Methods for determining reservoir recovery based on analogy principle

Abstract

This invention relates to a method for determining reservoir recovery based on analogy principles, comprising: statistically analyzing the number and types of reservoirs within the target reservoir area, selecting reservoir information contained within each reservoir type; assigning values ​​to parameters in the reservoir information, dimensionlessly converting the parameters, using recovery rate as the optimization objective, performing main control factor analysis using a neural network algorithm optimized by a genetic algorithm, setting thresholds for the number of sensitive parameters, and establishing a basic parameter sequence library of analogous reservoirs based on the SEC analogy criterion; inputting parameters of all reservoirs of the same type as the target reservoir within the oil area according to the basic parameter sequence library of analogous reservoirs, and performing three rounds of screening for similar reservoirs; determining the recovery rate based on the production dynamics of the target reservoir; if the target reservoir has a long development time and a declining trend, then using the decline rate of analogous reservoirs as an analogy, and calculating the final recovery rate using a dynamic method. The established analogous reservoir parameter sequence library is applicable to any reservoir within the entire oil-bearing area where the target reservoir is located, making the analogy process more systematic and standardized.

Description

Technical Field

[0001] This invention relates to the field of oil and gas field development recovery rate assessment, and particularly to a method for determining reservoir recovery rate based on the principle of analogy. Background Technology

[0002] Oil recovery rate is one of the most important parameters in oilfield development. It serves as the basis for formulating oilfield development plans, evaluating development effectiveness, and developing adjustment and potential tapping schemes. Therefore, it is essential to scientifically and accurately predict reservoir recovery rate. Currently, methods for determining recovery rate include analogy method, core analysis method, relative permeability curve method, empirical formula method, and numerical simulation method.

[0003] The analogy method plays a crucial role in oil and gas field recovery rate prediction and marketable reserve assessment. In oil and gas field recovery rate prediction, when studying the development patterns of reservoirs in their early stages of development and evaluating relevant development indicators, conventional formula methods and numerical simulations are often unusable due to insufficient dynamic production data. In such cases, the analogy method is frequently used to predict recovery rates. Similarly, the analogy method is important in marketable reserve assessment, especially in confirming undeveloped reserves. However, the use of the analogy method currently faces several problems. Firstly, differences in researchers' abilities and experience lead to variations in the selected analogy parameters or overly subjective judgments of the analogy results, significantly reducing their scientific validity and rationality. Secondly, there is a lack of systematic and standardized analogy procedures in literature or patents. Many researchers overlook the importance of analogy, relying solely on experience and simple comparisons of a few parameters, resulting in the underutilization of a large amount of effective reservoir information and failing to realize the practical value of the analogy method. Summary of the Invention

[0004] To address the aforementioned problems, the purpose of this invention is to provide a novel method for determining reservoir recovery based on analogy principles. This method solves the problem of analogy between the recovery rates of different types of reservoirs within the same oilfield. For different types of reservoirs, a neural network optimized by a genetic algorithm is used to rank the main controlling factors affecting reservoir recovery, overcoming the drawbacks of strong subjectivity in nonlinear numerical methods and the high difficulty and long simulation time of numerical simulation methods. Based on this, an analogy parameter sequence library for different types of reservoirs is established. The similarity between the analogous reservoir and the target reservoir is quantitatively analyzed using the analytic hierarchy process (AHP). The synergistic effect of AHP and the neural network greatly reduces the errors caused by human subjectivity in the hierarchical structure analysis and modeling process, making the analogy results more accurate and reasonable.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for determining reservoir recovery based on the analogy principle includes:

[0007] The number and types of oil reservoirs in the target oilfield are statistically analyzed, oil reservoir information is selected from the oil reservoir types, and data related to oil reservoir information, including geological structure data, oil reservoir development atlases, and production dynamic data, are collected.

[0008] The parameters in the reservoir information are assigned values ​​and dimensionless. The recovery rate is used as the optimization target. The main control factor analysis is performed using a neural network algorithm optimized by the genetic algorithm. A threshold for the number of sensitive parameters is set. The selection principle of the number threshold is to ensure that the selected parameter categories involve geological data, engineering data and development data. Parameters with sensitivity ranking before the threshold are included in the parameter sequence for analog analysis. A basic parameter sequence library of analog reservoirs in the oil-bearing area where the target reservoir is located is established.

