than the actual value of the minimum miscible pressure, then the CO 2 The method of displacing crude oil is near-miscible drive and non-miscible drive, which leads to the oil recovery rate and CO 2 The geological storage volume is greatly reduced, and the CCUS-EOR effect becomes worse; if the predicted value is greater than the actual value of the minimum miscible pressure, it will lead to a significant increase in the cost of crude oil production, and in severe cases it may cause formation rupture, which is not conducive to CO 2 Safety of geological storage.
[0003] Summary of existing technologies for determination of CO 2 There are mainly two methods for the minimum miscibility pressure of miscible oil displacement: experimental method and theoretical method. Commonly used experimental methods are mainly capillary experiment method, rising bubble instrument method, interfacial tension method, etc. However, the experimental method is cumbersome, time-consuming, labor-intensive, and expensive. Due to the shortcomings of the experiment, some experts and scholars have proposed to use theoretical methods to predict the minimum miscibility pressure. Among them, the theoretical methods mainly include empirical formula method, plate method, state equation method, and numerical simulation method. Among these theoretical methods, the empirical formula method and plate method are obtained based on a certain reservoir or specific fluid properties, with a narrow scope of application and poor accuracy. When calculating the minimum miscibility pressure, the state equation method needs to calculate the miscibility function according to the specific characteristics of the reservoir, but it is complicated and difficult to determine the specific miscibility function. The numerical simulation method first needs to fit the PVT parameters of the formation fluid, establish a numerical model, and finally determine the minimum miscibility pressure, which has poor accuracy. Summary of the invention
[0004] The embodiment of the present invention provides a method for predicting CO 2 The method and system for the minimum miscible pressure of miscible oil recovery at least partially solve the problem of predicting CO 2The minimum miscible pressure between the oil and the formation is complicated and has poor accuracy. 2 The technical effect of the minimum miscible pressure between the crude oil and the formation.
[0005] In the first aspect, to solve the above technical problems, the embodiments of the present invention provide the following technical solutions:
[0006] A prediction of CO 2 The method of minimum miscible pressure for miscible flooding includes:
[0007] Obtaining the preset type of reservoir crude oil data and making it into a data set;
[0008] The above data set is divided into training samples and test samples; based on adaptive fuzzy neural network and ant colony evolution algorithm, CO is obtained through training and learning. 2 Minimum miscible pressure prediction model for miscible oil displacement;
[0009] The data to be processed is input into the above prediction model to obtain the prediction results.
[0010] Optionally, the above training learns to obtain CO 2 The steps of the miscible flooding minimum miscible pressure prediction model also include:
[0011] Preset language rules;
[0012] Set the input layer to fuzzify the input variables and convert them into membership of different fuzzy sets;
[0013] Setting the second layer, by multiplying the above membership degrees of different fuzzy sets, the triggering strength of different fuzzy rules is calculated;
[0014] Set the third layer to normalize the above trigger strength;
[0015] Setting the fourth layer to calculate the output results of different rules based on the above preset language rules;
[0016] The fifth layer is set to defuzzify the above results and obtain the final result.
[0017] Optionally, the step of defuzzifying the above result further includes:
[0018] Take a weighted average of the results of each rule.
[0019] Optionally, after the above conversion into membership of different fuzzy sets, the above method further includes:
[0020] Based on the ant colony algorithm, according to the state transition rule, the path selected by the ants is calculated, the ants corresponding to the optimal path are obtained, and the membership degree is optimized based on the optimal path.
[0021] Optionally, after obtaining the ants corresponding to the optimal path, the method further includes:
[0022] Based on the pheromone update principle, the above ants release pheromones on the path they pass through and output the optimal path result; the update formula of the above pheromone is:
[0023] τ ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t)
[0024]
[0025] Where, ρ represents the pheromone volatility coefficient, 0<ρ<1; 1-ρ represents the pheromone residual coefficient; Represents the pheromone increment of the kth ant on its path, which is inversely proportional to the total length of the ant's path.
[0026] Optionally, after obtaining the prediction result, the method further includes:
[0027] The above prediction results were evaluated using the mean absolute relative error and the square of the correlation coefficient.
[0028] Optionally, after evaluating the above prediction results, the following steps are further included:
[0029] When the results of the above evaluation do not meet the preset requirements, the above prediction model is adjusted by increasing the number of the above data sets and / or adjusting the membership.
