A method for optimizing unconventional reservoir fracturing segmentation based on drilling and logging data

Abstract

The application discloses a kind of unconventional reservoir fracturing segmentation selection cluster optimization method based on drilling and logging data, comprising: S1: obtaining the logging-while-drilling, drilling data and post-fracturing production data of fractured well, determine the geologic evaluation parameter and engineering evaluation parameter;S2: carry out normal distribution conversion to geology, engineering evaluation parameter, and standardization processing, linear combination is principal component for each fracturing section after standardization parameter;S3: determine the membership degree of principal component for different evaluation grades, determine the weight of principal component, obtain the comprehensive score of each fracturing section by the membership degree matrix and weight matrix of principal component, and establish comprehensive sweet spot evaluation model;S4: obtain the comprehensive score profile of the well to be fractured, carry out segmentation selection cluster optimization.The technical scheme provided by the application carries out sweet spot comprehensive evaluation research based on logging-while-drilling and drilling data, with gamma, total hydrocarbon content and other parameters as geologic evaluation parameters, and mechanical specific energy as engineering evaluation parameter, which overcomes the defects in the prior art.

Description

Technical Field

[0001] This invention relates to the field of oil and gas engineering, and in particular to an optimization method for unconventional reservoir fracturing segment clusters based on drilling and logging data. Background Technology

[0002] Horizontal well fracturing with multiple clusters of fractures is a core technology in shale gas development. Long horizontal sections of shale reservoirs exhibit strong heterogeneity, and accurately selecting the reservoir's geological and engineering sweet spot is crucial for effective shale gas well development. Due to development cost constraints, some shale gas horizontal wells have not undergone conventional logging, making it impossible to predict the horizontal section sweet spot using traditional methods based on reservoir porosity, permeability, gas saturation, and engineering parameters such as Young's modulus, Poisson's ratio, and fracture toughness. Therefore, effectively utilizing logging-while-drilling (LOD) and well logging data to guide new well fracturing design is a pressing research area. Summary of the Invention

[0003] This invention mainly overcomes the shortcomings of existing technologies that require physical properties such as porosity, permeability, and gas saturation, as well as engineering parameters such as Young's modulus, Poisson's ratio, and fracture toughness when optimizing fracturing segment clusters. It proposes an unconventional reservoir fracturing segment cluster optimization method based on drilling and logging data, which can optimize fracturing segment clusters based on logging-while-drilling data and logging data.

[0004] The technical solution of the present invention is as follows:

[0005] An optimization method for unconventional reservoir fracturing segment cluster selection based on drilling and logging data includes the following steps:

[0006] S1: Obtain logging-while-drilling data, well logging data, and post-fracture production data of fractured wells in the target block to determine geological evaluation parameters and engineering evaluation parameters;

[0007] S2: The Box-Cox method is used to transform the geological evaluation parameters and the engineering evaluation parameters into a normal distribution. The transformed geological evaluation parameters and engineering evaluation parameters are then standardized. Principal component analysis is used to linearly combine the standardized parameters of each fracturing segment into principal components.

[0008] S3: Construct an interval type II Gaussian distribution membership function for each principal component, determine the membership degree of the principal component for different evaluation levels, determine the weight of the principal component by entropy weight method, combine the principal component membership matrix and weight matrix to obtain the comprehensive score of each fracturing segment, and establish a reservoir geology-engineering integrated sweet spot evaluation model.

[0009] S4: Obtain logging-while-drilling data and well logging data of the wells to be fractured in the target block, substitute the geological evaluation parameters and engineering evaluation parameters of the wells to be fractured into the reservoir geology-engineering integrated sweet spot evaluation model, obtain the comprehensive score profile of the layer to be fractured, and carry out segmented cluster optimization.

[0010] Preferably, the geological evaluation parameters include one or more of the following: gamma, total hydrocarbon content, methane content, and the ant body interpretation index of natural fractures.