[0009] Based on the analog reservoir basic parameter sequence library, input all reservoir parameters of the same type as the target reservoir in the oilfield, and perform three screenings of similar reservoirs: The first screening of similar reservoirs selects reservoirs with the same geological strata, sedimentary environment, geological structure, and driving mechanism as the target reservoir according to the SEC analogy basic criteria; the second screening of similar reservoirs selects reservoirs with similar values ​​from the first screening of similar reservoirs according to the principle of numerical proximity; the third screening of similar reservoirs uses the analytic hierarchy process to perform analogy index similarity analysis.

[0010] The recovery rate is determined based on the production dynamics of the target reservoir. If the target reservoir has been developed for a long time and has a declining trend, the decline rate of an analog reservoir is used as a comparison, and the final recovery rate is calculated using the dynamic method. If the target reservoir has not been developed or has been developed but does not have a declining trend, the recovery rate of the analog reservoir is used as the benchmark. The benchmark value of the recovery rate is adjusted according to the similarities and differences between the analog reservoir and the target oil and gas reservoir in terms of geological understanding and development characteristics, and is used as the final analog recovery rate.

[0011] Oil reservoir types are classified based on lithology, permeability, and driving mechanism.

[0012] There are three types of assignment: assigning a value between 0 and 1 to parameters without a specific value; assigning the average value of a parameter with a value within a certain range; and assigning the true value to a parameter with a definite value.

[0013] The threshold setting principle ensures that the selected parameter categories involve geological data, engineering data, and development data.

[0014] The parameters include any one of the following groups: permeability, effective thickness, injection-production ratio, porosity, viscosity, original reservoir pressure, well density, drive mode, well type, coefficient of variation, extraction mode, saturation pressure, burial depth, density, temperature, gas-oil ratio, development time, crude oil volume factor, oil-bearing area, lithology, reserve calculation parameters, reservoir and fluid properties, and development parameters.

[0015] The Analytic Hierarchy Process (AHP) includes constructing a comparison matrix based on an importance scale interpretation.

[0016] The comparison matrix is ​​simulated using a neural network method, with the importance of the parameters increasing in order. Each parameter corresponds to a sensitivity, and the importance scale of each type of parameter is determined based on the actual difference in sensitivity.

[0017] The present invention has the following advantages due to the adoption of the above technical solutions:

[0018] 1. Based on intelligent data analysis technology, this invention proposes a method for establishing analog reservoir parameter sequences, which transforms the complex and cumbersome parameter selection problem between analog reservoirs into an analysis problem of the main control factors of recovery rate dominated by optimized neural network technology. The established analog reservoir parameter sequence library is applicable to any reservoir in the entire oil-bearing area where the target reservoir is located, making the analogy process more systematic and standardized.

[0019] 2. The new method for determining reservoir recovery based on analogy principle disclosed in this invention combines the problem of establishing the judgment matrix in the analytic hierarchy process with the results of neural network simulation, which cleverly eliminates the drawback of excessive subjectivity in the analytic hierarchy process and improves the rationality and credibility of the analogy results.

[0020] The novel method for determining reservoir recovery based on analogy principles disclosed in this invention guides relevant personnel to make accurate and reasonable calculations and analyses on the matching optimization problem between target reservoirs and analog reservoirs by combining theoretical models with standardized diagrams. The method is simple and can be used immediately even by inexperienced personnel, minimizing the impact of research skills and experience. It has high applicability and promotion value. Attached Figure Description

[0021] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. In the drawings:

[0022] Figure 1 This is a technical roadmap for determining reservoir recovery rate based on the analogy principle in the implementation method of this invention;

[0023] Figure 2This is a technical roadmap for analyzing the main controlling factors of oil recovery rate based on the genetic algorithm-optimized neural network method in the implementation method of this invention;

[0024] Figure 3 This is a histogram showing the sensitivity of each factor to the recovery rate in an example of the present invention.

