Parameter weight determination method and device, equipment, storage medium and program product
By combining the analytic hierarchy process (AHP), XGBoost regression model, and game theory, the subjective and objective weights of shale gas production factors are calculated, which solves the problem of low accuracy in determining parameter weights and improves the accuracy and flexibility of shale gas well production prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- PETROCHINA CO LTD
- Filing Date
- 2024-11-01
- Publication Date
- 2026-05-08
AI Technical Summary
Existing methods for determining parameter weights rely on expert subjective experience and lack scientific basis, resulting in low accuracy.
By combining the analytic hierarchy process (AHP), the XGBoost regression model, and game theory, the subjective and objective weights of the factors influencing shale gas production capacity are calculated, consistency is tested, and finally, the comprehensive weights are calculated.
It improves the accuracy of shale gas well production capacity prediction, enhances the flexibility and adaptability of the method, simplifies the weight acquisition process, reduces reliance on professional knowledge, and improves user-friendliness.
Smart Images
Figure CN121998141A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of shale gas development technology, and in particular to a method, apparatus, equipment, storage medium, and program product for determining parameter weights. Background Technology
[0002] Currently, in multi-criteria decision analysis (MCDA), parameter weights reflect the relative importance of each decision criterion in the overall evaluation. Therefore, determining the parameter weights is crucial for the entire decision-making process.
[0003] Traditional methods for determining parameter weights mainly rely on the subjective experience of experts. While this approach is simple and easy to implement, it often lacks sufficient scientific basis and is easily influenced by personal preferences and irrational factors, resulting in low accuracy in determining parameter weights. Summary of the Invention
[0004] This invention provides a method, apparatus, device, storage medium, and program product for determining parameter weights, which can solve the technical problem of low accuracy in determining parameter weights in the prior art.
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] In a first aspect, embodiments of the present invention provide a method for determining parameter weights, the method comprising:
[0007] The subjective weights of each factor affecting shale gas production capacity were calculated using the analytic hierarchy process (AHP).
[0008] The objective weights of each factor affecting shale gas production capacity were calculated using the XGBoost regression model.
[0009] Consistency checks are performed on the subjective and objective weights of each factor.
[0010] When the consistency test result is passed, the comprehensive weight of each factor is calculated based on the subjective weight and objective weight of each factor.
[0011] Secondly, embodiments of the present invention provide a parameter weight determination device, the device comprising:
[0012] The first calculation module is configured to calculate the subjective weights of each factor affecting shale gas production capacity using the analytic hierarchy process.
[0013] The second calculation module is configured to calculate the objective weights of each factor affecting shale gas production capacity using the XGBoost regression model.
[0014] The detection module is configured to perform consistency detection on the subjective and objective weights of each of the factors;
[0015] The third calculation module is configured to calculate the comprehensive weight of each factor based on the subjective weight and objective weight of each factor when the consistency detection result is passed.
[0016] Thirdly, embodiments of the present invention also provide an electronic device, including: a memory and a processor; the processor is configured to read and execute a computer program stored in the memory to implement the steps of the aforementioned parameter weight determination method.
[0017] Fourthly, embodiments of the present invention also provide a computer storage medium storing computer-executable instructions, which, when executed, implement the steps of the aforementioned parameter weight determination method.
[0018] Fifthly, embodiments of the present invention also provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the aforementioned parameter weight determination method.
[0019] The beneficial effects of the technical solutions provided by the embodiments of the present invention include:
[0020] This invention calculates the subjective weights of each factor influencing shale gas production capacity using the Analytic Hierarchy Process (AHP); it then calculates the objective weights of each factor using an XGBoost regression model; a consistency check is performed on the subjective and objective weights of each factor; and when the consistency check passes, a comprehensive weight is calculated for each factor based on its subjective and objective weights. This invention combines the AHP, XGBoost machine learning algorithm, and game theory to more scientifically reflect the importance of each parameter in the comprehensive evaluation system, thereby improving the accuracy of shale gas well production capacity prediction. Furthermore, the weight acquisition method is not only applicable to shale gas well production capacity prediction but also possesses high flexibility and scalability, improving the model's adaptability and flexibility, simplifying the weight acquisition process, making it more efficient, easier to understand and apply, reducing reliance on specialized knowledge, and improving the method's accessibility and user-friendliness. This solves the technical problem of low accuracy in parameter weight determination in related technologies. Attached Figure Description
[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0022] Figure 1 This is a flowchart illustrating the first embodiment of the parameter weight determination method of the present invention;
[0023] Figure 2 This is a schematic diagram illustrating the objective weights of various factors affecting shale gas production capacity according to the present invention.