[0030] The second aspect is to provide a method for predicting CO 2 The system of minimum miscible pressure for miscible oil recovery includes:
[0031] A data acquisition module is used to acquire the preset type of reservoir crude oil data and make it into a data set;
[0032] The prediction model training module is used to divide the above data set into training samples and test samples; based on the adaptive fuzzy neural network and ant colony evolution algorithm, the CO 2 Minimum miscible pressure prediction model for miscible oil displacement;
[0033] The result output module is used to input the data to be processed into the above prediction model to obtain the prediction result.
[0034] In a third aspect, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps corresponding to the method of the first aspect when executing the computer program.
[0035] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps corresponding to the above method in the first aspect are implemented.
[0036] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:
[0037] Compared with the existing technology CO 2 The method for predicting the minimum miscible pressure of miscible displacement has more data points for each parameter, a wider range of values, higher calculation accuracy, and can also simultaneously analyze the non-pure CO 2 The minimum miscible pressure of miscible flooding provides a basis for CCUS development plans. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0039] Figure 1 The present invention provides a method for predicting CO 2 Flow chart of the method for minimum miscible pressure of miscible flooding;
[0040] Figure 2 The CO provided by the present invention 2 Cross-plot of the predicted value of the minimum miscible pressure of miscible displacement and the actual data;
[0041] Figure 3 The present invention provides a method for predicting CO 2 Schematic diagram of the system structure of the minimum miscible pressure of miscible oil displacement;
[0042] Figure 4 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION
[0043] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.
[0044] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0045] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0046] It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solutions of the present application, rather than limitations of the technical solutions of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments may be combined with each other.
[0047] In addition, the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.
[0048] The technical solution of the embodiment of the present invention is to solve the above technical problems, and the overall idea is as follows:
[0049] Based on the reservoir crude oil data as the data set, the adaptive fuzzy neural network (ANFIS) is used for deep learning to obtain the prediction model, and the ant colony evolution algorithm (ACO) is used to optimize the membership degree to achieve accurate boundary prediction of CO 2 The minimum miscible pressure of miscible displacement is predicted.
[0050] In an embodiment of the present invention, there is provided Figure 1 A predicted CO 2 A method for minimum miscible pressure of miscible oil displacement, the method comprising steps S101 to S103:
[0051] Step S101, obtaining a preset type of reservoir crude oil data and making it into a data set;
[0052] It should be noted that the reservoir crude oil data include reservoir temperature, composition of injected gas, C 5+ The molecular weight of the components, the ratio of the volatile components to the intermediate components in the crude oil, the critical temperature of the injected gas and the minimum miscible pressure data are shown in Table 1.
[0053]
[0054]
[0055] Table 1
[0056] Step S102: divide the data set into training samples and test samples; based on the adaptive fuzzy neural network and ant colony evolution algorithm, train and learn to obtain CO 2 Minimum miscible pressure prediction model for miscible oil displacement;
[0057] It should be noted that the input variables and output variables must be determined before training, where the input variable is the data set obtained in step S101, and the output variable is the predicted value of the minimum miscible pressure of the CO2 and crude oil system, as shown in Table 1. The data set is then divided into two different samples: training samples and test samples. The training sample contains about 75% of the data points for model training, that is, based on the adaptive fuzzy neural network and ant colony evolution algorithm, the variation law between the minimum miscible pressure of CO2 and crude oil and the input parameters is learned. The test sample includes the remaining 25% of the data points, which are used to evaluate the accuracy and effectiveness of the development model.
[0058] Detailed, training and learning to get CO 2 The steps of the miscible flooding minimum miscible pressure prediction model also include:
[0059] Preset language rules, the specific formula is:
[0060] Rule 1: if X = C 1 , Y=D 1 , then f 1 =u 1 X+v 1 Y+q 1 (1)
[0061] Rule 2: if X = C 2 , Y=D 2 , then f 2 =u 2 X+v 2 Y+q 2 (2)
[0062] Where: f 1 Represents the output result corresponding to rule 1; C 1 , D 1 is the nonlinear parameter corresponding to rule 1; u 1 , v 1 ,q 1 is the conclusion parameter corresponding to rule 1, f 2 Represents the output result corresponding to rule 2; C 2 , D 2 is the nonlinear parameter corresponding to Rule 2; u 2 , v2 ,q 2 is the conclusion parameter corresponding to rule 2. The adaptive fuzzy neural network (ANFIS) consists of a total of 5 layers, of which the first 3 layers are rule antecedents and the last 2 layers are rule consequents.