[0025] Figure 4 This is a hierarchical structure diagram of the target layer, criterion layer, and scheme layer in an example of the present invention;

[0026] Figure 5 This is a calibration chart of the importance scale determined based on sensitivity differences in an example of the present invention. Detailed Implementation

[0027] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the invention and to fully convey the scope of the invention to those skilled in the art.

[0028] According to some embodiments of this application, a method for determining reservoir recovery rate based on the analogy principle is provided, comprising the following steps:

[0029] Step 1

[0030] For a identified target reservoir, firstly, count the number and type of reservoirs in the oilfield where it is located, collect relevant data on these reservoirs, including geological structure data, reservoir development atlases and production dynamic data, and screen out all reservoir information contained in each major reservoir type.

[0031] Step 2

[0032] The main parameters contained in the collected data are assigned values, the parameters are made dimensionless, the recovery rate is used as the optimization target, and the main control factors are analyzed using a neural network algorithm optimized by a genetic algorithm.

[0033] Step 3

[0034] Set a threshold for the number of sensitive parameters. The principle for setting the threshold is to ensure that the selected parameter categories involve geological data, engineering data, and development data. Based on the analysis results, select the parameters with the highest sensitivity ranking above the threshold as the parameter sequence for analog analysis of this type of reservoir.

[0035] Step 4

[0036] Similarly, repeat steps 2-3 for other types of reservoirs, and establish a basic parameter sequence library of analog reservoirs in the oil-bearing area where the target reservoir is located, based on the criteria required for SEC analogy, namely the same geological strata, the same sedimentary environment, similar geological structure, and the same driving mechanism.

[0037] Step 5

[0038] According to the established analog reservoir basic parameter sequence library, input all reservoir parameters of the same type as the target reservoir in the oilfield, and perform three screenings for similar reservoirs;

[0039] Step 6

[0040] The first qualitative screening of similar reservoirs was conducted based on the SEC analogy criteria. The second quantitative screening followed the principle of numerical proximity, specifically selecting reservoirs from the first screening results whose main influencing parameters were similar to the target reservoir as analogous reservoirs. The third quantitative screening involved a comprehensive analysis of the influence of all parameters, specifically using the analytic hierarchy process (AHP) to analyze the similarity of analogy indicators. The three screening methods are progressively more advanced.

[0041] Step 7

[0042] The recovery rate is determined based on the production dynamics of the target reservoir. If the target reservoir has been developed for a long time and has a declining trend, the decline rate is compared with that of the target reservoir, and the final recovery rate is calculated using the dynamic method. If the target reservoir has not been developed or has been developed but does not have a declining trend, the recovery rate of the analog reservoir is used as the benchmark. Based on the similarities and differences between the analog reservoir and the target reservoir in terms of geological understanding and development characteristics, the benchmark value of the recovery rate is appropriately adjusted as the final analog recovery rate.

[0043] This invention provides a novel method for determining reservoir recovery based on the analogy principle, comprising the following steps:

[0044] For a identified target reservoir, the first step is to statistically analyze the number and types of reservoirs within its oilfield. Reservoir types are classified based on lithology, permeability, and driving mechanism, such as high-permeability sandstone reservoirs developed through water injection, and low-permeability conglomerate reservoirs with natural energy. The main reservoir types within the oilfield are then identified and statistically analyzed.

[0045] After determining the reservoir type, information on all reservoirs included in each major reservoir type was screened, and relevant data on these reservoirs were collected, including geological structure data, reservoir development atlases, and production dynamic data. After data collection, a main controlling factor analysis was performed.

[0046] To conduct the main control factor analysis, the first step is to assign values ​​to the key parameters contained in the collected relevant data. The assignment is divided into three types: for parameters without specific numerical values, such as geological strata, sedimentary type, and lithology, these parameters are assigned values ​​between 0 and 1 according to their inherent categories, eliminating the need for dimensionless assignment; for parameters with values ​​within a certain range, such as porosity and permeability, these parameters are assigned a scalar average; for parameters with definite numerical values, such as temperature and pressure, their actual values ​​are used. After dimensionless assignment of these parameters, with recovery rate as the optimization objective, a neural network algorithm optimized by a genetic algorithm is used for main control factor analysis. The genetic algorithm is used to optimize the initial threshold of the neural network, significantly improving the network training time and accuracy. The technical route is detailed in the appendix. Figure 2 .

