An intelligent well completion optimization design method and device based on multi-objective optimization

Abstract

The present invention discloses an intelligent well completion optimization design method and device based on multi-objective optimization. The method comprises: clustering a first data set according to a weighted clustering model, and evaluating the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result. The first data set includes reservoir physical property parameters along the depth of the horizontal well. The weighted clustering model is constructed based on the weights of each reservoir physical property parameter; dividing the fracturing stages according to the engineering sweet spot evaluation results and preset stage length constraints, wherein the engineering sweet spot evaluation results are determined based on geomechanical parameters; solving a multi-objective function constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results, and the preset cluster spacing constraints within the divided fracturing stages, and performing well completion optimization design based on the objective solution of the multi-objective function. The present invention can perform well completion optimization design efficiently, meticulously, and intelligently.

Description

Technical Field

[0001] The present invention relates to the technical field of reservoir stimulation and transformation, and in particular to an intelligent well completion optimization design method and device based on multi-objective optimization. Background Art

[0002] Refined and intelligent completion optimization design is crucial for the development and transformation of unconventional oil reservoirs.

[0003] However, existing technologies that use clustering algorithms to identify and evaluate sweet spots require significant manpower and time to interpret the clustering results, limiting the ability to optimize completion designs based on sweet spots. Furthermore, existing technologies often employ geometric or single-point completion design approaches, making it difficult to achieve balanced fracturing stimulation. Consequently, existing completion optimization design methods suffer from low intelligence and efficiency, poor design results, and difficulty achieving balanced fracturing stimulation.

[0004] To address the above issues, no effective solutions have been proposed so far. Summary of the Invention

[0005] The embodiments of this specification provide an intelligent completion optimization design method and device based on multi-objective optimization to solve the problems of existing completion optimization design methods such as low intelligence and efficiency, poor design effect, and difficulty in achieving balanced fracturing transformation.

[0006] In a first aspect, the embodiments of this specification provide an intelligent completion optimization design method based on multi-objective optimization, including:

[0007] Clustering a first data set according to a weighted clustering model, and evaluating a geological sweet spot of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes reservoir physical property parameters along a horizontal well depth, and the weighted clustering model is constructed according to a weight of each reservoir physical property parameter;

[0008] Dividing the fracturing stages according to the engineering sweet spot evaluation results and the preset stage length constraints, wherein the engineering sweet spot evaluation results are determined according to geomechanical parameters;

[0009] Within the divided fracturing sections, a multi-objective function constructed based on the geological sweet spot evaluation results, engineering sweet spot evaluation results, and preset cluster spacing constraints is solved, and completion optimization design is performed based on the target solution results of the multi-objective function.

[0010] In some embodiments, the method further comprises:

[0011] Calculate the correlation coefficient between each reservoir physical property parameter and oil production profile data;

[0012] Determine the sum of the absolute values ​​of the correlation coefficients between each reservoir physical property parameter and the oil production profile data;

[0013] The weight of each reservoir physical property parameter is determined based on the correlation coefficient between the reservoir physical property parameters and the oil production profile data and the sum of the absolute values ​​of the correlation coefficients.

[0014] In some embodiments, the weighted clustering model is constructed based on the weights of the reservoir physical property parameters, including:

[0015] According to the weights of the reservoir physical parameters, weighted Euclidean distance function, criterion function and rating function are constructed;

[0016] A weighted clustering model is constructed based on the weighted Euclidean distance function, criterion function and rating function.

[0017] In some embodiments, clustering the first data set according to the weighted clustering model includes:

[0018] selecting a preset number of cluster centers from the first data set;

[0019] Calculating the weighted Euclidean distance from each first sample point in the first data set to the cluster center according to the weighted Euclidean distance function, wherein each first sample point corresponds to each reservoir physical property parameter at different horizontal well depths;

[0020] Comparing the weighted Euclidean distances, and dividing each first sample point into a cluster corresponding to a weighted Euclidean distance less than a preset distance threshold according to the comparison result;

[0021] According to the criterion function, a new cluster center is selected from the cluster clusters;

[0022] Repeat the process of calculating the weighted Euclidean distance, dividing each first sample point, and selecting a new cluster center until the number of iterations reaches a preset iteration threshold or the new cluster center is unchanged, thereby obtaining a first clustering result after clustering;

[0023] The geological sweet spot of each cluster after clustering is evaluated to obtain a geological sweet spot evaluation result, including:

[0024] The clustered geological sweet spots in the first clustering result are evaluated using a rating function to obtain a geological sweet spot evaluation result.

[0025] In some embodiments, the use of a rating function to evaluate the clustered geological sweet spots in the first clustering result to obtain a geological sweet spot evaluation result includes:

[0026] Using the rating function, determine the weighted geological sweet spot index of each cluster in the first clustering result;

[0027] Compare the weighted geological sweet spot indexes of each cluster, sort the comparison results of the weighted geological sweet spot indexes in descending order, and obtain the sorting results of the weighted geological sweet spot indexes;

[0028] According to the ranking result of the weighted geological dessert index, the corresponding geological dessert is graded into different levels to obtain a geological dessert evaluation result, which includes a geological dessert category label.

[0029] In some embodiments, the engineering sweet spot evaluation result includes geomechanical strength along the depth of the horizontal well; accordingly, dividing the fracturing stages according to the engineering sweet spot evaluation result and the preset stage length constraint includes:

[0030] constructing a second data set based on the geomechanical strength along the horizontal well depth and the well depth, wherein each second sample point in the second data set corresponds to the geomechanical strength of different well depths and different horizontal well depths;

[0031] constructing a geomechanical similarity term based on the geomechanical strength of the second sample point and the average geomechanical strength of all second sample points;

[0032] Construct a segment length penalty term based on the well depth range of the fracturing segment and the preset segment length constraint;

[0033] According to the geomechanical similarity term and the segment length penalty term, the objective function for dividing the fracturing segments is constructed;

[0034] A dynamic programming algorithm is used to solve the objective function and determine the segmentation points of the fracturing stage;

[0035] The fracturing segments are divided according to the segmentation points to obtain a fracturing segment division result, wherein the fracturing segment division result includes a segment label.

[0036] In some embodiments, the method further comprises:

[0037] Determine the weight of the geological dessert according to the geological dessert category label in the geological dessert evaluation result;

[0038] According to the segment labels in the fracturing segment division results, the fracturing segments are identified and constraints are imposed on the fracturing clusters within the fracturing segments, so that the constrained fracturing clusters are located within a preset segment depth range;

[0039] Generating candidate regions according to the geomechanical strength along the depth of the horizontal well, and generating candidate clusters from the candidate regions;

[0040] The first goal is to make the sum of the weights of the geological sweet spots greater than the preset weight threshold, the second goal is to make the geomechanical similarity term less than the preset variance threshold, and the preset cluster spacing requirement is used as the constraint condition;

[0041] A multi-objective function is constructed based on the first objective, the second objective and the constraints.

[0042] In some embodiments, solving a multi-objective function constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results, and the preset cluster spacing constraint includes:

[0043] A multi-objective optimization algorithm is used to solve the multi-objective function and determine the target solution of the multi-objective function from the candidate clusters;

[0044] The well completion optimization design is performed according to the objective solution results of the multi-objective function, including:

[0045] The target solution result is used as the target design scheme of the fracturing cluster within the fracturing stage, and the completion optimization design is carried out according to the target design scheme of the fracturing cluster.

[0046] In a second aspect, the embodiments of this specification further provide an intelligent well completion optimization design device based on multi-objective optimization, comprising:

[0047] a geological sweet spot evaluation module, configured to cluster a first data set according to a weighted clustering model, and evaluate the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes reservoir physical property parameters along the depth of the horizontal well, and the weighted clustering model is constructed based on the weights of the reservoir physical property parameters;

[0048] A fracturing segment division module, configured to divide the fracturing segments according to the engineering sweet spot evaluation results determined based on geomechanical parameters and preset segment length constraints;

[0049] The completion optimization design module is used to solve the multi-objective function constructed based on the geological sweet spot evaluation results, engineering sweet spot evaluation results and preset cluster spacing constraints within the divided fracturing stages, and perform completion optimization design based on the target solution results of the multi-objective function.

[0050] In a third aspect, an embodiment of this specification further provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the steps of the above-mentioned intelligent completion optimization design method based on multi-objective optimization.