[0011] Preferably, the engineering evaluation parameters include bottom hole mechanical specific energy.

[0012] Preferably, the method for calculating the bottom hole mechanical specific energy is as follows:

[0013]

[0014] MSE b Specific energy, MPa; W e For effective drilling pressure, kN; A b The drill bit area is in mm. 2 N b T is the drill bit rotation speed, r / min; b V represents the drill bit torque, in kN·m. pc η is the mechanical drilling rate, m / h; η is the drill bit water power coefficient; Δp b is the drill bit pressure drop, MPa; Q is the drilling fluid discharge rate, L / s.

[0015] Preferably, in step s2, the Box-Cox method is used to transform the geological evaluation parameters and the engineering evaluation parameters to a normal distribution. The method is as follows:

[0016]

[0017] In the formula, y i λ represents the original data, and λ represents the data transformation parameter.

[0018] Preferably, the principal component representation method is as follows:

[0019]

[0020] In the formula, z n v is the nth principal component; nn It is the nth component of the nth eigenvector; This is the value of the nth parameter of the nth factor.

[0021] Preferably, the K-Medoids clustering algorithm is used to optimize segmented cluster selection based on the overall similarity of desserts.

[0022] Preferably, the optimization criterion for the fracturing segment in the segmented clustering is as follows: when the length of the fracturing segment in the clustering result is less than α, it is merged into the adjacent segments for joint division; when the length of the fracturing segment is greater than β, it is divided into the maximum value between α and β, where α is the minimum length of the fracturing segment and β is the maximum length of the fracturing segment.

[0023] Preferably, the optimization criteria for the perforation cluster position in the segmented cluster selection are as follows: the middle of the two fracturing sections is the bridge plug position, no perforation is performed within 10-15m on both sides of the bridge plug, and the remaining positions are selected as the perforation cluster positions based on the cluster spacing requirements and the position with the higher comprehensive score.

[0024] Compared with the prior art, the present invention has the following advantages: The present invention provides an unconventional reservoir fracturing segment cluster optimization method based on drilling and logging data. The improved technical solution uses the gas production index of the production profile test well as the evaluation index. Based on the logging-while-drilling data, it uses gamma, total hydrocarbon content, methane content and natural fracture index as geological characteristic parameters, and mechanical specific energy as engineering characteristic parameter to carry out a comprehensive evaluation study of the geological engineering sweet spot, thus overcoming the defects in the prior art. Attached Figure Description

[0025] Figure 1 This is the original gamma-ray distribution map during drilling in an embodiment of the present invention;

[0026] Figure 2 This is a graph showing the gamma distribution during drilling after normal transformation in an embodiment of the present invention.

[0027] Figure 3 This is a graph showing the principal component analysis results in an embodiment of the present invention;

[0028] Figure 4 This is a principal component loading coefficient diagram in an embodiment of the present invention;

[0029] Figure 5 This is the interval type II Gaussian membership function of principal component 1 in this embodiment of the invention;

[0030] Figure 6 This is a principal component weight distribution diagram in an embodiment of the present invention;

[0031] Figure 7 This is a graph showing the fitting relationship between the gas extraction index and the comprehensive score after normal transformation in an embodiment of the present invention.

[0032] Figure 8 This is a schematic diagram of the basic parameters and fracturing segmentation results of well Q1 in an embodiment of the present invention;

[0033] Figure 9 This is a diagram showing the location selection result of the perforation cluster in well Q1 in an embodiment of the present invention. Detailed Implementation

[0034] The present invention will be further described below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and technical features described in this application can be combined with each other. It should also be pointed out that, unless otherwise indicated, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terms "comprising" or "including" and similar words used in this invention refer to elements or objects preceding the word that encompass the elements or objects listed following the word and their equivalents, without excluding other elements or objects.