[0063] Set the input layer, fuzzify the input variables, and convert them into membership of different fuzzy sets; the specific formula is:
[0064] P i 1 =μC i (X),i=1,2 (3)
[0065] P i 1 =μD j (Y),j=3,4 (4)
[0066] Where: i represents the node; X, Y represent the input variables of node i; μC i (X), μD j (Y) represents the membership function of input variables X and Y respectively; P i 1 is the membership value corresponding to the first-layer node i.
[0067] Set up the second layer, and calculate the triggering strength of different fuzzy rules by multiplying the membership of different fuzzy sets; the specific formula is:
[0068]
[0069] Set the third layer and normalize the trigger strength; P i 3 is the trigger ratio of the ith rule in all rule bases, that is, the degree to which the ith rule is used in the entire reasoning process. The specific calculation formula is:
[0070]
[0071] Set the fourth layer to calculate the output results of different rules based on the preset language rules; the specific calculation formula is:
[0072]
[0073] Set the fifth layer to defuzzify the results and get the final result. That is, the results of each rule are weighted averaged to get the exact output, where the weight is the normalized triggering degree of different rules. The specific formula is:
[0074]
[0075] Furthermore, after converting into membership of different fuzzy sets, the method further comprises:
[0076] Based on the ant colony algorithm, according to the state transition rules, the path selected by the ants is calculated, the ants corresponding to the optimal path are obtained, and the membership is optimized based on the optimal path.
[0077] It should be noted that the parameter that has the greatest impact on the performance of the adaptive fuzzy neural network (ANFIS) is the membership degree Pi, so the ant colony evolution algorithm (ACO) is used to optimize the membership degree Pi. In detail, the movement criteria of individual ants in the ant colony algorithm are determined by the state transition rules, and the connection between individual ants is determined by the released pheromones. At the same time, the pheromones will gradually evaporate, thereby eliminating the influence of the pheromones released by the previous ant individuals and guiding the new ant individuals to search in new paths and directions. Specifically:
[0078] First, perform population initialization operation, initialize the pheromone and path of each node on the path, and place the ants evenly and randomly at each node. During initialization, each path has the same probability of being selected by the ants, and randomly generate the initial population. N is the population size.
[0079] Secondly, according to the state transition rule, the ants search and move forward between the paths. When all the ants have finished searching, the paths selected by the ants are calculated to obtain the ants corresponding to the optimal path. The probability formula for the ants’ path selection is:
[0080]
[0081] In the formula, represents the probability that the kth ant moves from point i to point j at time t; τ ij (t) represents the pheromone of ants on a certain path, α represents the pheromone factor; η il (t) represents the amount of information remaining on the ant's path at time t, which is generally the inverse of the distance between i and j; β is the inspiration factor; M k Represents the set of target points that ant k may pass through at the next moment.
[0082] Finally, after obtaining the ants corresponding to the optimal path, pheromone update is required. That is, based on the pheromone update principle, the ants release pheromones on the path they pass through, expand the search range of the path, avoid falling into the local optimum, and output the optimal path result. The pheromone update formula is:
[0083] τ ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t) (10)
[0084]
[0085] Where, ρ represents the pheromone volatility coefficient, 0<ρ<1; 1-ρ represents the pheromone residual coefficient; Represents the pheromone increment of the kth ant on its path, which is inversely proportional to the total length of the ant's path.
[0086] Step S103: input the data to be processed into the prediction model to obtain the prediction result.
[0087] Correspondingly, the parameters of the ant colony evolution algorithm (ACO) are set as follows: the population size is 60, the maximum number of iterations is 200, the pheromone factor value is 2.7, the heuristic influence factor value is 1.5, and the pheromone volatility coefficient is 0.3. The results are shown in Table 2.
[0088]
[0089]
[0090] Table 2
[0091] Furthermore, after obtaining the prediction result, the method further includes:
[0092] The prediction results were evaluated using mean absolute relative error and squared correlation coefficient.
[0093] It should be noted that when the value of the average absolute relative error percentage (AARE) is smaller and the value of the square of the correlation coefficient (R2) is closer to 1, it indicates that the accuracy of the prediction result is higher, thereby verifying and evaluating the effectiveness of the prediction model. The specific calculation formula is as follows:
[0094] The average absolute relative error (AARE) is:
[0095]
[0096] in
[0097]
[0098] The square of the correlation coefficient (R 2 )for:
[0099]
[0100] Where n is the data point for constructing the model; MMP i实验 is the experimental data of MMP in group i; MMP i预测 is the predicted value of the model for group i; is the mean of the model predictions.