[0051] The embodiments of this specification provide an intelligent completion optimization design method and device based on multi-objective optimization. First, a first data set is clustered according to a weighted clustering model, and the geological sweet spots of each cluster after clustering are evaluated to obtain a geological sweet spot evaluation result. The first data set includes the physical property parameters of each reservoir along the depth of the horizontal well, and the weighted clustering model is constructed according to the weights of each reservoir physical property parameter. Then, the fracturing section is divided according to the engineering sweet spot evaluation results and the preset section length constraints. The engineering sweet spot evaluation results are determined according to geomechanical parameters. Finally, within the divided fracturing section, a multi-objective function constructed according to the geological sweet spot evaluation results, the engineering sweet spot evaluation results, and the preset cluster spacing constraints is solved, and the completion optimization design is performed according to the objective solution results of the multi-objective function. In the embodiments of this specification, by introducing a weighted clustering model, intelligent evaluation of geological sweet spots can be achieved without spending a lot of manpower and time costs to interpret the clustering results. Geomechanical parameters enable accurate engineering sweet spot evaluation. Determining the engineering sweet spot evaluation results, combined with preset segment length constraints, enables intelligent fracturing stage design, resolving the issue of uneven fracturing caused by traditional geometric or single-stage spacing designs. Solving multi-objective functions enables intelligent and refined completion optimization design, improving engineering applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the 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 work. In the drawings:

[0053] Figure 1 This is a flow chart of an intelligent well completion optimization design method based on multi-objective optimization provided in an embodiment of this specification;

[0054] Figure 2 This is a schematic diagram of an embodiment of an intelligent well completion optimization design method based on multi-objective optimization provided by an embodiment of this specification, in a scenario example;

[0055] Figure 3 This is a schematic diagram of the structure of an intelligent well completion optimization design device based on multi-objective optimization provided in an embodiment of this specification;

[0056] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0057] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments derived by those skilled in the art based on the embodiments in this specification without creative effort shall fall within the scope of protection of this specification.

[0058] Unconventional oil and gas resources, primarily shale formations, are playing an increasingly important role in energy supply. Multi-stage hydraulic fracturing is a key technology for developing unconventional reservoirs. Implementing effective completion design plays a crucial role in the development and transformation of unconventional reservoirs, aiming to achieve balanced production increases across all fracture stages. Accurately assessing sweet spots (areas with favorable geological and engineering characteristics) is a prerequisite for effective fracturing design. Identifying these sweet spots and optimizing completion design are crucial for targeting high-quality reservoir formations and achieving uniform fracture initiation and propagation.

[0059] The concept of "sweet spots" is widely used in oil and gas exploration and development, and can include geological sweet spots and engineering sweet spots. Geological sweet spots refer to areas with high hydrocarbon content and good rock physical properties, and are usually evaluated using parameters such as clay content, total organic carbon, porosity, and permeability. Engineering sweet spots refer to areas that are conducive to efficient hydraulic fracturing and are evaluated based on geomechanical indicators such as Young's modulus, Poisson's ratio, and in-situ stress. However, the strong nonlinear relationships and complexity of these parameters make it challenging to discern their relative importance.

[0060] Currently, few studies have comprehensively considered the geological and engineering "dual sweet spots" and applied them to completion design. Among these studies, a common approach is to determine optimal parameter classification thresholds using single-factor evaluation methods, multi-factor correlation analysis, or radar regional models. This approach then constructs a coupled classification matrix of geological and engineering sweet spots, forming a comprehensive "dual sweet spot" evaluation table. Based on this evaluation table, perforation clusters (also known as fracturing clusters) increase during fracturing stages with favorable "dual sweet spots" and become sparse during fracturing stages with unfavorable "dual sweet spots." However, this approach has significant limitations in evaluating and balancing geological and engineering sweet spots. First, these classification thresholds often vary depending on the selected method and reservoir characteristics, resulting in challenges such as non-uniqueness and limited regional applicability. Second, while cluster density is increased in high-quality "dual sweet spot" fracturing stages, clusters are located using geometric or single designs within each stage without considering variations in fracture initiation pressure between clusters. Increasing monitoring data indicates that traditional geometric or single completion designs struggle to achieve balanced fracturing stimulation.

[0061] With the application of artificial intelligence in oil and gas exploration and development, clustering algorithm has become a promising sweet spot identification method, but it is difficult to distinguish and integrate geological sweet spots and engineering sweet spots. It requires a lot of effort to combine mechanistic cognition to interpret the clustering results, which limits its ability to fully consider the "double sweet spots" in completion optimization design.

[0062] In summary, the existing completion optimization design methods have problems such as low intelligence and efficiency, poor design effect, and difficulty in achieving balanced fracturing transformation.

[0063] To address the aforementioned issues, embodiments of this specification provide an intelligent completion optimization design method and apparatus based on multi-objective optimization. First, a first data set is clustered according to a weighted clustering model, and the geological sweet spots of each cluster are evaluated to obtain a geological sweet spot evaluation result. The first data set includes reservoir physical property parameters along the depth of the horizontal well. The weighted clustering model is constructed based on the weights of each reservoir physical property parameter. Then, based on the engineering sweet spot evaluation results and preset segment length constraints, the fracturing stages are divided. The engineering sweet spot evaluation results are determined based on geomechanical parameters. Finally, within the divided fracturing stages, a multi-objective function constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results, and the preset cluster spacing constraints is solved, and completion optimization design is performed based on the objective solution of the multi-objective function.

[0064] By introducing a weighted clustering model, intelligent evaluation of geological sweet spots is possible without the extensive human and time-consuming effort required to interpret clustering results. Geomechanical parameters enable accurate engineering sweet spot evaluation. Determining the engineering sweet spot evaluation results, combined with pre-set segment length constraints, enables intelligent fracturing stage design, resolving the issue of uneven fracturing initiation associated with traditional geometric or single-stage spacing designs. Solving multi-objective functions enables intelligent and refined completion optimization design, improving engineering applicability.

[0065] It should be noted that the terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, for the purposes of describing the embodiments of this application.

[0066] It is understood that the above methods provided in the embodiments of this specification can be applied to electronic devices, which can refer to electronic devices with data computing, processing, and storage capabilities. The electronic device can be a terminal such as a PC (Personal Computer), a tablet computer, a smartphone, a wearable device, an intelligent robot, etc.; it can also be a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0067] See Figure 1 As shown, the embodiment of this specification provides an intelligent completion optimization design method based on multi-objective optimization. In specific implementation, the method may include the following contents:

[0068] S101: Clustering a first data set according to a weighted clustering model, and evaluating the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes reservoir physical property parameters along the depth of a horizontal well, and the weighted clustering model is constructed based on the weights of the reservoir physical property parameters.

[0069] In some embodiments, prior to the above-mentioned S101, readily available and valuable data from the drilling, completion, and production processes may be collected, such as logging data, drilling data, and post-fracturing monitoring data. Logging data may include, but is not limited to, gamma ray logging (GR), density logging (DEN), acoustic logging (DT), neutron logging (CNL), and resistivity logging (RT). Drilling data may include, but is not limited to, weight on bit (WOB), torque (T), rotational speed (N), rate of penetration (ROP), and drill bit diameter (Ab). Monitoring data may include, but is not limited to, oil production profile data obtained through tracer diagnosis and fluid distribution data obtained through fiber optic monitoring.

[0070] Because the quality of collected data significantly impacts the performance of subsequent models and algorithms, and raw data often contains discontinuities and anomalies caused by environmental factors, instrument performance, and human error, data preprocessing is necessary. Data preprocessing can include filling outliers and missing values ​​in the collected data using the KNN algorithm, standardizing the accuracy of logging data and drilling data to 0.125m, and other methods, though this specification does not provide specific limitations.

[0071] Afterwards, the reservoir physical parameters and geomechanical parameters can be calculated based on the pre-processed drilling data and / or logging data. The specific calculation formula can refer to the existing technology and will not be described in detail in this specification. Among them, the reservoir physical parameters may include at least one of the following: mud content, porosity, oil saturation, and permeability. The geomechanical parameters may include at least one of the following: bottom hole mechanical specific energy and minimum horizontal principal stress. Reservoir physical parameters can be used to evaluate geological sweet spots, and geomechanical parameters can be used to evaluate engineering sweet spots. Reservoir physical parameters and geomechanical parameters can be stored in ascending order by well depth, with an interval of 0.125m. Reservoir physical parameters and geomechanical parameters can effectively characterize the flow capacity, storage performance and mechanical characteristics of the reservoir. They are calculated from oilfield logging and drilling data without the need for special logging programs, which can ensure the practical applicability of large-scale field implementation.

[0072] Afterwards, in order to reduce noise and improve the reliability of subsequent cluster evaluation, the reservoir physical properties and geomechanical parameters can be denoised and smoothed.

[0073] For example, since the curves of reservoir physical parameters all represent spatially ordered data along the wellbore depth, the depth axis is analogous to a time series. Empirical mode decomposition (EMD) can be used to perform noise reduction on the reservoir physical parameters. The specific processing process is as follows:

[0074] 1) IMF extraction: Iteratively extract the signal extreme envelope to generate IMF components from high to low frequency;

[0075] 2) Noise component identification: Calculate the sample entropy of the first two order IMFs. If the entropy exceeds the threshold of 1.2, it is determined to be the noise-dominant component.

[0076] 3) Signal reconstruction: After removing the noise IMF, the remaining IMF and the residual term are superimposed to obtain the denoised reservoir physical property parameters.

[0077] The reservoir physical property parameters after noise reduction are then normalized. For details, please refer to existing technologies and will not be described in detail in this specification. By using this normalization method, the reservoir physical property parameters can be converted to a standard normal distribution with a mean of 0 and a standard deviation of 1, eliminating the impact of different parameter dimensions on the data while maintaining the data distribution characteristics.