[0024] Figure 3 This is a schematic diagram illustrating the subjective weights of various factors affecting shale gas production capacity in this invention.
[0025] Figure 4 This is a schematic diagram showing the comprehensive weights of various factors affecting shale gas production capacity according to the present invention.
[0026] Figure 5 This is a functional module diagram of an embodiment of the parameter weight determination device of the present invention;
[0027] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation
[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0029] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0030] In a first aspect, embodiments of the present invention provide a method for determining parameter weights.
[0031] In one embodiment, reference is made to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the parameter weight determination method of the present invention. Figure 1 As shown, the methods for determining parameter weights include:
[0032] Step S10: The subjective weights of each factor affecting shale gas production capacity are calculated using the analytic hierarchy process (AHP).
[0033] In some specific embodiments, step S10 includes:
[0034] The importance ratios among various factors affecting shale gas production capacity were determined using the 1-9 scaling method, and a judgment matrix was constructed.
[0035] The subjective weights of each factor affecting shale gas production capacity are calculated based on the scaling values of each factor in the judgment matrix.
[0036] In this embodiment, the 1-9 scale is a method for quantifying subjective judgments, primarily used in the Analytic Hierarchy Process (AHP) to determine the relative importance of factors by comparing them pairwise. The scale values range from 1 to 9, representing different degrees of importance between two factors. Specifically, 1 indicates that both factors are equally important, 3 indicates that the former is slightly more important than the latter, 5 indicates that the former is significantly more important than the latter, 7 indicates that the former is strongly more important than the latter, and 9 indicates that the former is extremely more important than the latter. 2, 4, 6, and 8 represent the median values for these adjacent judgments. Values at symmetrical positions are in reciprocal form. The scale values and their meanings are shown in Table 1.
[0037] Table 1
[0038]
[0039] A judgment matrix is constructed by comparing the importance of each factor pairwise. For example, if factor A is slightly more important than factor B, then 3 is entered in the row corresponding to A and column corresponding to B in the judgment matrix; if factor A is extremely more important than factor B, then 9 is entered. For instance, taking factors 1, 2, and 3 as examples, the importance of factors 1, 2, and 3 is compared pairwise. If factor 1 is slightly more important than factor 2, then 3 (3 times more important) is entered in the row corresponding to factor 1 and column corresponding to factor 2 in the judgment matrix; if factor 1 is half as important as factor 3, then 3 is entered in the row corresponding to factor 1 and column corresponding to factor 3 in the judgment matrix. ( (This is extremely important).
[0040] Furthermore, if factor 2 is less important than factor 1 ( It is twice as important as factor 3, and one-quarter as important as factor 4. The importance of factor 3 is twice that of factor 1, and the importance of factor 3 is four times that of factor 2. The constructed judgment matrix is shown in Table 2.
[0041] Table 2
[0042]
[0043] First, calculate the sum of the scale values in each column of the judgment matrix. The sum of the first column is 3.33, the sum of the second column is 8, and the sum of the third column is 1.75. Then, normalize the scale values in each row of the judgment matrix. (First row:)
[0044]
[0045] Second line:
[0046] Third line:
[0047] Then, the sum of the normalized scale values of each row is divided by the total number of factors in each row, and the quotient is used as the weight of each factor in the judgment matrix.
[0048] Factor 1 weight:
[0049] Factor 2 weight:
[0050] Factor 3 weight:
[0051] The weights of each factor are normalized to obtain the subjective weight of each factor.
[0052] Factor 1 Subjective Weighting:
[0053] Factor 2: Subjective weighting
[0054] Factor 3: Subjective weighting
[0055] That is, the subjective weight of factor 1 is 0.51, the subjective weight of factor 2 is 0.195, and the subjective weight of factor 3 is 0.295. In this embodiment, the factors affecting shale gas production capacity include: brittle mineral content, sand addition intensity, TOC (total organic carbon content), porosity, fluid intensity, vertical depth, thickness of sub-layer I, drilling length of type I reservoir, gas saturation, fracturing section length, average inter-section spacing, and average displacement. Similarly, by comparing the importance of each factor pairwise, a judgment matrix is constructed, and then the subjective weight of each factor affecting shale gas production capacity can be calculated based on the scaling values of each factor in the judgment matrix.