[0047] Set a threshold for the number of sensitive parameters. The principle for setting the threshold is to ensure that the selected parameter categories involve geological data, engineering data, and development data. Based on the analysis results, select the parameters with the highest sensitivity ranking above the threshold as the basic parameters for analog analysis of this type of reservoir.

[0048] Similarly, repeat steps 2-4 for other types of reservoirs, and establish a basic parameter sequence library of the target reservoir in the oil-bearing area based on the criteria required for SEC analogy, namely the same geological strata, the same sedimentary environment, similar geological structure, and the same driving mechanism. In addition to the above 5 parameters, the sequence library should also include development time, reserve calculation parameters, reservoir and fluid properties, and development parameters.

[0049] Filter all reservoirs in the oilfield that are of the same type as the target reservoir, and input each parameter in sequence based on the analog reservoir basic parameter library established in steps 1-5.

[0050] The first screening is conducted based on the SEC analogy criteria, which selects reservoirs with the same geological strata, sedimentary environment, geological structure, and driving mechanism as the target reservoirs as analog reservoirs. The second screening method is to select reservoirs with similar main controlling factors to the target reservoirs from the first reservoir screening results as analog reservoirs, based on the principle of numerical similarity.

[0051] The specific method for the third screening is to use the analytic hierarchy process to perform analogy index similarity analysis, screen out the analog reservoirs that are most similar to the target reservoir, and if the final result includes more than two analog reservoirs, then multiple reservoirs are used for subsequent analogies, and finally the uncertainty range of the recovery rate assessment of the target oil and gas reservoir is given.

[0052] The recovery rate is determined based on the production dynamics of the target reservoir. If the target reservoir has been developed for a long time and has a declining trend, the decline rate is compared with that of the target reservoir, and the final recovery rate is calculated using the dynamic method. If the target reservoir has not been developed or has been developed but does not have a declining trend, the recovery rate of the analog reservoir is used as the benchmark. Based on the similarities and differences between the analog reservoir and the target reservoir in terms of geological understanding and development characteristics, the benchmark value of the recovery rate is appropriately adjusted as the final analog recovery rate.

[0053] The methods for establishing the comparison matrices at the criterion level and the comparison matrices at the alternative level in the Analytic Hierarchy Process (AHP), and the theoretical basis for rationality analysis, are as follows:

[0054] Assume that there are 5 controlling factors established by the rules described in steps 4 and 5, and their importance is ranked as follows: parameter 1, parameter 2, parameter 3, parameter 4, parameter 5;

[0055] The comparable reservoirs selected based on steps 6 and 7 are: comparable reservoir 1, comparable reservoir 2, and comparable reservoir 3.

[0056] Interpret the importance scale diagram as follows: Figure 5 As shown, establish a comparison matrix.

[0057] The comparison matrix of the criteria layer is shown in Table 1.

[0058] Table 1 Example of Criterion Layer Comparison Matrix

[0059] Parameter 1 Parameter 2 Parameter 3 Parameter 4 Parameter 5 Parameter 1 1 2 3 4 5 Parameter 2 1 / 2 1 2 3 4 Parameter 3 1 / 3 1 / 2 1 2 3 Parameter 4 1 / 4 1 / 3 1 / 2 1 2 Parameter 5 1 / 5 1 / 4 1 / 3 1 / 2 1

[0060] In the comparison matrix of the above criteria layer, the second row and third column represent that the importance of parameter 1 is one level higher than that of parameter 2, and the third row and fifth column represent that the importance of parameter 2 is two levels higher than that of parameter 4. The principle behind establishing this matrix is: simulation is performed using a neural network method, with the importance of parameters 1-5 increasing sequentially, and each parameter corresponding to a sensitivity. The importance scale of each type of parameter is determined based on the actual difference in sensitivity. Since the sensitivity obtained by the neural network simulation method has physical meaning, representing the rate of change in recovery rate with parameter values, this method of determining the importance scale is feasible.

[0061] If the difference in sensitivity is small, the scale difference is 1; if the difference in sensitivity is large, the scale difference is 2 or higher. The specific determination needs to be made based on the actual sensitivity values.