[0078] In some embodiments, before the above S101, the specific implementation may also include:

[0079] Calculate the correlation coefficient between each reservoir physical property parameter and oil production profile data;

[0080] Determine the sum of the absolute values ​​of the correlation coefficients between each reservoir physical property parameter and the oil production profile data;

[0081] The weight of each reservoir physical property parameter is determined based on the correlation coefficient between the reservoir physical property parameters and the oil production profile data and the sum of the absolute values ​​of the correlation coefficients.

[0082] Specifically, we can first calculate the reservoir physical parameters along the horizontal well depth (such as mud content V sh , porosity f, oil saturation So, permeability K) to construct the first data set X1:

[0083]

[0084] Among them, each row vector x i The reservoir physical parameters (φ, S o ,k,V sh )vector.

[0085] Then, the correlation analysis of each reservoir physical property parameter and the oil production profile data can be performed to calculate the correlation coefficient between each reservoir physical property parameter and the oil production profile data. Since the relationship between reservoir physical properties and oil production profile data is usually nonlinear, the Spearman method can be selected to perform correlation analysis on each reservoir physical property parameter after noise reduction and smoothing with the oil production profile data in the monitoring data to calculate the Spearman correlation coefficient. The Spearman method is a non-parametric statistical method that can evaluate monotonic relationships (linear or nonlinear) without the need for normality assumptions and is robust to outliers. Specifically, the correlation coefficient between each reservoir physical property parameter and the oil production profile data can be calculated according to the following formula:

[0086]

[0087] Where r is the correlation coefficient; is the rank difference between the observed values ​​of the jth reservoir physical property parameter and the oil production profile data; n is the number of first sample points. The rank difference can be determined by sorting the reservoir physical property parameters and the oil production profile data, determining the rank of the sorted reservoir physical property parameters and the rank of the sorted oil production profile data, and then taking the difference between the two to determine the rank difference between the observed values ​​of the reservoir physical property parameters and the oil production profile data.

[0088] The sum of the absolute values ​​of the correlation coefficients between each reservoir physical property parameter and the oil production profile data can be determined according to the following formula:

[0089]

[0090] The weight of each reservoir physical property parameter can be determined according to the following formula based on the correlation coefficient between the reservoir physical property parameter and the oil production profile data and the sum of the absolute values ​​of the correlation coefficients (the weight is assigned according to the absolute value of the correlation coefficient, and the sum of the weights of the reservoir physical property parameters is ensured to be equal to 1):

[0091]

[0092] Among them, w j is the weight of the jth reservoir physical property parameter; r j is the correlation coefficient between the jth reservoir physical property parameter and the oil production profile data; p is the total number of reservoir physical property parameters; |r j | is the absolute value of the correlation coefficient of the j-th reservoir physical property parameter; It is the sum of the absolute values ​​of the correlation coefficients between each reservoir physical property parameter and oil production profile data.

[0093] In some embodiments, the weighted clustering model in S101 is constructed based on the weights of the reservoir physical property parameters. In specific implementation, it may include:

[0094] According to the weights of the reservoir physical parameters, weighted Euclidean distance function, criterion function and rating function are constructed;

[0095] A weighted clustering model is constructed based on the weighted Euclidean distance function, criterion function and rating function.

[0096] Specifically, the weighted Euclidean distance function can be constructed according to the following formula:

[0097]

[0098] Among them, D(x i ,c J ) is the weighted Euclidean distance, which can be used as a weighted Euclidean distance function; x i is the reservoir physical property parameter vector of the first sample point i; c J is the centroid of the Jth cluster; p is the total number of reservoir physical property parameters; |w j | is the absolute value of the weight of the jth reservoir physical property parameter; x ij is the jth reservoir property parameter vector of the ith first sample point; c Jj is the centroid of the jth reservoir property parameter in the Jth cluster. The weighted Euclidean distance function can be used to evaluate the distance between each first sample point and the corresponding cluster center within each cluster. To enhance the influence of key petrophysical parameters on the clustering process, the weighted Euclidean distance can be used as a similarity measure between samples.

[0099] The criterion function can be constructed according to the following formula:

[0100]

[0101] Among them, WSSE is the weighted sum of square errors of K clusters, which can be used as a criterion function; K is the total number of preset clusters; x iis the reservoir physical property parameter vector of the first sample point i; C J are all the first samples belonging to the Jth cluster; p is the total number of reservoir physical property parameters; |w j | is the absolute value of the weight of the jth reservoir physical property parameter; x ij is the jth reservoir property parameter vector of the ith first sample point; c Jj is the centroid of the jth reservoir property parameter of the Jth cluster. K can be determined based on the "elbow rule." When K is below an appropriate value, increasing K significantly reduces the WSSE. However, when K exceeds this value, the reduction in WSSE is negligible, and this value is the optimal K. A criterion function can be used to evaluate clustering effectiveness.

[0102] The rating function can be constructed according to the following formula:

[0103]

[0104] Among them, WGSSI J is the weighted geological sweet spot index of cluster J, which can be used as a rating function; f is porosity; So is oil saturation; K is permeability; V sh is the mud content; w j is the weight of the j-th reservoir physical property parameter; is the 25th percentile of the j-th reservoir physical property parameter in the J-th cluster; is the 75th percentile of the jth reservoir property parameter in the Jth cluster. Among them, the horizontal well interval is divided into K different clusters, but their relative quality cannot be directly evaluated. In order to quantitatively evaluate the reservoir quality and link the clustering results with mechanistic understanding, the weighted geological sweet spot index (WGSSI) is proposed. This index integrates the weights obtained from the correlation of oil production contribution to reflect the relative contribution of rock physical parameters to reservoir performance. In order to minimize the impact of outliers, the analysis can focus on the interquartile range (i.e., the 25th to 75th percentile) of the normalized values ​​of each reservoir property parameter in each cluster. WGSSI J The larger the value, the better the geological sweet spot and the more favorable the reservoir quality is for production.

[0105] Subsequently, a mechanism-guided weighted K-medoids clustering model or algorithm can be constructed based on the weighted Euclidean distance function, criterion function, and rating function, enabling intelligent identification of geological sweet spots consistent with mechanism-based cognition. The K-medoids clustering algorithm improves clustering effectiveness by repeatedly replacing the cluster center reference point with each data point. Correlation analysis is performed between reservoir physical properties and oil production profile data, and a weighted distance function, criterion function, and rating function are constructed that emphasize key reservoir physical properties. This dual focus on cluster centers and weighting functions enhances robustness to outliers and better captures the underlying geological mechanisms.

[0106] In some embodiments, clustering the first data set according to the weighted clustering model in S101 may include:

[0107] selecting a preset number of cluster centers from the first data set;

[0108] Calculating the weighted Euclidean distance from each first sample point in the first data set to the cluster center according to the weighted Euclidean distance function, wherein each first sample point corresponds to each reservoir physical property parameter at different horizontal well depths;

[0109] Comparing the weighted Euclidean distances, and dividing each first sample point into a cluster corresponding to a weighted Euclidean distance less than a preset distance threshold according to the comparison result;

[0110] According to the criterion function, a new cluster center is selected from the cluster clusters;

[0111] Repeat the process of calculating the weighted Euclidean distance, dividing each first sample point, and selecting a new cluster center until the number of iterations reaches a preset iteration threshold or the new cluster center is unchanged, thereby obtaining a first clustering result after clustering;

[0112] The above S101 evaluates the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result. When specifically implemented, it may include:

[0113] The clustered geological sweet spots in the first clustering result are evaluated using a rating function to obtain a geological sweet spot evaluation result.

[0114] In some embodiments, the above-mentioned evaluation of the clustered geological sweet spots in the first clustering result using the rating function to obtain the geological sweet spot evaluation result may include:

[0115] Using the rating function, determine the weighted geological sweet spot index of each cluster in the first clustering result;

[0116] Compare the weighted geological sweet spot indexes of each cluster, sort the comparison results of the weighted geological sweet spot indexes in descending order, and obtain the sorting results of the weighted geological sweet spot indexes;

[0117] According to the ranking result of the weighted geological dessert index, the corresponding geological dessert is graded into different levels to obtain a geological dessert evaluation result, which includes a geological dessert category label.

[0118] Specifically, the steps of clustering the first data set using the weighted K-medoids clustering algorithm (i.e., the weighted clustering model described above) are as follows:

[0119] 1) Initialization: Randomly select a preset number (e.g., K) cluster centers from the first data set.

[0120] 2) Weighted Euclidean distance allocation: According to the weighted Euclidean distance function, the weighted Euclidean distance of each first sample point to the K cluster centers is calculated, and each weighted Euclidean distance is compared. According to the comparison result, it is divided into the cluster cluster with the smallest weighted Euclidean distance (that is, the weighted Euclidean distance is less than the preset distance threshold. The preset distance threshold can be set according to actual needs and is not specifically limited in this specification).

[0121] 3) Cluster center update: For each cluster, a new cluster center is selected by minimizing the criterion function WSSE.