[0056] Step S20: Calculate the objective weights of each factor affecting shale gas production capacity using the XGBoost regression model.
[0057] In some specific embodiments, step S20 includes:
[0058] The XGBoost regression model is trained based on the factors that affect shale gas production capacity and the final recoverable reserves of the actual single wells corresponding to each factor. When the loss function of the XGBoost regression model converges, the trained XGBoost regression model is obtained.
[0059] Each factor affecting shale gas production capacity is sequentially input into the trained XGBoost regression model to obtain the final recoverable reserves of a single well corresponding to each factor output by the trained XGBoost regression model.
[0060] The final recoverable reserves of the predicted single-well assessment corresponding to each of the factors are normalized, and the normalized final recoverable reserves of the predicted single-well assessment corresponding to each of the factors are used as the objective weights of each factor.
[0061] In this embodiment, the XGBoost regression model is trained based on various factors affecting shale gas production capacity (brittle mineral content, sand addition intensity, TOC (total organic carbon content), porosity, fluid intensity, vertical depth, thickness of sublayer I, drilling length of type I reservoir, gas saturation, fracturing section length, average interval between sections, and average displacement), and the actual EUR (Estimated Ultimate Recovery, the final recoverable reserves of a single well) corresponding to each factor. When the loss function of the XGBoost regression model converges, the trained XGBoost regression model is obtained.
[0062] XGBoost (Xtreme Gradient Boosting) is an improvement on the Gradient Boosting algorithm, using a regression tree as its internal decision tree. This algorithm utilizes Newton's method to find the extreme value of the loss function. The objective function of XGBoost includes a loss function and a tree-based regularization term. Since direct optimization of the original objective function is difficult, the loss function is Taylor-expanded to a second order, and a regularization term is added. The loss function of the XGBoost regression model is as follows:
[0063]
[0064] In the formula, Let y represent the loss function of the XGBoost regression model. i This represents the final recoverable reserves of the actual single well corresponding to the i-th factor. This represents the final recoverable reserves of the predicted single well corresponding to the i-th factor. f represents the regularization term. kLet L(φ) represent the k-th tree in the decision tree model, L(φ) represent the objective function, and n represent the total number of factors affecting shale gas production capacity. T represents the number of leaf nodes in the decision tree, γ and λ are constants, and ω represents the vector formed by all the leaf node values in the decision tree.
[0065] Each factor affecting shale gas production capacity is sequentially input into the trained XGBoost regression model to obtain the predicted EUR corresponding to each factor output by the trained XGBoost regression model.
[0066] The predicted EURs corresponding to each factor affecting shale gas production capacity are normalized so that their sum is 1. The normalized predicted EURs corresponding to each factor affecting shale gas production capacity are used as the objective weights of each factor affecting shale gas production capacity.
[0067] Step S30: Perform a consistency check on the subjective weight and objective weight of each factor;
[0068] In some specific embodiments, step S30 includes:
[0069] Substitute the subjective and objective weights of each factor into the consistency detection formula to calculate the consistency index between the subjective and objective weights of all factors.
[0070] The consistency index is compared with the threshold.
[0071] If the consistency index is greater than or equal to the threshold, the consistency test result is determined to be unsuccessful.
[0072] If the consistency index is less than the threshold, the consistency test result is determined to be passed;
[0073] The consistency detection formula is as follows:
[0074]
[0075] In the formula, S represents the consistency index between the subjective and objective weights of all the factors, and a i b represents the subjective weight of the i-th factor. i represents the objective weight of the i-th factor, and n represents the total number of factors affecting shale gas production capacity.
[0076] In this embodiment, after obtaining the subjective and objective weights of each factor affecting shale gas production capacity, a consistency check needs to be performed on these weights. The subjective and objective weights of each factor affecting shale gas production capacity are substituted into the consistency check formula. This allows for the calculation of the consistency index between the subjective and objective weights of all factors affecting shale gas production capacity. In the formula, S represents the consistency index between the subjective and objective weights of all factors affecting shale gas production capacity, and a... i b represents the subjective weight of the i-th factor affecting shale gas production capacity. i denoted by , where represents the objective weight of the i-th factor affecting shale gas production capacity, and n represents the total number of all factors affecting shale gas production capacity.