[0062] In this example, the sensitivity differences between the parameters are assumed to be close to an arithmetic sequence, so the importance scale of each parameter differs by 1.

[0063] The details of the established scheme-level comparison matrix are shown in Table 2.

[0064] Table 2 Example of Scheme Layer Comparison Matrix

[0065] Parameter 1 Analogous reservoir 1 Analog reservoir 2 Analogous reservoir 3 Analogous reservoir 1 1 2 3 Analog reservoir 2 1 / 2 1 2 Analogous reservoir 3 1 / 3 1 / 2 1

[0066] The second row and third column of the above scheme comparison matrix represent that, in terms of parameter 1, analog reservoir 1 scores one level higher than analog reservoir 2.

[0067] The principle behind establishing this matrix is: for parameters with definite values, the mathematical model for their calculation is as follows:

[0068]

[0069] In the formula, S represents the similarity between the target reservoir parameter 1 and the analog reservoir parameter 1; A i B represents the i-th reservoir parameter value of the target reservoir; i This represents the i-th reservoir parameter value in the analog reservoir.

[0070] The importance scale of each parameter is determined based on the actual difference in similarity.

[0071] If the similarity difference is small, the scale difference is 1; if the similarity difference is large, the scale difference is 2 or higher. The specific situation needs to be determined based on the actual similarity values.

[0072] The comparison matrix for parameters 2-5 is constructed in the same way as that for parameter 1.

[0073] The comparison matrix above also needs to undergo a consistency check. The basic principle is as follows: First, calculate the index CI, which measures the deviation of the comparison matrix from consistency. The calculation formula is as follows:

[0074]

[0075] The formula for calculating the test coefficient CR is:

[0076]

[0077] In the formula, CI represents the consistency index; λ Max represents the largest eigenvalue of the comparison matrix; n represents the sum of the diagonal elements of the comparison matrix; CR represents the consistency test coefficient; RI represents the random consistency index.

[0078] Generally, when CR≦0.1, the comparison matrix can be considered to have satisfactory consistency. Otherwise, the judgment matrix needs to be adjusted until satisfactory consistency is achieved.

[0079] Example:

[0080] This example illustrates a lithologic structural oil and gas reservoir in low-permeability sandstone within a continental oil-bearing basin. Reservoir classification and data collection have already been explained in the specific implementation plan, and will not be covered in this example. The example will focus on the step of establishing an analogous reservoir parameter sequence for the lithologic structural oil and gas reservoir in low-permeability sandstone.

[0081] After data collection and parameter assignment, the first step is to analyze the controlling factors. The analysis method is a neural network algorithm optimized from a genetic algorithm. This involves determining the key parameters of the neural network. These parameters are usually determined through trial and error. For this example, the model parameters of the established neural network and genetic algorithm are as follows: 10 individuals in the genetic algorithm, 30 generations, crossover probability 0.2, mutation probability 0.1, 10 input nodes, 15 hidden nodes, 1 output node, and target error 10⁻⁵. The final sensitivity ranking is shown in the appendix. Figure 3 A threshold for the number of sensitive parameters is set, with the principle of ensuring that the selected parameter categories involve geological data, engineering data, and development data. Based on the analysis results, this example selects 10 parameters as the lithological and structural oil and gas reservoir analogy sequence parameters for low-permeability sandstone.

[0082] The first screening, based on the SEC analogy criteria, selected reservoirs with similar geological strata, sedimentary environment, geological structure, and driving mechanism to the target reservoir as analog reservoirs, resulting in six similar reservoirs. The second screening principle was quantitative coarse screening, selecting reservoirs from the first screening results whose main influencing parameters were similar to those of the target reservoir as analog reservoirs. In this example, the main parameters affecting the recovery rate were permeability, effective thickness, development time, porosity, and viscosity. Therefore, these parameters of the six similar reservoirs were compared with those of the target reservoir, ultimately selecting three similar reservoirs, as detailed in Table 3.