[0122] 4) Iteration: Repeat steps 2)-3) until convergence (selection of a new cluster center remains unchanged) or the number of iterations reaches a preset iteration threshold. The preset iteration threshold can be set according to actual needs and is not specifically limited in this specification.

[0123] 5) Results: The first clustering results are labeled using a rating function, and the geological sweet spots are divided or rated into four categories (I, II, III, and IV), corresponding to "Level 1 (high quality)", "Level 2 (good)", "Level 3 (fair)", and "Level 4 (poor)", respectively, corresponding to the weighted geological sweet spot index (WGSSI). Specifically, the rating function can be used to determine the weighted geological sweet spot index of each cluster in the first clustering result. The weighted geological sweet spot indexes of each cluster are then compared and sorted in descending order based on the comparison results of the weighted geological sweet spot index. Finally, based on the sorted weighted geological sweet spot index, the corresponding geological sweet spots are graded into different levels to obtain the geological sweet spot evaluation results. For example, the top-ranked cluster has the highest weighted geological sweet spot index and is the best geological sweet spot, and can be rated as a Level 1 geological sweet spot. The bottom-ranked cluster has the lowest weighted geological sweet spot index and is the worst geological sweet spot, and can be rated as a Level 4 geological sweet spot. I, II, III, and IV can be used as geological sweet spot category labels.

[0124] The main goal of geological sweet spot identification and evaluation is to identify reservoirs with good reservoir flow capacity and reservoir performance (such as high porosity, high oil saturation, high permeability and low clay content or low mud content). Clustering algorithms are usually used to classify well sections with similar reservoir properties into the same category. However, unsupervised clustering techniques based on Euclidean methods usually assign the same weight to reservoir rock physical parameters, while ignoring their differential contributions to production dynamics. In order to address this limitation, the present application constructs a mechanism-guided weighted clustering model based on the weights of each reservoir physical parameter. Compared with traditional technologies, this application can focus on reservoir physical parameters that have a great impact on oil production, reduce the impact of reservoir physical parameters that have a small impact on oil production on clustering results, and realize intelligent evaluation of geological sweet spots.

[0125] S102: Divide the fracturing stages according to the engineering sweet spot evaluation result and the preset stage length constraint, wherein the engineering sweet spot evaluation result is determined according to geomechanical parameters.

[0126] In some embodiments, the engineering sweet spot evaluation result in S102 is determined based on geomechanical parameters, and in specific implementation, may include:

[0127] Based on the bottom hole mechanical specific energy and the minimum horizontal principal stress, the geomechanical strength is determined according to the following formula:

[0128] P Geo =MSE b +S hmin

[0129] Among them, P Geo is the geomechanical strength, MPa; MSE b is the bottom hole mechanical specific energy, MPa; S hmin is the minimum horizontal principal stress, MPa;

[0130] The engineering sweet spot is evaluated according to the distribution of geomechanical strength to obtain the engineering sweet spot evaluation result.

[0131] Specifically, bottom hole mechanical specific energy MSE b It can be used as a representation of the mechanical strength of reservoir rocks and can effectively evaluate the crack resistance of rocks under different pressure conditions. hmin It plays a key role in the initiation and propagation of fractures, influencing rock deformation and failure patterns. By comprehensively considering these two factors, we can more comprehensively evaluate the geomechanical characteristics of the reservoir, providing a scientific basis for engineering design and ultimately determining engineering sweet spot parameters, which provide an important basis for well completion optimization. The above geomechanical strength distribution can intuitively demonstrate the engineering sweet spot of the horizontal well section.

[0132] In some embodiments, each perforation cluster within a certain fracturing section along a horizontal well has a corresponding fracture initiation pressure. When the wellbore pressure exceeds this threshold, the corresponding fracture will initiate. The fracture initiation condition can be expressed as:

[0133] P w >T str,eff,c +s init,c +Ds e,c +Ds i,c

[0134] Among them, P w is the wellbore pressure (MPa); T str,eff,c is the effective tensile strength (MPa); s init,c is the stress shadow of the previous stage (MPa); Ds e,c is the stress shadow of other cracks in the same stage (MPa); Ds i,c is the initial magnitude of the minimum principal stress at the cluster (MPa).

[0135] In the fracturing initiation stage, when the cracks have not yet expanded on a large scale and the interference between cracks can be ignored, the fracturing initiation pressure condition can be further simplified as:

[0136] P W 'V≥P Geo =MSE b +S hmin

[0137] Among them, P W ' is the cracking pressure, MPa.

[0138] Engineering sweet spots can also be evaluated based on the distribution of geomechanical strength and fracture initiation pressure, yielding engineering sweet spot evaluation results. For example, moderate geomechanical strength and low fracture initiation pressure indicate high fracturing efficiency and controllable fractures, making development a priority. The corresponding engineering sweet spot can be rated as a first-class (high-quality) engineering sweet spot. Excessively high or low geomechanical strength or excessively high fracture initiation pressure indicates a high fracturing risk and is not recommended for development. The corresponding engineering sweet spot can be rated as a non-sweet spot.

[0139] In some embodiments, the engineering sweet spot evaluation results in S102 may include geomechanical strength along the depth of the horizontal well. Accordingly, the division of the fracturing stages according to the engineering sweet spot evaluation results and the preset stage length constraints in S102 may include:

[0140] constructing a second data set based on the geomechanical strength along the horizontal well depth and the well depth, wherein each second sample point in the second data set corresponds to the geomechanical strength of different well depths and different horizontal well depths;

[0141] constructing a geomechanical similarity term based on the geomechanical strength of the second sample point and the average geomechanical strength of all second sample points;

[0142] Construct a segment length penalty term based on the well depth range of the fracturing segment and the preset segment length constraint;

[0143] According to the geomechanical similarity term and the segment length penalty term, the objective function for dividing the fracturing segments is constructed;

[0144] A dynamic programming algorithm is used to solve the objective function and determine the segmentation points of the fracturing stage;

[0145] The fracturing segments are divided according to the segmentation points to obtain a fracturing segment division result, wherein the fracturing segment division result includes a segment label.

[0146] Specifically, the geomechanical strength and well depth along the horizontal well can be collected first to construct or form a second data set. Among them, x I The well depth and geomechanical strength information corresponding to the second sample point I. Determine the preset segment length constraint, such as setting the target segment length interval [a, b].

[0147] Then according to the well depth set S in the Qth fracturing section Q The geomechanical strength X of the second sample point I in I and the average geomechanical strength m of all second sample points in the Qth fracturing stage Q , the geomechanical similarity term is constructed according to the following formula, that is, for the Q-th fracturing stage, its internal sample X I The mean square error can be expressed as:

[0148]

[0149] Among them, the geomechanical similarity term can be used to measure the difference in geomechanical strength between well sections within the same fracturing section. This application uses the sum of squared errors as a measurement indicator.

[0150] Then according to the well depth range L of the Qth fracturing section Q and preset segment length constraint L target (i.e., the target segment length interval, such as [a, b]), construct the segment length penalty term Penalty (L Q ):

[0151] Penalty(L Q )=(L Q -L target ) 2

[0152] To ensure that the well depth range of each fracturing stage falls within the target stage length interval [a, b] (this interval can be determined based on the geological and engineering characteristics of different blocks), the above-mentioned stage length penalty term can be introduced when there is a deviation between the actual stage length and the target stage length (such as the median of the interval).

[0153] According to the geomechanical similarity term and the segment length penalty term Penalty(L Q ), and construct the objective function for dividing the fracturing stages according to the following formula:

[0154]

[0155] Among them, M is the objective function for dividing the fracturing stages; K is the total number of fracturing stages (i.e. the total number of clusters); S Q is the well depth set in the Qth fracturing section; X I is the geomechanical strength of the second sample point I; m Q is the average geomechanical strength of all second sample points in the Qth fracturing stage; λ is the weight factor that regulates the balance between the geomechanical similarity term and the segment length penalty term; L Q is the well depth range of the Qth fracturing stage.

[0156] Afterwards, the dynamic programming algorithm (DP algorithm) can be used to solve the objective function M to determine the segmentation points of the fracturing section. The fracturing section can then be divided according to the segmentation points to obtain the fracturing section division results. The specific steps are as follows:

[0157] 1) State definition: Let D I It represents the optimal objective function value of the first I second sample points (sample index is 0 to I-1), and records the best segmentation point seg[I], that is, the index of the end of the previous segment.

[0158] 2) State transition: For each second sample point I, only consider the candidate starting point G so that the depth difference = Depth[I-1]-Depth[G] falls within the target segment length interval [a,b], and calculate the objective function value M(G,I) within the interval [G,I). The state transition equation is:

[0159]

[0160] 3) Global optimal solution and backtracking: After recursively calculating D(n) using dynamic programming, the optimal segment boundary index is obtained through backtracking to determine the start and end positions of each fracturing segment, thereby realizing intelligent division of fracturing segments.

[0161] For example, assume the following well depth and geomechanical strength data (constituting the second data set):

[0162]

[0163] The target segment length interval is [a, b] = [150, 250], and the weight factor λ = 1.