[0101] Based on Table 2, the evaluation results are as follows: average absolute relative error (AARE) = 1.82, correlation coefficient square (R 2 )=0.9915, which proves that this method has high calculation accuracy. Figure 2 shown.
[0102] Furthermore, after evaluating the prediction results, it also includes:
[0103] When the evaluation results do not meet the preset requirements, the prediction model is adjusted by increasing the number of data sets and / or adjusting the membership.
[0104] It should be noted that when the result does not meet the verification requirements, the factors with greater influencing factors are adjusted. One is to increase the number of data sets and training sets, thereby improving accuracy. The second is to improve the accuracy of the membership, such as using the ant colony algorithm for optimization in the present invention.
[0105] Based on the same inventive concept, the embodiment of the present invention provides a method for predicting CO 2 Systems with minimum miscible pressure for miscible flooding, such as Figure 3 As shown, including:
[0106] A data acquisition module is used to acquire the preset type of reservoir crude oil data and make it into a data set;
[0107] The prediction model training module is used to divide the data set into training samples and test samples; based on the adaptive fuzzy neural network and ant colony evolution algorithm, the CO 2 Minimum miscible pressure prediction model for miscible oil displacement;
[0108] The result output module is used to input the data to be processed into the prediction model to obtain the prediction result.
[0109] Based on the same inventive concept, an embodiment of the present invention provides an electronic device, such as Figure 4 As shown, the invention comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a method for predicting CO is implemented. 2 The method of minimum miscible pressure for miscible oil recovery.
[0110] Based on the same inventive concept, this embodiment provides a computer-readable storage medium on which a computer program is stored, characterized in that when the program is executed by a processor, a method for predicting CO 2 The method of minimum miscible pressure for miscible oil recovery.
[0111] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0112] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0113] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A method for predicting the minimum miscible pressure of CO2 miscible flooding, characterized in that: The method comprises: Obtaining the preset type of reservoir crude oil data and making it into a data set; The data set is divided into training samples and test samples; based on the adaptive fuzzy neural network and ant colony evolution algorithm, a minimum miscible pressure prediction model of CO2 miscible flooding is obtained through training and learning; The data to be processed is input into the prediction model to obtain the prediction result.
2. The method according to claim 1, characterized in that The step of training and learning to obtain the minimum miscible pressure prediction model for CO2 miscible flooding also includes: Preset language rules; Set the input layer to fuzzify the input variables and convert them into membership of different fuzzy sets; Setting a second layer to calculate the triggering strength of different fuzzy rules by multiplying the membership degrees of different fuzzy sets; Setting a third layer to normalize the trigger strength; Setting a fourth layer to calculate output results of different rules based on the preset language rules; A fifth layer is set to defuzzify the result to obtain the final result.
3. The method according to claim 2, characterized in that The step of defuzzifying the result further includes: Take a weighted average of the results of each rule.
4. The method according to claim 2, characterized in that After the conversion to the membership of different fuzzy sets, the method further comprises: Based on the ant colony algorithm and according to the state transfer rule, the path selected by the ants is calculated to obtain the ants corresponding to the optimal path, and the membership degree is optimized based on the optimal path.
5. The method according to claim 4, characterized in that After obtaining the ants corresponding to the optimal path, the method further includes: Based on the pheromone update principle, the ants release pheromones on the path they pass through and output the optimal path result; the pheromone update formula is: t ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t) Where, ρ represents the pheromone volatility coefficient, 0<ρ<1; 1-ρ represents the pheromone residual coefficient; Represents the pheromone increment of the kth ant on its path, which is inversely proportional to the total length of the ant's path.
6. The method according to claim 1, characterized in that After obtaining the prediction result, the method further includes: The prediction results were evaluated using mean absolute relative error and squared correlation coefficient.
7. The method according to claim 6, characterized in that After evaluating the prediction results, the method further comprises: When the evaluation result does not meet the preset requirements, the prediction model is adjusted by increasing the number of data sets and / or adjusting the membership.
8. A system for predicting the minimum miscible pressure of CO2 miscible flooding, characterized in that: The system comprises: A data acquisition module is used to acquire the preset type of reservoir crude oil data and make it into a data set; A prediction model training module is used to divide the data set into training samples and test samples; based on an adaptive fuzzy neural network and an ant colony evolution algorithm, a minimum miscible pressure prediction model for CO2 miscible flooding is obtained through training and learning; The result output module is used to input the data to be processed into the prediction model to obtain the prediction result.
9. An electronic device, characterized in that: The electronic device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the method steps described in any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps corresponding to the method according to any one of claims 1 to 7 are implemented.