[0077] Taking a threshold of 0.4 as an example, if the calculated consistency index S is greater than or equal to the threshold of 0.4, the consistency test result is determined to be unsuccessful.
[0078] If the calculated consistency index S is less than the threshold of 0.4, it is considered that the subjective and objective weights meet the consistency test, and the consistency test result is determined to be passed.
[0079] Step S40: When the consistency detection result is passed, the comprehensive weight of each factor is calculated based on the subjective weight and objective weight of each factor.
[0080] In some specific embodiments, step S40 includes:
[0081] The subjective and objective weights of each factor are arbitrarily linearly combined using a game-theoretic combination weighting method to obtain the comprehensive weight of each factor.
[0082] In this embodiment, the game-theoretic combination weighting method aims to minimize the deviation between the combined weight values and the subjective and objective weight values, thereby constructing the objective function. Simply put, it seeks an optimal position in the "game" between subjective and objective weights. Specifically, let the subjective weight 'a' of each factor affecting shale gas production capacity be... i The vector formed is Let b be the objective weight of each factor affecting shale gas production capacity. i The vector formed is Any linear combination of these two weight vectors is: In the formula, W is the overall weight, and α q These are the coefficients of the linear combination.
[0083] According to game theory, the optimal strategy for optimizing the coefficients of the linear combinatorial equation is:
[0084]
[0085] Based on the differential properties of matrices, the optimal first derivative condition for the above equation is obtained as follows:
[0086]
[0087] Therefore, by solving the system of linear combination equations, the linear combination coefficients α1 and α2 can be obtained.
[0088] The coefficients α1 and α2 of the linear combination are normalized. In the formula, This represents the normalized linear combination coefficients. Substituting the normalized linear combination coefficients into the formula... This allows us to obtain the comprehensive weight of each factor affecting shale gas production capacity.
[0089] Specifically, refer to Figure 2 , Figure 3 as well as Figure 4 . Figure 2 This is a schematic diagram illustrating the objective weights of various factors affecting shale gas production capacity according to the present invention. Figure 3 This is a schematic diagram illustrating the subjective weights of various factors affecting shale gas production capacity in this invention. Figure 4 This is a schematic diagram illustrating the comprehensive weights of various factors affecting shale gas production capacity in this invention. Taking vertical depth as an example, the objective weight of vertical depth is 0.079, and the subjective weight of vertical depth is 0.08. After solving for the linear combination coefficients α1 and α2, the objective and subjective weights of vertical depth are substituted into the formula. The calculated comprehensive weight of vertical depth is 0.08. The calculation process for the comprehensive weights of other factors affecting shale gas production capacity is the same as that for vertical depth, and will not be repeated here.
[0090] In this embodiment, the subjective weights of each factor affecting shale gas production capacity are calculated using the Analytic Hierarchy Process (AHP); the objective weights of each factor are calculated using the XGBoost regression model; a consistency check is performed on the subjective and objective weights of each factor; when the consistency check result is passed, the comprehensive weight of each factor is calculated based on the subjective and objective weights. This embodiment combines the AHP, XGBoost machine learning algorithm, and game theory to more scientifically reflect the importance of each parameter in the comprehensive evaluation system, thereby improving the accuracy of shale gas well production capacity prediction. Furthermore, the weight acquisition method is not only applicable to shale gas well production capacity prediction but also possesses high flexibility and scalability, improving the model's adaptability and flexibility, simplifying the weight acquisition process, making it more efficient, easier to understand and apply, reducing reliance on professional knowledge, improving the method's accessibility and user-friendliness, and solving the technical problem of low accuracy in parameter weight determination in related technologies.
[0091] Optionally, in one embodiment, after step S10, the following is included:
[0092] Based on the classification of various factors affecting shale gas production capacity, the system is divided into three levels: target level, criterion level, and scheme level. The target level is the final proposed result of a problem; the criterion level includes various factors affecting shale gas production capacity; and the scheme level refers to reservoir stimulation methods.