[0164] Step 1. Initialization:

[0165] D(0)=0

[0166] seg[0]=-1

[0167] Step 2: State transfer:

[0168] For each second sample point I from 1 to 9, calculate D I :

[0169] For I=1:

[0170] Candidate starting point G = 0

[0171] Calculate M(0,1)=40000;

[0172] D(1)=D(0)+M(0,1)==0+40000=40000

[0173] seg[1]=0

[0174] For I=2:

[0175] Candidate starting point G = 0, 1

[0176] Calculate M(0,2) and M(1,2);

[0177] Select the smallest:

[0178] D(2)=min(D(0)+M(0,2),D(1)+M(1,2))

[0179] Update seg[2]

[0180] Repeat the above process until I=9.

[0181] Step 3: Backtracking: The optimal segment boundary index is obtained through backtracking, thereby determining the start and end positions (segmentation points) of each fracturing segment, thereby dividing the fracturing segment. The fracturing segment division result may include segment labels.

[0182] The above-mentioned dynamic programming algorithm is an effective method for solving sequential optimization problems, especially for seeking the global optimal segmentation in a non-convex solution space. The dynamic programming algorithm divides the second data set (including well depth and geomechanical strength along the horizontal well depth) into K segments and finds the optimal segment division solution by minimizing the objective function. Its goal is to divide the wellbore into fracturing segments, ensuring that the geomechanical properties within the fracturing segments are as similar as possible while meeting the predetermined segment length requirements. By adopting the dynamic programming algorithm, well segments with similar geomechanical strength can be divided into the same fracturing segment. This method integrates well depth data and geomechanical strength indicators while maintaining the spatial continuity of the reservoir interval. It uses the similar geomechanical strength within the segment as the objective function and imposes a segment length constraint penalty term according to the construction requirements, realizing the intelligent division of the fracturing segments and providing support for the subsequent intelligent optimization of the perforation cluster.

[0183] S103: In the divided fracturing stages, a multi-objective function constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results, and the preset cluster spacing constraint is solved, and completion optimization design is performed based on the target solution results of the multi-objective function.

[0184] In some embodiments, before the above S103, in specific implementation, the following steps may also be included:

[0185] Determine the weight of the geological dessert according to the geological dessert category label in the geological dessert evaluation result;

[0186] According to the segment labels in the fracturing segment division results, the fracturing segments are identified and constraints are imposed on the fracturing clusters within the fracturing segments, so that the constrained fracturing clusters are located within a preset segment depth range;

[0187] Generating candidate regions according to the geomechanical strength along the depth of the horizontal well, and generating candidate clusters from the candidate regions;

[0188] The first goal is to make the sum of the weights of the geological sweet spots greater than the preset weight threshold, the second goal is to make the geomechanical similarity term less than the preset variance threshold, and the preset cluster spacing requirement is used as the constraint condition;

[0189] A multi-objective function is constructed based on the first objective, the second objective and the constraints.

[0190] Specifically, the geological sweet spot weight W can be determined according to the geological sweet spot category label using the following formula:

[0191] W i =5-Class i

[0192] When i = 1, it represents label I (Class I), with a maximum weight of W1 = 4. When i = 4, it represents label IV (Class IV), with a maximum weight of W4 = 1. These geological sweet spot weights are used to construct a multi-objective function, enabling multi-objective optimization to select perforation or fracturing clusters.

[0193] Then, you can use segment labels to identify fracture segments and impose constraints on fracture clusters within each fracture segment, such that all fracture clusters within a fracture segment must lie within a preset segment depth range:

[0194]

[0195] in, is the well depth coordinate of the Jth cluster in the nth fracturing stage.

[0196] Then, the engineering sweet spot labels (geomechanical strength along the horizontal well depth) can be used to generate candidate areas. Unlike the traditional fixed threshold method, the candidate areas in this application are dynamically defined: the second sample point within the range of ±10% of the engineering sweet spot mean m is determined as a high-quality engineering sweet spot area P Eng , and is the input sample set used for subsequent clustering:

[0197] P Eng ∈[0.9m n ,1.1m n ]

[0198] Then, the DBSCAN clustering algorithm can be used to further refine and select reasonable initial cluster centers within the candidate area (i.e., generate candidate clusters from the candidate area), thereby creating better convergence conditions for subsequent multi-objective optimization. DBSCAN can automatically identify clusters based on data density without defining the number of clusters. DBSCAN can automatically identify high-density areas, ensuring that candidate clusters do not span well sections with drastic changes in geomechanical properties. Specifically, the steps for generating candidate clusters using the DBSCAN clustering algorithm are as follows:

[0199] Step 1: Data standardization:

[0200] In the candidate region, the two-dimensional data composed of Depth and Pgeo are standardized so that DBSCAN can perform clustering on an unbiased scale.

[0201] The hyperparameters that need to be set include the neighborhood radius (eps) and the minimum number of samples (min_samples). eps (neighborhood radius) defines the density range of a point. This value can be adjusted based on the data distribution, typically ranging from 0.1 to 0.5; in this paper, it is set to 0.3. min_samples (minimum number of samples) sets the minimum number of sample points a cluster must contain to ensure that only sufficiently dense clusters are formed. For candidate regions with fewer data points (<1000), this value typically ranges from 3 to 5; in this paper, it is set to 5.

[0202] Step 2: Clustering and noise filtering:

[0203] DBSCAN identifies the density-reachable relationship of data points based on eps and min_samples, classifies interconnected high-density areas into the same cluster, and marks low-density areas as noise points (label = -1), filtering out noise points that do not belong to any dense area.

[0204] Step 3: Generation of candidate cluster centers:

[0205] A set of candidate cluster centers is generated, representing dense regions of high-quality engineering sweet spot data that meet predefined engineering criteria.

[0206] By selecting cluster centers (generating candidate clusters), a well-constrained and physically meaningful initial solution space can be provided for subsequent multi-objective optimization.

[0207] Then, the sum of the weights of the geological sweet spots can be greater than the preset weight threshold as the first goal (i.e., maximizing the geological sweet spot score) to ensure that the selected cluster center corresponds to the area with favorable reservoir conditions. Specifically, the construction formula of the first goal f1 is as follows:

[0208]

[0209] Among them, W i is the geological sweet spot weight of the first sample point i; S J is the first sample point set of the J-th cluster.

[0210] The second objective (i.e., minimizing the geomechanical strength variance) can be set as the geomechanical similarity term being less than a preset variance threshold. This objective minimizes the geomechanical strength differences between clusters to ensure that the clusters exhibit similar mechanical properties. Specifically, the formula for constructing the second objective f2 is as follows:

[0211]

[0212] Among them, X I is the geomechanical strength of the second sample point I; m Qis the average geomechanical strength of all second sample points in the Qth fracturing stage; S Q is the set of well depths within the Qth fracturing stage.

[0213] The preset cluster spacing requirement can be used as a constraint. To prevent the selected clusters from being too close or too far apart, adjacent clusters must maintain a spacing of 10m-30m to reduce fracture interference and stimulate production efficiency. Specifically, when the depth difference is less than 10m or greater than 30m, a fixed penalty is applied as follows:

[0214] 10m￡|c J -c J+1 |£30m

[0215] Among them, c J is the well depth of the J-th cluster.

[0216] Finally, a multi-objective function can be constructed based on the first objective, the second objective and the constraints.

[0217] In some embodiments, the multi-objective function constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results, and the preset cluster spacing constraint in the above S103 may include:

[0218] A multi-objective optimization algorithm is used to solve the multi-objective function and determine the target solution of the multi-objective function from the candidate clusters;

[0219] The above-mentioned S103, performing the well completion optimization design according to the objective solution results of the multi-objective function, may include:

[0220] The target solution result is used as the design scheme of the fracturing cluster within the fracturing stage, and the completion optimization design is carried out according to the design scheme of the fracturing cluster.

[0221] Specifically, a multi-objective optimization algorithm (such as the NSGA-II algorithm) can be used to solve the multi-objective function, and the optimal solution of the multi-objective function (i.e., the target solution) can be determined from the candidate clusters as the target design scheme for the fracturing cluster or perforation cluster. The target design scheme can include the optimal perforation cluster set, which can provide key input for intelligent completion design to ensure the optimal perforation location and number.

[0222] The specific steps of the NSGA-II algorithm are as follows:

[0223] 1) Population encoding and initialization: Candidate solutions are encoded as binary vectors, with each gene corresponding to a candidate cluster (1 for selection and 0 for exclusion). The initial population size is set to 100, and the initialization is a 50% biased random distribution.

[0224] 2) Fitness evaluation: For each individual, the geosweetness score and geomechanical strength variance are calculated by assigning the data to clusters according to the nearest depth. Spatial constraints are integrated into the objective function value.

[0225] 3) Non-dominated sorting and crowding distance: All individuals are sorted using non-dominated sorting to generate a Pareto front. Within each Pareto front, the crowding distance is calculated to maintain population diversity.