[0035] This invention provides an unconventional reservoir fracturing segment cluster optimization method based on drilling and logging data, comprising the following steps:

[0036] S1: Obtain logging-while-drilling data, well logging data, and post-fracture production data of fractured wells in the target block to determine geological evaluation parameters and engineering evaluation parameters.

[0037] In practical applications, since some new wells have not undergone logging, it is impossible to use logging data to guide fracturing optimization in new well design. Specifically, based on the logging-while-drilling, logging, and production data of historical wells in the target block, the gamma ray, total hydrocarbon content, methane content, and ant body interpretation natural fracture index of each fracturing section of the historical wells can be statistically analyzed as geological evaluation parameters, and the bottom hole mechanical specific energy can be calculated based on drilling data as engineering evaluation parameters.

[0038] Mechanical specific energy is the mechanical work required to break a unit volume of rock during drilling, reflecting the mechanical properties of the reservoir in its in-situ state. The corrected model for bottom hole mechanical specific energy is as follows:

[0039]

[0040] MSE b Specific energy, MPa; W e For effective drilling pressure, kN; A b The drill bit area is in mm. 2 N b T is the drill bit rotation speed, r / min; b V represents the drill bit torque, in kN·m. pc η is the mechanical drilling rate, m / h; η is the drill bit water power coefficient; Δp b is the drill bit pressure drop, MPa; Q is the drilling fluid discharge rate, L / s.

[0041] The results of the fracturing test are a direct evaluation index of the effectiveness of fracturing. In order to eliminate the influence of the length of the fracturing section and the fracturing test regime on the analysis results, the production of the fracturing test is converted into the gas production index per unit length meter in this invention.

[0042] S2: The Box-Cox method is used to transform the geological evaluation parameters and the engineering evaluation parameters into a normal distribution. The transformed geological evaluation parameters and engineering evaluation parameters are then standardized. Principal component analysis is used to linearly combine the standardized parameters of each fracturing segment into principal components.

[0043] Since field statistical data often exhibits an irregular or skewed distribution, it cannot satisfy the normal distribution assumption of the Gaussian membership function in the interval type II fuzzy comprehensive evaluation model, which will affect the model's evaluation accuracy. Therefore, this invention employs the Box-Cox transformation method, which converts the data into a normal distribution by determining a suitable exponent λ. This effectively reduces the skewness of the data and makes it closer to a normal distribution, thereby improving the model's reliability and effectiveness.

[0044] Specifically, the transformation equation is:

[0045]

[0046] In the formula y i λ represents the original data, and λ represents the data transformation parameter.

[0047] Principal component analysis (PCA) enables dimensionality reduction and feature extraction, helping to reduce data complexity, remove redundant information, and reveal underlying patterns and structures. PCA projects the original high-dimensional data into a low-dimensional subspace by identifying the principal directions of variance (i.e., principal components), thus preserving as much data variance as possible. Principal components are linear combinations of the original features, arranged in descending order of variance; therefore, the first few principal components typically capture most of the variability in the data.

[0048] To eliminate dimensional differences between parameters and ensure that principal component analysis is based on data variability without being affected by data scale, standardization preprocessing is required for each parameter. The standardization equation is as follows:

[0049]

[0050] In the formula The standardized parameter value; x ij The original value of the parameter; x min,ij The minimum value of the parameter; x max,ij This represents the maximum value of the parameter.

[0051] Eigenvalue decomposition is performed on the covariance matrix of the standardized data to obtain eigenvalues ​​and their corresponding eigenvectors. The eigenvectors represent the principal variance directions of the data, while the eigenvalues ​​represent the magnitude of the variance along these directions. The principal component composition can be represented as:

[0052]

[0053] In the formula z n v is the nth principal component; nn It is the nth component of the nth eigenvector; This is the value of the nth parameter of the nth factor.

[0054] The number of principal components is selected based on the cumulative variance contribution rate. Generally, it is considered that when the variance of the current k principal components accounts for more than 95%, they can contain the main information of the data. The calculation formula is as follows:

[0055]

[0056] In the formula H k h is the cumulative variance contribution rate of the first k principal components. j Let be the variance of the j-th principal component.