[0093] Based on the subjective weights of each factor and the weights of each scheme in the scheme layer with respect to each factor, the formula is used to... The subjective weights of each scheme in the scheme layer are calculated;
[0094] In the formula, w j w represents the subjective weight of the j-th option in the option layer. ij C represents the weight of the j-th scheme in the scheme layer with respect to the i-th factor. i represents the subjective weight of the i-th factor, and n represents the total number of factors affecting shale gas production capacity.
[0095] In this embodiment, the target layer and criterion layer are first determined. The target layer is the final proposed result of a problem; in this embodiment, the target layer is the production capacity assessment of shale gas wells. Based on the geological conditions and development characteristics of shale gas wells, factors affecting shale gas production capacity are determined as criterion layers. These factors include: brittle mineral content, sand addition intensity, TOC (total organic carbon content), porosity, fluid strength, vertical depth, thickness of sublayer I, length of Class I reservoir encountered, gas saturation, length of fractured section, average interval between sections, and average discharge rate.
[0096] The ultimate goal of the entire decision-making process is to predict the production capacity of shale gas wells, and to list possible solutions: Based on the factors in the criterion layer, all possible solutions or measures should be listed, and these solutions should be able to impact the target layer. In the context of shale gas well production capacity prediction, solutions include different reservoir stimulation methods such as fracturing technologies. A solution layer is constructed, with all feasible solutions as elements of the solution layer. These solutions will be used for subsequent weight calculations and decision analysis. Pairwise comparisons are performed: In the solution layer, each solution is compared pairwise with each criterion to determine their importance or priority.
[0097] Based on the subjective weights of each factor affecting shale gas production capacity and the weights of each factor affecting shale gas production capacity in each scheme at the scheme level, the formula is used to... The subjective weights of each option in the alternative layer are calculated, and the importance or priority of each option is determined based on these subjective weights. In the formula, w j w represents the subjective weight of the j-th option in the option layer. ij C represents the weight of the j-th scheme in the scheme layer with respect to the i-th factor affecting shale gas production capacity. irepresents the subjective weight of the i-th factor affecting shale gas production capacity, and n represents the total number of all factors affecting shale gas production capacity.
[0098] The weights of the factors in the criterion layer at the scheme layer are determined in the following two cases:
[0099] 1) If the importance comparison values of each scheme in the scheme layer at the criterion layer can be quantified, then this type of numerical value can be directly determined after normalization.
[0100] 2) If the importance of each option in the option layer cannot be directly quantified at the criterion layer, then pairwise comparisons should be made according to the scaling criteria.
[0101] If the impact of each option in the solution layer on the criteria layer can be measured using specific data or clear quantitative indicators, then the values of these importance comparisons can be quantified. For example, if the criterion is cost, and each option in the solution layer has clear cost data, then this data can be used for quantitative analysis.
[0102] If the importance comparison values cannot be directly measured by data but rely on the subjective judgment or experience of experts, then these importance comparison values cannot be directly quantified. In such cases, a pairwise comparison method is usually used, with expert scoring used to determine relative importance.
[0103] Secondly, embodiments of the present invention also provide a parameter weight determination device.
[0104] In one embodiment, reference is made to Figure 5 , Figure 5 This is a functional module diagram of an embodiment of the parameter weight determination device of the present invention. Figure 5 As shown, the parameter weight determination device includes:
[0105] The first calculation module 10 is configured to calculate the subjective weights of each factor affecting shale gas production capacity using the analytic hierarchy process.
[0106] The second calculation module 20 is configured to calculate the objective weights of each factor affecting shale gas production capacity using the XGBoost regression model.
[0107] The detection module 30 is configured to perform consistency detection on the subjective weight and objective weight of each of the factors;
[0108] The third calculation module 40 is configured to calculate the comprehensive weight of each factor based on the subjective weight and objective weight of each factor when the consistency detection result is passed.
[0109] Optionally, in one embodiment, the first computing module 10 is configured to:
[0110] The importance ratios among various factors affecting shale gas production capacity were determined using the 1-9 scaling method, and a judgment matrix was constructed.
[0111] The subjective weights of each factor affecting shale gas production capacity are calculated based on the scaling values of each factor in the judgment matrix.
[0112] Optionally, in one embodiment, the second computing module 20 is configured to:
[0113] The XGBoost regression model is trained based on the factors that affect shale gas production capacity and the final recoverable reserves of the actual single wells corresponding to each factor. When the loss function of the XGBoost regression model converges, the trained XGBoost regression model is obtained.