[0226] 4) Selection, Crossover, and Mutation: Binary tournament selection with crowding comparison was used. Simulated binary crossover (SBX) and random exponential mutation were applied to generate offspring. The crossover probability was set to 0.7, and the mutation probability was set to 0.3.

[0227] 5) Population update and iteration: Merge the parent population and the child population, and use non-dominated sorting and crowding distance to select the next generation. This process is iterated for 100 generations or until convergence.

[0228] 6) Optimal solution extraction and post-processing: Extract the Pareto optimal set from the final population, select an optimal individual from it, and apply a small random perturbation to the selected cluster to obtain the final optimized cluster set.

[0229] Since the optimization of perforation clusters requires comprehensive consideration of geological sweet spots, engineering sweet spots and construction constraints, it is a typical nonlinear optimization problem, and it is necessary to construct reasonable objective functions and constraints. This application first uses the threshold method to obtain ±10% of the mean value of geomechanical strength in the fracturing section, and obtains candidate clusters based on the threshold and the collar position. Then, the DBSCAN density clustering algorithm is used to generate candidate cluster centers. As a density-based clustering method, DBSCAN does not require the predefined number of clusters, and autonomously identifies high-density areas, thereby preventing candidate clusters from crossing well sections with significant changes in geomechanical strength, and finally forming candidate clusters. The NSGA-Ⅱ multi-objective optimization algorithm is then used to construct a multi-objective function with the minimum difference in geomechanical strength of the preferred clusters within the section (engineering sweet spot angle), the maximum geological sweet spot score, and the penalty term for the construction cluster spacing requirement. The Pareto frontier search is used in the candidate clusters to generate the optimal set, which is the optimal perforation cluster set. Finally, a specific completion plan can be formulated based on the optimized perforation cluster design plan to ensure the maximization of fracturing and mining effects.

[0230] For example, suppose a fracturing section contains the following 10 sample points:

[0231]

[0232] First, perform data processing:

[0233] 1. Geological sweet spot weight distribution:

[0234] Classi =Class1=Ⅰ(weight 4), Ⅱ(weight 3), Ⅲ(weight 2), Ⅳ(weight 1).

[0235] For example: the weight of sample 1 is W1=4, and the weight of sample 4 is W4=1

[0236] 2. Intra-segment constraints:

[0237] All clusters must be within the depth range of segment 1 (2000m-2050m).

[0238] 3. Generation of engineering sweet spot candidate areas:

[0239] The calculated mean value of the engineering sweet spot strength is m = 31.9 MPa, and the dynamic range is 31.9 ± 3.19 MPa (i.e., 28.7 to 35.1 MPa);

[0240] Filter the samples that meet the conditions: Samples 1 (35), 2 (32), 3 (30), 5 (36), 7 (33), 8 (34), 9 (37).

[0241] Candidate area samples: samples 1, 2, 3, 5, 7, 8, 9.

[0242] Then, the candidate clusters are initialized based on DBSCAN:

[0243] 1. Data standardization:

[0244] The well depth (Depth) and geological sweet spot weight (Pgeo) are standardized to eliminate the dimension effect.

[0245] 2. DBSCAN parameter settings:

[0246] eps=0.3 (neighborhood radius), min_samples=5 (minimum number of samples).

[0247] 3. Clustering results:

[0248] Cluster 1: samples 1 (2000m), 2 (2005m), 3 (2010m), 5 (2020m), 7 (2030m), 8 (2035m);

[0249] Cluster 2: Sample 9 (2040m);

[0250] Noise point: Sample 5 (2020m) was filtered due to insufficient density (hypothesis).

[0251] 4. Extraction of candidate cluster centers:

[0252] Cluster 1 center: well depth 2016.7m, average geological sweet spot weight 3.5;

[0253] Cluster 2 center: well depth 2040m, geological sweet spot weight 4.

[0254] Then, the perforation cluster optimization based on NSGA-II is performed:

[0255] 1. Optimization objectives and constraints:

[0256] Objective 1: Minimize the variance of geomechanical strength between clusters;

[0257] Objective 2: Maximize the sum of geological sweet spot scores;

[0258] Constraints: The distance between clusters must be within the range of 10 to 30 meters.

[0259] 2. Set NSGA-II parameters:

[0260] Population size: 100, iteration number: 100;

[0261] Crossover probability: 0.7, mutation probability: 0.3.

[0262] 3. Instance optimization process:

[0263] 1) Population code:

[0264] The candidate clusters are cluster 1 and cluster 2, encoded as binary vectors (e.g. [1,0] indicates that cluster 1 is selected)

[0265] 2) Fitness calculation:

[0266] Scenario 1: Cluster 1 and Cluster 2 are selected (encoded as [1,1]).

[0267] Variance of geomechanical strength: Cluster 1 (30.8-36 MPa), Cluster 2 (37 MPa), the variance is relatively large.

[0268] Geological sweet spot score: Cluster 1 weight sum = 4 + 3 + 4 + 4 = 15, Cluster 2 = 4, total score = 19.

[0269] Constraint penalty: cluster spacing = 2040 - 2016.7 = 23.3m (satisfied, no penalty).

[0270] Scenario 2: Only cluster 1 (encoding [1,0]) is selected.

[0271] Geomechanical strength variance: Variance within cluster 1 = 2.8.

[0272] Geological dessert score = 15.

[0273] No cluster spacing constraint (single cluster only).

[0274] 3) Pareto frontier solution:

[0275] Solution A (select cluster 1 + cluster 2): total score 19, high variance.

[0276] Solution B (select only cluster 1): total score 15, low variance.

[0277] 4) Optimal solution selection:

[0278] If balancing the total score and variance is preferred, choose solution A and apply random perturbations (such as adjusting the center of cluster 1 to 2015m).

[0279] Ultimately, through the collaborative optimization of DBSCAN and NSGA-II, cluster 1 (2000-2035 m) was selected as the optimal perforation cluster. It has a high geological sweet spot score (15), a low geomechanical strength variance (2.8), and meets the intra-segment depth constraint. If multiple clusters are required, the candidate cluster spacing needs to be adjusted or the constraints need to be relaxed.

[0280] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. For details, please refer to the description of the aforementioned related processing embodiments, and no further description is given here.

[0281] The above describes the present invention. However, it is worth noting that this specific embodiment is only intended to better illustrate the present application and to describe specific embodiments of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0282] In a specific implementation scenario, see Figure 2 As shown in Figure 2, the process of the intelligent completion optimization design method based on multi-objective optimization can be as follows:

[0283] 1. Data Processing

[0284] Collect readily available and valuable data from the drilling, completion, and production processes, such as well logging data (0.125 m), drilling data (1 m), and post-fracturing monitoring data. Exploratory Data Analysis (EDA) is used to understand the structure, distribution, and quality of logging and drilling data through statistical methods and visualization. The logging and drilling data are then preprocessed, including unified metrics, outlier processing, and missing value filling. Reservoir physical and geomechanical parameters are calculated based on the preprocessed drilling and / or logging data. The reservoir physical and geomechanical parameters are then smoothed and de-noised to form a single-well dataset. Post-fracturing monitoring data can include tracer monitoring and fiber optic monitoring. Tracer testing can obtain oil production contribution rates / segments (i.e., oil production profile data), while fiber optic monitoring can obtain fluid distribution amounts / clusters, which can be used for monitoring interpretation.

[0285] 2. Model and Algorithm

[0286] First, correlation analysis is performed between each reservoir property parameter and oil production profile data, and the Spearman correlation coefficient is calculated to determine the weights of each reservoir property parameter. A weighted Euclidean distance function, a criterion function, and a ranking function are then constructed based on the weights of each reservoir property parameter. Furthermore, based on the weighted Euclidean distance function, the criterion function, and the ranking function, a mechanism-guided weighted K-medoids clustering model or algorithm is constructed. The total number of clusters, K, is determined based on the "elbow rule." Clustering is performed using the weighted Euclidean distance function and the criterion function, and the ranking function is used to evaluate the clustered geological sweet spots in the first clustering result to obtain a geological sweet spot evaluation result.

[0287] The bottom hole mechanical specific energy MSE b and the minimum horizontal principal stress S hmin Determine the geomechanical strength and then evaluate the engineering sweet spot. An objective function for segmenting fracturing stages can be constructed based on the geomechanical similarity term (similar geomechanical strength) and the segment length penalty term (fracture segment length limit penalty). A dynamic programming (DP) algorithm is used to solve the objective function for segmenting fracturing stages, performing a global optimization search to determine the start and end locations (segmentation points) of each fracturing stage. This allows for segmentation and optimization.

[0288] After segmenting the fracturing stages, a threshold method is used to determine ±10% of the mean geomechanical strength within the fracturing stages. Based on the threshold and collar location, candidate clusters are generated. The DBCAN density clustering algorithm is then used to generate candidate cluster centers, ultimately forming an initial candidate cluster or candidate clusters. A multi-objective function is constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results, and the preset cluster spacing requirements (cluster spacing constraints). The NSGA-II multi-objective optimization algorithm is used to further select the optimal set of perforation clusters from the candidate clusters, while satisfying physical constraints. This allows for optimized perforation cluster design, providing key input for intelligent completion design and ensuring optimal perforation location and quantity.