[0057] S3: Construct an interval type II Gaussian distribution membership function for each principal component, determine the membership degree of the principal component for different evaluation levels, determine the weight of the principal component by entropy weight method, combine the principal component membership matrix and weight matrix to obtain the comprehensive score of each fracturing segment, and establish a reservoir geology-engineering integrated sweet spot evaluation model.

[0058] In this invention, a type II Gaussian distribution membership function is used to transform the three-dimensional spatial membership function into a two-dimensional uncertainty domain, and its upper and lower membership functions are expressed as follows:

[0059]

[0060]

[0061] In the formula It is the membership function; μ is the membership function; c is the principal component value corresponding to the peak value of the Gaussian distribution; and σ The upper and lower boundaries of the standard deviation are calculated using equation (8):

[0062]

[0063] In the formula, σ is the original standard deviation; ξ is the membership function adjustment factor, ξ∈[0.3,1].

[0064] To achieve differentiated quantitative evaluation of reservoirs, membership functions are divided into four levels based on principal component differences, with Level I = 100, Level II = 75, Level III = 50, and Level IV = 25. The upper membership function (UMF) and lower membership function (LMF) are determined according to equations (7) and (8), and a structure can be constructed as follows: Figure 5 The interval is shown as a type II membership function. Figure 5For each principal component value in the sample, there are two corresponding upper and lower membership values ​​at fuzzy levels. The average of the upper and lower membership values ​​is taken as the membership degree of the principal component with respect to that level. This yields the membership degrees of all principal components for each sample with respect to the four levels, thus constructing the membership degree matrix R. k×4 This lays the foundation for evaluating the impact of each principal component's individual factors on the overall evaluation results.

[0065] In this invention, the entropy weight method is used to measure the degree of uncertainty of each indicator by the information entropy of each indicator, and then to determine the objective weight of each indicator. It has strong versatility in multi-indicator decision-making and evaluation problems.

[0066] The formula for calculating entropy is:

[0067]

[0068] In the formula, m is the sample size; q ij e represents the proportion of each sample point in the principal component; j Let be the entropy value of the j-th principal component.

[0069] In the entropy weight method, the weight of a parameter is inversely proportional to its information entropy. The principal component entropy weight is:

[0070]

[0071] In the formula w j Let be the weight of the j-th principal component.

[0072] S4: Obtain logging-while-drilling data and well logging data of the wells to be fractured in the target block, substitute the geological evaluation parameters and engineering evaluation parameters of the wells to be fractured into the reservoir geology-engineering integrated sweet spot evaluation model, obtain the comprehensive score profile of the layer to be fractured, and carry out segmented cluster optimization.

[0073] Based on the weight analysis results, a weight matrix W can be constructed. 1×k Combining the weight matrix and the membership matrix yields the comprehensive evaluation matrix F for each sample. 1×4 This refers to the degree of membership of each principal component to the four reservoir quality grades. By multiplying the membership degree of each principal component corresponding to the four grades of each sample with the score assigned to each grade and then summing the results, the evaluation result can be quantified into a comprehensive score, thereby constructing a geological-engineering integrated sweet spot evaluation model.

[0074] Based on the logging-while-drilling data and well logging data of the wells to be fractured, geological and engineering evaluation parameters are obtained and substituted into the comprehensive evaluation model to obtain the comprehensive score profile of the reservoir to be stimulated. A higher score indicates better source lithology and fracturing capability at that location. Based on this, optimization of cluster and construction parameter differences is carried out. The K-Medoids clustering algorithm is used to divide the reservoir based on the comprehensive sweet spot similarity.

[0075]

[0076] In the formula, J(u) is the sum of squared errors of each category of samples; K is the number of clusters; m' j x is the number of samples in class j; i For sample points; u j Let j be the j-th cluster center.