[0114] Each factor affecting shale gas production capacity is sequentially input into the trained XGBoost regression model to obtain the final recoverable reserves of a single well corresponding to each factor output by the trained XGBoost regression model.
[0115] The final recoverable reserves of the predicted single-well assessment corresponding to each of the factors are normalized, and the normalized final recoverable reserves of the predicted single-well assessment corresponding to each of the factors are used as the objective weights of each factor.
[0116] Optionally, in one embodiment, the loss function of the XGBoost regression model is as follows:
[0117]
[0118] In the formula, Let y represent the loss function of the XGBoost regression model. i This represents the final recoverable reserves of the actual single well corresponding to the i-th factor. This represents the final recoverable reserves of the predicted single well corresponding to the i-th factor. f represents the regularization term. k Let L(φ) represent the k-th tree in the decision tree model, L(φ) represent the objective function, and n represent the total number of factors affecting shale gas production capacity. T represents the number of leaf nodes in the decision tree, γ and λ are constants, and ω represents the vector formed by all the leaf node values in the decision tree.
[0119] Optionally, in one embodiment, the detection module 30 is configured to:
[0120] Substitute the subjective and objective weights of each factor into the consistency detection formula to calculate the consistency index between the subjective and objective weights of all factors.
[0121] The consistency index is compared with the threshold.
[0122] If the consistency index is greater than or equal to the threshold, the consistency test result is determined to be unsuccessful.
[0123] If the consistency index is less than the threshold, the consistency test result is determined to be passed;
[0124] The consistency detection formula is as follows:
[0125]
[0126] In the formula, S represents the consistency index between the subjective and objective weights of all the factors, and a i b represents the subjective weight of the i-th factor. i represents the objective weight of the i-th factor, and n represents the total number of factors affecting shale gas production capacity.
[0127] Optionally, in one embodiment, the third computing module 40 is configured to:
[0128] The subjective and objective weights of each factor are arbitrarily linearly combined using a game-theoretic combination weighting method to obtain the comprehensive weight of each factor.
[0129] Optionally, in one embodiment, the parameter weight determination device further includes a fourth calculation module configured to:
[0130] Based on the classification of various factors affecting shale gas production capacity, the system is divided into three levels: target level, criterion level, and scheme level. The target level is the final proposed result of a problem; the criterion level includes various factors affecting shale gas production capacity; and the scheme level refers to reservoir stimulation methods.
[0131] Based on the subjective weights of each factor and the weights of each scheme in the scheme layer with respect to each factor, the formula is used to... The subjective weights of each scheme in the scheme layer are calculated;
[0132] In the formula, w j w represents the subjective weight of the j-th option in the option layer. ij C represents the weight of the j-th scheme in the scheme layer with respect to the i-th factor. i represents the subjective weight of the i-th factor, and n represents the total number of factors affecting shale gas production capacity.
[0133] The functions of each module in the above-mentioned parameter weight determination device correspond to the steps in the above-mentioned parameter weight determination method embodiment, and their functions and implementation processes will not be described in detail here.
[0134] Thirdly, embodiments of the present invention also provide an electronic device, the structure of which is as follows: Figure 6 As shown, it includes: a memory and a processor, wherein the processor is used to read and execute the computer program stored in the memory to implement the aforementioned parameter weight determination method.
[0135] Fourthly, embodiments of the present invention also provide a computer storage medium storing computer-executable instructions, which, when executed, implement the aforementioned parameter weight determination method.
[0136] Fifthly, embodiments of the present invention provide a computer program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described parameter weight determination method embodiments, and can achieve the same technical effect. To avoid repetition, it will not be described again here.
[0137] Finally, it should be noted that while some processes described in the embodiments of the present invention include multiple operations or steps that appear in a specific order, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of the present invention, or may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.
[0138] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for determining parameter weights, characterized in that, The method includes: The subjective weights of each factor affecting shale gas production capacity were calculated using the analytic hierarchy process (AHP). The objective weights of each factor affecting shale gas production capacity were calculated using the XGBoost regression model. Consistency checks are performed on the subjective and objective weights of each factor. When the consistency test result is passed, the comprehensive weight of each factor is calculated based on the subjective weight and objective weight of each factor.