[0083] Table 3. Analog Reservoir Characteristic Parameters After Three Screenings

[0084]

[0085]

[0086] The third screening method involves using the Analytic Hierarchy Process (AHP) to perform analogy index similarity analysis, selecting the analogous reservoirs most closely similar to the target reservoir. Hierarchical structure is the foundation of AHP; the so-called hierarchical structure means that AHP decomposes the decision into three levels: the target level, the criteria level, and the alternative level. The hierarchical structure diagram for this example is attached. Figure 4 .

[0087] The next step is to establish the comparison matrix at the criterion level and the comparison matrix at the alternative level. The comparison matrix at the criterion level is composed of pairwise comparisons of 10 parameters, and the importance scale is determined by the sensitivity difference and the actual meaning of the parameters. See the chart for the importance scale calibration. Figure 5 The criterion layer comparison matrix established based on this result is shown in Table 4.

[0088] Table 4. Criterion Layer Comparison Matrix

[0089]

[0090]

[0091] After determining the comparison matrix at the criterion level, it is necessary to determine the comparison matrix at the scheme level. There are 10 comparison matrices at the scheme level. Taking permeability as an example, the process of establishing the comparison matrix at the scheme level is introduced. The similarity of the permeability parameter between the analogous reservoir and the target reservoir is calculated according to the similarity mathematical model. The calculation results are 0.99, 0.93, and 0.89, respectively. The three similarity values ​​are close to an arithmetic sequence, indicating that for the permeability parameter, the differences between different analogous reservoirs are similar in magnitude. Therefore, the comparison matrix is ​​established as shown in Table 5.

[0092] The process of establishing the comparison matrix for other parameter schemes is the same as that for permeability.

[0093] Table 5 Scheme Layer Comparison Matrix

[0094] Penetration Analogous reservoir 1 Analog reservoir 2 Analogous reservoir 3 Analogous reservoir 1 1 2 3 Analog reservoir 2 1 / 2 1 2 Analogous reservoir 3 1 / 3 1 / 2 1

[0095] The CR value for the consistency test of the criterion-level comparison matrix is ​​0.026. The CR values ​​for the consistency test of the scheme-level comparison matrix are 0.0048, 0, 0.0048, 0.0133, 0.0253, 0.0403, 0.0512, 0, 0, and 0, respectively. All of these values ​​are less than 0.1, indicating that the established comparison matrix is ​​reasonable.

[0096] The weights of the criterion layer and the scheme layer were calculated, as detailed in Tables 6 and 7.

[0097] Table 6. Details of Criterion Layer Weights

[0098]

[0099] Table 7. Details of Scheme Layer Weights

[0100] Analogous reservoir 1 Analog reservoir 2 Analogous reservoir 3 Penetration 0.5400 0.2971 0.1629 Effective thickness 0.2500 0.5000 0.2500 Injection ratio 0.2971 0.5400 0.1629 Porosity 0.1688 0.3876 0.4436 Viscosity 0.6303 0.1080 0.2617 Original oil pressure 0.1566 0.2489 0.5944 Well network density 0.0859 0.6439 0.2701 driving method 0.3333 0.3333 0.3333 Well type 0.4545 0.0909 0.4545 coefficient of variation 0.3333 0.3333 0.3333