[0077] The method for determining the number of clusters K is as follows:

[0078]

[0079] In the formula K max K represents the maximum number of clusters. min α is the minimum number of clusters, L is the length of the horizontal section of the well to be fractured, α is the minimum length of the fractured section, and β is the maximum length of the fractured section.

[0080] Specifically, in this invention, the criteria for dividing fracturing segments are as follows: when the length of a fracturing segment in the clustering results is less than α, it is merged into adjacent segments for joint division; when the length of a fracturing segment is greater than β, it is averaged into the maximum value between α and β. The criteria for selecting the perforation cluster location are as follows: the middle of two fracturing segments is the bridge plug location, no perforation is performed within 10-15m on both sides of the bridge plug, and the remaining locations are selected as perforation cluster locations based on the cluster spacing requirements and the higher comprehensive score.

[0081] Example 1

[0082] In this invention, a segmented cluster optimization application was carried out for the Q1 well in the southeastern Sichuan shale gas reservoir.

[0083] (1) Data processing and model building

[0084] A sample database was constructed based on the geological and engineering parameters of 63 fractured sections from three shale gas horizontal wells in the same block and development strata as well as the meter gas production index under a 50,000 cubic meter production profile testing regime. Some section parameters are shown in Table 1. Among these parameters, the gamma ray while drilling ranges from 205.7 to 391.9 API, total hydrocarbon content from 3.1% to 13.1%, methane content from 2.6% to 12.5%, natural fracture index from 0.02 to 0.65, and mechanical specific energy from 21.4 to 46.3 MPa. The wide distribution of geological and engineering evaluation parameters provides strong guidance for the fracturing design of new wells in the same block and strata.

[0085] Table 1 Partial Sample Data

[0086]

[0087]

[0088] A Box-Cox normal distribution transformation is performed on the original data. Example 1 illustrates the process using drilling gamma transformation as an example. The original drilling gamma distribution is as follows: Figure 1 As shown, the p-value for the normality hypothesis test is 0.012, which is less than the significance level (p>0.05), indicating a significant difference between the original distribution and the normal distribution, necessitating a normality transformation. The gamma distribution during drilling after the Box-Cox transformation is shown below. Figure 2 As shown, the normality test p-value is 0.818, indicating that the transformed data satisfies a normal distribution, laying the foundation for constructing an interval type-2 membership function.

[0089] The principal component analysis results are as follows: Figure 3 As shown, the cumulative variance contribution rate of the first four principal components reaches 99.53%, exceeding 95%. Therefore, the five factors affecting fracturing effectiveness are linearly combined into four principal components. The principal component loading coefficients are as follows: Figure 4 As shown, a principal component expression can be constructed from this, for example, Principal Component 1 = 0.145 * Drilling Gamma - 0.308 * Total Hydrocarbon Content - 0.269 * Methane Content + 0.884 * Natural Fracture Index + 0.172 * MSE b The loading coefficient represents the contribution of each original factor to each principal component. The cumulative loading of each parameter to the principal components is not significantly different, indicating that the principal components can reflect the comprehensive influence of geological and engineering parameters.

[0090] Setting the membership function adjustment factor to 0.6 allows us to construct interval type II membership functions for principal component 1, as shown below. Figure 5 As shown, the membership degrees of each principal component value for each sample with respect to the four levels can be obtained, thus constructing the membership degree matrix R. k×4 .

[0091] The weighting analysis results are as follows Figure 6 As shown, principal component 1 has the largest weight, followed by principal components 3 and 4. Combined with... Figure 3 Analysis revealed that the natural fracture index and total hydrocarbon content contributed the most to principal component 1, while mechanical specific energy and drilling gamma contributed the most to principal components 3 and 4, further demonstrating that the comprehensive evaluation model fully considered the contributions of geological evaluation parameters and engineering evaluation parameters.