2. The parameter weight determination method according to claim 1, characterized in that, The step of calculating the subjective weights of each factor affecting shale gas production capacity using the analytic hierarchy process includes: The importance ratios among various factors affecting shale gas production capacity were determined using the 1-9 scaling method, and a judgment matrix was constructed. The subjective weights of each factor affecting shale gas production capacity are calculated based on the scaling values of each factor in the judgment matrix.
3. The parameter weight determination method according to claim 1, characterized in that, The step of calculating the objective weights of each factor affecting shale gas production capacity using the XGBoost regression model includes: The XGBoost regression model is trained based on the factors that affect shale gas production capacity and the final recoverable reserves of the actual single wells corresponding to each factor. When the loss function of the XGBoost regression model converges, the trained XGBoost regression model is obtained. Each factor affecting shale gas production capacity is sequentially input into the trained XGBoost regression model to obtain the final recoverable reserves of a single well corresponding to each factor output by the trained XGBoost regression model. The final recoverable reserves of the predicted single-well assessment corresponding to each of the factors are normalized, and the normalized final recoverable reserves of the predicted single-well assessment corresponding to each of the factors are used as the objective weights of each factor.
4. The parameter weight determination method according to claim 3, characterized in that, The loss function of the XGBoost regression model is as follows: In the formula, Let y represent the loss function of the XGBoost regression model. i This represents the final recoverable reserves of the actual single well corresponding to the i-th factor. This represents the final recoverable reserves of the predicted single well corresponding to the i-th factor. f represents the regularization term. k Let L(φ) represent the k-th tree in the decision tree model, L(φ) represent the objective function, and n represent the total number of factors affecting shale gas production capacity. T represents the number of leaf nodes in the decision tree, γ and λ are constants, and ω represents the vector formed by all the leaf node values in the decision tree.
5. The parameter weight determination method according to claim 1, characterized in that, The step of performing consistency detection on the subjective weights and objective weights of each factor includes: Substitute the subjective and objective weights of each factor into the consistency detection formula to calculate the consistency index between the subjective and objective weights of all factors. The consistency index is compared with the threshold. If the consistency index is greater than or equal to the threshold, the consistency test result is determined to be unsuccessful. If the consistency index is less than the threshold, the consistency test result is determined to be passed; The consistency detection formula is as follows: In the formula, S represents the consistency index between the subjective and objective weights of all the factors, and a i b represents the subjective weight of the i-th factor. i represents the objective weight of the i-th factor, and n represents the total number of factors affecting shale gas production capacity.
6. The parameter weight determination method according to claim 1, characterized in that, The step of calculating the comprehensive weight of each factor based on the subjective and objective weights of each factor includes: The subjective and objective weights of each factor are arbitrarily linearly combined using a game-theoretic combination weighting method to obtain the comprehensive weight of each factor.
7. The parameter weight determination method according to claim 1, characterized in that, After calculating the subjective weights of each factor affecting shale gas production capacity using the analytic hierarchy process (AHP), the following steps are included: Based on the classification of various factors affecting shale gas production capacity, the system is divided into three levels: target level, criterion level, and scheme level. The target level is the final proposed result of a problem; the criterion level includes various factors affecting shale gas production capacity; and the scheme level refers to reservoir stimulation methods. Based on the subjective weights of each factor and the weights of each scheme in the scheme layer with respect to each factor, the formula is used to... The subjective weights of each scheme in the scheme layer are calculated; In the formula, w j w represents the subjective weight of the j-th option in the option layer. ij C represents the weight of the j-th scheme in the scheme layer with respect to the i-th factor. i represents the subjective weight of the i-th factor, and n represents the total number of factors affecting shale gas production capacity.
8. A parameter weight determination device, characterized in that, The device includes: The first calculation module is configured to calculate the subjective weights of each factor affecting shale gas production capacity using the analytic hierarchy process. The second calculation module is configured to calculate the objective weights of each factor affecting shale gas production capacity using the XGBoost regression model. The detection module is configured to perform consistency detection on the subjective and objective weights of each of the factors; The third calculation module is configured to calculate the comprehensive weight of each factor based on the subjective weight and objective weight of each factor when the consistency detection result is passed.
9. An electronic device, characterized in that, include: Memory and processor; The processor is configured to read and execute the computer program stored in the memory to implement the steps of the parameter weight determination method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed, implement the steps of the parameter weight determination method as described in any one of claims 1-7.
11. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the parameter weight determination method as described in any one of claims 1-7.