[0101] Multiplying the weights of the criterion layer and the scheme layer, the final calculated total score for analog reservoir 1 is 0.3748, for analog reservoir 2 it is 0.3529, and for analog reservoir 3 it is 0.2723. Therefore, analog reservoir 1 is the reservoir most similar to the target reservoir. Since the target reservoir has not yet been developed, the recovery rate of analog reservoir 1 (36%) is selected as the baseline for the target reservoir's recovery rate. Considering that the target reservoir has better porosity, flow coefficient, homogeneity, and microfracture development than the analog reservoir, 40% is selected as the final analog recovery rate for the target reservoir.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining reservoir recovery rate based on the principle of analogy, characterized in that, include: The number and types of oil reservoirs in the target oil reservoir area are statistically analyzed, oil reservoir information is selected from the oil reservoir types, and data related to oil reservoir information are collected, including geological structure data, oil reservoir development atlases and production dynamic data. The parameters in the reservoir information are assigned values ​​and dimensionless. The recovery rate is used as the optimization target. The main control factor analysis is performed using the neural network algorithm optimized by the genetic algorithm. The threshold for the number of sensitive parameters is set. The parameters with the highest sensitivity ranking are included in the parameter sequence for analog analysis. A basic parameter sequence library of analog reservoirs in the oil-bearing area where the target reservoir is located is established. According to the analog reservoir basic parameter sequence library, input all reservoir parameters of the same type as the target reservoir in the oil area, and perform three screenings of similar reservoirs: the first screening of similar reservoirs, based on the SEC analogy basic criteria, selects reservoirs with the same geological strata, sedimentary environment, geological structure and driving mechanism as the target reservoir as analog reservoirs. The second screening of similar reservoirs was conducted based on the principle of numerical proximity, selecting reservoirs with similar numerical values ​​from the first screening of similar reservoirs as analog reservoirs. The third screening of similar reservoirs used the analytic hierarchy process (AHP) for analogy index similarity analysis. The recovery rate is determined based on the production dynamics of the target reservoir: if the target reservoir has been developed for a long time and has a decreasing trend, the decline rate of an analog reservoir is used as an analogy, and the final recovery rate is calculated using the dynamic method; if the target reservoir has not been developed or has been developed but does not have a decreasing trend, the recovery rate of the analog reservoir is used as the benchmark, and the benchmark value of the recovery rate is adjusted according to the similarities and differences between the analog reservoir and the target oil and gas reservoir in terms of geological understanding and development characteristics, and is used as the final analog recovery rate. The analytic hierarchy process includes constructing a comparison matrix based on an importance scale interpretation. The comparison matrix is ​​simulated using a neural network method, with the importance of the parameters increasing in order. Each parameter corresponds to a sensitivity, and the importance scale of each type of parameter is determined according to the actual difference in sensitivity. The third screening uses the Analytic Hierarchy Process (AHP) to perform analogy index similarity analysis, and selects analogous reservoirs that are most similar to the target reservoir. The hierarchical structure is the basis of the AHP, which decomposes the decision into three levels: target level, criterion level, and alternative level.

2. The method for determining reservoir recovery based on the analogy principle according to claim 1, characterized in that, Oil reservoir types are classified based on lithology, permeability, and driving mechanism.

3. The method for determining reservoir recovery based on the analogy principle according to claim 1, characterized in that, There are three types of assignment: assigning a value between 0 and 1 to parameters without a specific value; assigning the average value of a parameter with a value within a certain range; and assigning the true value to a parameter with a definite value.

4. The method for determining reservoir recovery based on the analogy principle according to claim 1, characterized in that, The principle for setting the number threshold is to ensure that the selected parameter categories involve geological data, engineering data, and development data.

5. The method for determining reservoir recovery based on the analogy principle according to claim 1, characterized in that, The parameters include any one of the following groups: permeability, effective thickness, injection-production ratio, porosity, viscosity, original reservoir pressure, well density, drive mode, well type, coefficient of variation, extraction mode, saturation pressure, burial depth, density, temperature, gas-oil ratio, development time, crude oil volume factor, oil-bearing area, lithology, reserve calculation parameters, reservoir and fluid properties, and development parameters.

6. The method for determining reservoir recovery based on the analogy principle according to claim 1, characterized in that, In the comparison matrix, for parameters with definite values, the mathematical model for calculation is as follows: （1） In the formula, Indicates the similarity between the target reservoir parameters and the parameters of the analogous reservoir; This represents the i-th reservoir parameter value of the target reservoir; This represents the i-th reservoir parameter value in the analog reservoir.

7. The method for determining reservoir recovery based on the analogy principle according to claim 6, characterized in that, A consistency check is performed on the comparison matrix, and the index CI, which measures the deviation of the comparison matrix from consistency, is calculated using the following formula: （2） CI stands for Consistency Index; This represents the largest eigenvalue of the comparison matrix; This represents the sum of the diagonal elements of the comparison matrix.

8. The method for determining reservoir recovery based on the analogy principle according to claim 7, characterized in that, The formula for calculating the test coefficient CR is: （3） In the formula, CR represents the consistency test coefficient; CI stands for Consistency Index; RI stands for Random Consistency Index; When CR≦0.1, it indicates that the comparison matrix has satisfactory consistency; Otherwise, adjust the judgment matrix until satisfactory consistency is achieved.

Images