[0092] Based on the weight analysis results, a weight matrix W can be constructed. 1×k = [0.467 0.113 0.225 0.195], combining the weight matrix and membership matrix yields the comprehensive evaluation matrix F for each sample. 1×4This refers to the degree of membership of each principal component to the four reservoir quality grades. By multiplying the membership degrees of each of the four principal components corresponding to the four grades for each sample by the assigned scores for each grade and then summing the results, the evaluation result can be quantified into a comprehensive score. The comprehensive evaluation results of the sample data are as follows: Figure 7 As shown, the correlation coefficient between the gas production index and the comprehensive score is 0.9451, indicating that the distribution of each parameter after transformation satisfies the normal distribution assumption of the type II interval membership function. The model can effectively reflect the reservoir production potential under the comprehensive influence of geological engineering parameters.

[0093] (2) Segmentation and clustering optimization of wells to be fractured

[0094] Well Q1 has the same target layer as the sample well, with a horizontal section length of 630m (5050m~5680m). The logging curve is as follows. Figure 9 As shown, the total hydrocarbon content and mechanical energy curves fluctuate significantly, indicating strong heterogeneity of the reservoir. Evaluation results show that the overall score for the horizontal section ranges from 38.6 to 74.9, indicating a moderate sweet spot compared to the sample. The 5050–5114m interval has a higher overall score, with significant advantages in total hydrocarbon and methane content, demonstrating that the sweet spot evaluation effectively reflects the reservoir's heterogeneity and shows good correlation with geological and engineering parameters.

[0095] The block length ranges from 50m to 100m, therefore the number of clusters should be selected from 6 to 12, and the number of clusters should be set to 10. The reservoir segmentation of well Q1 is as follows. Figure 8 As shown, the 630m horizontal section is divided into 10 fracturing sections, with section lengths ranging from 58m to 67m.

[0096] Taking the first and second segments as examples, the optimal location of the perforation clusters is selected. The cluster spacing in this block is 6m to 11m, and the cluster length is 0.5m. According to the perforation cluster selection criteria, each segment is designed with 5 perforation clusters, with a cluster spacing of 6.5m to 10.5m and a distance of 11.5m to 14.5m from the bridge plug. The positions of the bridge plug and the perforation clusters are as follows. Figure 9 As shown.

[0097] This invention provides an unconventional reservoir fracturing segmentation and clustering optimization method based on drilling and logging data. It employs Box-Cox normal transformation and principal component analysis to perform feature engineering on the original data, and combines interval type II fuzzy logic and entropy weight method to quantify the reservoir's geological and engineering sweet spots. Furthermore, this invention uses the K-Medoids clustering algorithm to divide reservoirs based on the similarity of comprehensive sweet spots, and optimizes the segmentation and clustering location differences by matching reservoir geological and engineering characteristics. This achieves fracturing segmentation and clustering optimization using only logging-while-drilling data.