[0289] By using correlation analysis to quantify the relationship between reservoir physical properties and oil production profile data, a mechanism-guided weighted Euclidean clustering was constructed, and a mechanism-guided weighted clustering model was established, which enables sweet spot evaluation that is more consistent with geological principles.

[0290] By constructing the objective function for dividing fracturing stages based on the geomechanical similarity term and the segment length penalty term, it not only considers the geomechanical strength but also integrates the constraints of fracturing segment length and spatial continuity. This enables the DP algorithm to automatically generate a reasonable fracturing stage design without human intervention.

[0291] Based on the threshold method and DBSCAN density clustering to generate candidate clusters, a multi-objective optimization algorithm is adopted, incorporating the geological sweet spot, engineering sweet spot, and perforation cluster spacing into its objective function and penalty terms. This not only ensures that the candidate clusters have similar geomechanical conditions but also maximizes the quality of the geological sweet spot, reduces inter-fracture interference and collar damage, and thus improves the effectiveness of the transformation.

[0292] Although this specification provides examples such as the following examples or the accompanying Figure 3 The method operation steps or device structure shown, but based on routine or no creative labor, the method or device may include more or fewer operation steps or module units after partial merger. In the steps or structures where there is no necessary causal relationship logically, the execution order of these steps or the module structure of the device is not limited to the execution order or module structure shown in the embodiments or drawings of this specification. When the method or module structure described is applied to actual devices, servers or terminal products, it can be executed sequentially or in parallel according to the method or module structure shown in the embodiments or drawings (for example, in an environment of parallel processors or multi-threaded processing, or even in an implementation environment of distributed processing and server clusters). Based on the above-mentioned intelligent completion optimization design method based on multi-objective optimization, the embodiments of this specification also propose an embodiment of an intelligent completion optimization design device based on multi-objective optimization. As Figure 3 As shown, the device may specifically include the following modules:

[0293] The geological sweet spot evaluation module 301 can be used to cluster the first data set according to a weighted clustering model and evaluate the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result. The first data set includes various reservoir physical property parameters along the depth of the horizontal well. The weighted clustering model is constructed based on the weights of the various reservoir physical property parameters.

[0294] The fracturing stage division module 302 may be used to divide the fracturing stages according to the engineering sweet spot evaluation results determined based on geomechanical parameters and preset stage length constraints;

[0295] The completion optimization design module 303 can be used to solve a multi-objective function constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results and the preset cluster spacing constraints within the divided fracturing stages, and perform completion optimization design based on the target solution results of the multi-objective function.

[0296] In some embodiments, before the above-mentioned geological sweet spot evaluation module 301, it can be specifically used to calculate the correlation coefficient between each reservoir physical property parameter and the oil production profile data; determine the sum of the absolute values ​​of the correlation coefficients between each reservoir physical property parameter and the oil production profile data; and determine the weight of each reservoir physical property parameter based on the correlation coefficient between the reservoir physical property parameter and the oil production profile data and the sum of the absolute values ​​of the correlation coefficients.

[0297] In some embodiments, the above-mentioned geological sweet spot evaluation module 301 can be specifically used to construct a weighted Euclidean distance function, a criterion function and a rating function according to the weights of each reservoir physical property parameter; and to construct a weighted clustering model according to the weighted Euclidean distance function, the criterion function and the rating function.

[0298] In some embodiments, the above-mentioned geological sweet spot evaluation module 301 can also be specifically used to select a preset number of cluster centers from the first data set; according to the weighted Euclidean distance function, calculate the weighted Euclidean distance from each first sample point in the first data set to the cluster center, and each first sample point corresponds to each reservoir physical property parameter at different horizontal well depths; compare the weighted Euclidean distance, and according to the comparison result, divide each first sample point into a cluster cluster corresponding to a weighted Euclidean distance less than a preset distance threshold; according to the criterion function, select a new cluster center from the cluster cluster; repeat the process of calculating the weighted Euclidean distance, dividing each first sample point, and selecting a new cluster center until the number of iterations reaches the preset iteration number threshold or the new cluster center selected does not change, and obtain the first clustering result after clustering; use the rating function to evaluate the clustered geological sweet spots in the first clustering result to obtain a geological sweet spot evaluation result.

[0299] In some embodiments, the above-mentioned geological sweet spot evaluation module 301 can also be specifically used to use a rating function to determine the weighted geological sweet spot index of each cluster in the first clustering result; compare the weighted geological sweet spot index of each cluster, and sort the comparison results of the weighted geological sweet spot index in descending order to obtain the sorting results of the weighted geological sweet spot index; according to the sorting results of the weighted geological sweet spot index, the corresponding geological sweet spot is classified into different levels to obtain a geological sweet spot evaluation result, and the geological sweet spot evaluation result includes a geological sweet spot category label.

[0300] In some embodiments, the above-mentioned engineering sweet spot evaluation results include the geomechanical strength along the depth of the horizontal well; accordingly, the above-mentioned fracturing segment division module 302 can be specifically used to construct a second data set based on the geomechanical strength along the depth of the horizontal well and the well depth, and each second sample point in the second data set corresponds to the geomechanical strength of different well depths and different horizontal well depths; based on the geomechanical strength of the second sample point and the average geomechanical strength of all second sample points, a geomechanical similarity term is constructed; based on the well depth range of the fracturing segment and the preset segment length constraint, a segment length penalty term is constructed; based on the geomechanical similarity term and the segment length penalty term, an objective function for dividing the fracturing segment is constructed; a dynamic programming algorithm is used to solve the objective function to determine the segmentation points of the fracturing segment; the fracturing segment is divided according to the segmentation points to obtain a fracturing segment division result, and the fracturing segment division result includes a segment label.

[0301] In some embodiments, the above-mentioned completion optimization design module 303 can be used to determine the geological sweet spot weight based on the geological sweet spot category label in the geological sweet spot evaluation result; identify the fracturing segment based on the segment label in the fracturing segment division result and impose constraints on the fracturing cluster within the fracturing segment, so that the fracturing cluster after imposing constraints is located within a preset segment depth range; generate candidate areas based on the geomechanical strength along the horizontal well depth, and generate candidate clusters from the candidate areas; set the sum of the geological sweet spot weights greater than the preset weight threshold as the first goal, set the geomechanical similarity term less than the preset variance threshold as the second goal, and set the preset cluster spacing requirement as the constraint condition; construct a multi-objective function based on the first goal, the second goal and the constraint condition.

[0302] In some embodiments, the completion optimization design module 303 can be specifically used to adopt a multi-objective optimization algorithm to solve a multi-objective function, and determine the target solution result of the multi-objective function from the candidate cluster; use the target solution result as the target design scheme of the fracturing cluster in the fracturing section, and perform completion optimization design according to the target design scheme of the fracturing cluster.

[0303] As can be seen from the above, the intelligent completion optimization design device based on multi-objective optimization provided in the embodiments of this specification can realize the automated and intelligent design of fracturing sections and perforation cluster positions, thereby providing a new technical approach for the optimized completion design of unconventional oil and gas reservoirs, and has important engineering applicability.

[0304] An embodiment of this specification also provides an electronic device based on the above-mentioned intelligent completion optimization design method based on multi-objective optimization, including a processor and a memory for storing processor executable programs / instructions. When the processor is specifically implemented, it can perform the following steps according to the program / instructions: clustering a first data set according to a weighted clustering model, and evaluating the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes reservoir physical property parameters along the depth of the horizontal well, and the weighted clustering model is constructed according to the weight of each reservoir physical property parameter; dividing the fracturing section according to the engineering sweet spot evaluation result and the preset section length constraint, and the engineering sweet spot evaluation result is determined according to the geomechanical parameters; within the divided fracturing section, solving the multi-objective function constructed according to the geological sweet spot evaluation result, the engineering sweet spot evaluation result and the preset cluster spacing constraint, and performing completion optimization design according to the target solution result of the multi-objective function.

[0305] In order to complete the above instructions more accurately, refer to Figure 4 As shown, the embodiment of this specification also provides another specific electronic device, wherein the electronic device includes a network communication port 401, a processor 402 and a memory 403, and the above structures are connected through internal cables so that each structure can perform specific data interaction.

[0306] Among them, the processor 402 can be specifically used to cluster the first data set according to the weighted clustering model, and evaluate the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes various reservoir physical property parameters along the depth of the horizontal well, and the weighted clustering model is constructed according to the weights of each reservoir physical property parameter; the fracturing section is divided according to the engineering sweet spot evaluation result and the preset section length constraint, and the engineering sweet spot evaluation result is determined according to the geomechanical parameters; within the divided fracturing section, a multi-objective function constructed according to the geological sweet spot evaluation result, the engineering sweet spot evaluation result and the preset cluster spacing constraint is solved, and completion optimization design is performed according to the target solution result of the multi-objective function.

[0307] The memory 403 may be specifically used to store corresponding instruction programs.