[0098] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An optimization method for unconventional reservoir fracturing segment clusters based on drilling and logging data, comprising the following steps: S1: Obtain logging-while-drilling data, well logging data, and post-fracture production data of fractured wells in the target block to determine geological evaluation parameters and engineering evaluation parameters; The geological evaluation parameters include one or more of the following: gamma, total hydrocarbon content, methane content, and the ant body interpretation index of natural fractures. The engineering evaluation parameters include bottom-hole mechanical specific energy; S2: The Box-Cox method is used to transform the geological evaluation parameters and the engineering evaluation parameters into a normal distribution. The transformed geological evaluation parameters and engineering evaluation parameters are then standardized. Principal component analysis is used to linearly combine the standardized parameters of each fracturing segment into principal components. S3: Construct an interval type II Gaussian distribution membership function for each principal component, determine the membership degree of the principal component for different evaluation levels, determine the weight of the principal component by entropy weight method, combine the principal component membership matrix and weight matrix to obtain the comprehensive score of each fracturing segment, and establish a reservoir geology-engineering integrated sweet spot evaluation model. S4: Obtain logging-while-drilling data and well logging data of the wells to be fractured in the target block, substitute the geological evaluation parameters and engineering evaluation parameters of the wells to be fractured into the reservoir geology-engineering integrated sweet spot evaluation model, obtain the comprehensive score profile of the layer to be fractured, and carry out segmented cluster optimization. The optimization criterion for fracturing segments in segmented clustering is: when the length of the fracturing segment in the clustering result is less than... α When the length of the fracturing section is greater than a certain value, it is incorporated into the adjacent section for joint division; when the length of the fracturing section is greater than a certain value, it is incorporated into the adjacent section for joint division. β At that time, divide it into equal parts. α and β The maximum value between α This is the minimum length of the fracturing section. β This represents the maximum length of the fracturing section; The optimization criteria for the perforation cluster position in the segmented cluster selection are as follows: the middle of the two fracturing sections is the bridge plug position, no perforation is allowed within 10-15m on both sides of the bridge plug, and the remaining positions are selected as the perforation cluster positions based on the cluster spacing requirements and the position with the higher comprehensive score.

2. The unconventional reservoir fracturing segment cluster optimization method as described in claim 1, wherein the bottom hole mechanical specific energy calculation method is as follows: ； (1) In the formula MSE b Specific energy, MPa; W e For effective drilling pressure, kN; A b The drill bit area is in mm. 2 ; N b The drill bit rotation speed is r / min; T b The drill bit torque is expressed in kN·m. V pc The drilling speed is mechanical, in m / h; η Δ is the drill bit water power coefficient; p b The drill bit pressure drop is measured in MPa. Q This represents the drilling fluid discharge rate, in L / s.

3. The unconventional reservoir fracturing segmented cluster optimization method as described in claim 1, wherein in step S2, the Box-Cox method is used to transform the geological evaluation parameters and the engineering evaluation parameters to a normal distribution, the method being: (2)； In the formula, y i The original data, λ These are the data transformation parameters.

4. The unconventional reservoir fracturing segmented cluster optimization method as described in claim 1, wherein the principal component representation method is as follows: (4)； In the formula, z n For the first n One principal component; v nn For the first n The th eigenvector of the th feature vector n One component; For the first n The first factor n Each parameter value.

5. The unconventional reservoir fracturing segmented cluster optimization method as described in claim 1 employs the K-Medoids clustering algorithm to perform segmented cluster optimization based on the comprehensive sweet spot similarity.

6. An unconventional reservoir fracturing segment cluster optimization system based on drilling and logging data, used to implement the unconventional reservoir fracturing segment cluster optimization method according to any one of claims 1-5. Includes the following modules: The evaluation parameter acquisition module acquires logging-while-drilling data, well logging data, and post-fracture production data of fractured wells in the target block to determine geological evaluation parameters and engineering evaluation parameters. The principal component determination module uses the Box-Cox method to transform the geological evaluation parameters and the engineering evaluation parameters into a normal distribution. The transformed geological evaluation parameters and engineering evaluation parameters are then standardized. Finally, the standardized parameters of each fracturing segment are linearly combined into principal components using principal component analysis. The comprehensive sweet spot evaluation module constructs an interval type II Gaussian distribution membership function for each principal component, determines the membership degree of the principal component for different evaluation levels, determines the weight of the principal component through the entropy weight method, and combines the principal component membership matrix and weight matrix to obtain the comprehensive score of each fracturing segment, thus establishing a reservoir geology-engineering comprehensive sweet spot evaluation model. The segmented cluster optimization module acquires logging-while-drilling data and well logging data of the wells to be fractured in the target block, substitutes the geological evaluation parameters and engineering evaluation parameters of the wells to be fractured into the reservoir geology-engineering integrated sweet spot evaluation model, obtains the comprehensive score profile of the layer to be fractured, and carries out segmented cluster optimization.

Images