[0308] In this embodiment, the network communication port 401 can be a virtual port that is bound to different communication protocols, thereby being capable of sending or receiving different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0309] In this embodiment, the processor 402 may be implemented in any suitable manner. For example, the processor may take the form of a microprocessor or a processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, an embedded microcontroller, etc. This specification is not intended to limit this.

[0310] In this embodiment, the memory 403 may include multiple levels. In a digital system, anything that can store binary data can be a memory. In an integrated circuit, a circuit with a storage function that has no physical form is also called a memory, such as RAM, FIFO, etc. In a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0311] An embodiment of this specification also provides a computer storage medium based on the above-mentioned intelligent completion optimization design method based on multi-objective optimization, wherein the computer storage medium stores a computer program / instruction, which, when executed, implements: clustering a first data set according to a weighted clustering model, and evaluating the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes reservoir physical property parameters along the depth of the horizontal well, and the weighted clustering model is constructed according to the weights of each reservoir physical property parameter; dividing the fracturing section according to the engineering sweet spot evaluation result and the preset section length constraint, wherein the engineering sweet spot evaluation result is determined according to the geomechanical parameters; within the divided fracturing section, solving a multi-objective function constructed according to the geological sweet spot evaluation result, the engineering sweet spot evaluation result and the preset cluster spacing constraint, and performing completion optimization design according to the target solution result of the multi-objective function.

[0312] In this embodiment, the storage medium includes, but is not limited to, random access memory (RAM), read-only memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured in accordance with the standards specified by the communication protocol for network connection communication.

[0313] In this embodiment, the functions and effects specifically implemented by the program instructions stored in the computer storage medium can be explained in comparison with other implementations and will not be repeated here.

[0314] Although this specification provides the method operation steps as described in the embodiments or flow charts, more or fewer operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiments is only one way of executing the order of many steps and does not represent the only execution order. When the device or client product in practice is executed, it can be executed in sequence or in parallel according to the method shown in the embodiments or the drawings (for example, a parallel processor or a multi-threaded processing environment, or even a distributed data processing environment). The term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, product or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, product or device. In the absence of more restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or device including the elements. Words such as first and second are used to represent names and do not represent any particular order.

[0315] Those skilled in the art will also appreciate that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to implement the same functionality by logically programming the method steps in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be considered structures within the hardware component. Alternatively, the devices for implementing various functions can be considered both software modules implementing the method and structures within the hardware component.

[0316] This specification may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, classes, and the like that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0317] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that this specification can be implemented by means of software plus the necessary general hardware platform. Based on this understanding, the technical solution of this specification can essentially be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of this specification.

[0318] The various embodiments in this specification are described in a progressive manner. References to the common or similar parts of the various embodiments are sufficient. Each embodiment focuses on the differences from the other embodiments. This specification can be used in a variety of general-purpose or specialized computer system environments or configurations. For example, personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above systems or devices.

[0319] Although the present specification has been described with reference to the embodiments, persons skilled in the art will appreciate that there are many variations to the present specification without departing from the spirit of the present specification, and it is intended that the appended claims encompass such variations without departing from the spirit of the present specification.

Claims

1. An intelligent well completion optimization design method based on multi-objective optimization, characterized in that: include: Clustering a first data set according to a weighted clustering model, and evaluating a geological sweet spot of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes reservoir physical property parameters along a horizontal well depth, and the weighted clustering model is constructed according to a weight of each reservoir physical property parameter; Dividing the fracturing stages according to the engineering sweet spot evaluation results and the preset stage length constraints, wherein the engineering sweet spot evaluation results are determined according to geomechanical parameters; Within the divided fracturing sections, a multi-objective function constructed based on the geological sweet spot evaluation results, engineering sweet spot evaluation results, and preset cluster spacing constraints is solved, and completion optimization design is performed based on the target solution results of the multi-objective function.

2. The method according to claim 1, characterized in that The method further comprises: Calculate the correlation coefficient between each reservoir physical property parameter and oil production profile data; Determine the sum of the absolute values ​​of the correlation coefficients between each reservoir physical property parameter and the oil production profile data; The weight of each reservoir physical property parameter is determined based on the correlation coefficient between the reservoir physical property parameters and the oil production profile data and the sum of the absolute values ​​of the correlation coefficients.

3. The method according to claim 1, characterized in that The weighted clustering model is constructed based on the weights of various reservoir physical property parameters, including: According to the weights of the reservoir physical parameters, weighted Euclidean distance function, criterion function and rating function are constructed; A weighted clustering model is constructed based on the weighted Euclidean distance function, criterion function and rating function.

4. The method according to claim 3, characterized in that Clustering the first data set according to the weighted clustering model includes: selecting a preset number of cluster centers from the first data set; Calculating the weighted Euclidean distance from each first sample point in the first data set to the cluster center according to the weighted Euclidean distance function, wherein each first sample point corresponds to each reservoir physical property parameter at different horizontal well depths; Comparing the weighted Euclidean distances, and dividing each first sample point into a cluster corresponding to a weighted Euclidean distance less than a preset distance threshold according to the comparison result; According to the criterion function, a new cluster center is selected from the cluster clusters; Repeat the process of calculating the weighted Euclidean distance, dividing each first sample point, and selecting a new cluster center until the number of iterations reaches a preset iteration threshold or the new cluster center is unchanged, thereby obtaining a first clustering result after clustering; The geological sweet spot of each cluster after clustering is evaluated to obtain a geological sweet spot evaluation result, including: The clustered geological sweet spots in the first clustering result are evaluated using a rating function to obtain a geological sweet spot evaluation result.

5. The method according to claim 4, characterized in that The use of the rating function to evaluate the clustered geological sweet spots in the first clustering result to obtain a geological sweet spot evaluation result includes: Using the rating function, determine the weighted geological sweet spot index of each cluster in the first clustering result; Compare the weighted geological sweet spot indexes of each cluster, sort the comparison results of the weighted geological sweet spot indexes in descending order, and obtain the sorting results of the weighted geological sweet spot indexes; According to the ranking result of the weighted geological dessert index, the corresponding geological dessert is graded into different levels to obtain a geological dessert evaluation result, which includes a geological dessert category label.

6. The method according to claim 1, wherein The engineering sweet spot evaluation results include geomechanical strength along the depth of the horizontal well. Accordingly, the fracturing stages are divided according to the engineering sweet spot evaluation results and the preset stage length constraints, including: constructing a second data set based on the geomechanical strength along the horizontal well depth and the well depth, wherein each second sample point in the second data set corresponds to the geomechanical strength of different well depths and different horizontal well depths; constructing a geomechanical similarity term based on the geomechanical strength of the second sample point and the average geomechanical strength of all second sample points; Construct a segment length penalty term based on the well depth range of the fracturing segment and the preset segment length constraint; According to the geomechanical similarity term and the segment length penalty term, the objective function for dividing the fracturing segments is constructed; A dynamic programming algorithm is used to solve the objective function and determine the segmentation points of the fracturing stage; The fracturing segments are divided according to the segmentation points to obtain a fracturing segment division result, wherein the fracturing segment division result includes a segment label.

7. The method according to claim 1, characterized in that The method further comprises: Determine the weight of the geological dessert according to the geological dessert category label in the geological dessert evaluation result; According to the segment labels in the fracturing segment division results, the fracturing segments are identified and constraints are imposed on the fracturing clusters within the fracturing segments, so that the constrained fracturing clusters are located within a preset segment depth range; Generating candidate regions according to the geomechanical strength along the depth of the horizontal well, and generating candidate clusters from the candidate regions; The first goal is to make the sum of the weights of the geological sweet spots greater than the preset weight threshold, the second goal is to make the geomechanical similarity term less than the preset variance threshold, and the preset cluster spacing requirement is used as the constraint condition; A multi-objective function is constructed based on the first objective, the second objective and the constraints.

8. The method according to claim 7, characterized in that The solution is a multi-objective function constructed based on the geological sweet spot evaluation results, the engineering sweet spot evaluation results and the preset cluster spacing constraint, including: A multi-objective optimization algorithm is used to solve the multi-objective function and determine the target solution of the multi-objective function from the candidate clusters; The well completion optimization design is performed according to the objective solution results of the multi-objective function, including: The target solution result is used as the target design scheme of the fracturing cluster within the fracturing stage, and the completion optimization design is carried out according to the target design scheme of the fracturing cluster.

9. An intelligent well completion optimization design device based on multi-objective optimization, characterized in that: include: a geological sweet spot evaluation module, configured to cluster a first data set according to a weighted clustering model, and evaluate the geological sweet spots of each cluster after clustering to obtain a geological sweet spot evaluation result, wherein the first data set includes reservoir physical property parameters along the depth of the horizontal well, and the weighted clustering model is constructed based on the weights of the reservoir physical property parameters; A fracturing segment division module, configured to divide the fracturing segments according to the engineering sweet spot evaluation results determined based on geomechanical parameters and preset segment length constraints; The completion optimization design module is used to solve the multi-objective function constructed based on the geological sweet spot evaluation results, engineering sweet spot evaluation results and preset cluster spacing constraints within the divided fracturing stages, and perform completion optimization design based on the target solution results of the multi-objective function.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images