Coal seam hydraulic fracturing parameter dynamic optimization method based on deep learning

Abstract

The invention provides a coal seam hydraulic fracturing parameter dynamic optimization method based on deep learning, and belongs to the technical field of coal seam parameter optimization based on computer data processing. The method comprises the following steps: firstly, acquiring a coal seam slice CT image with spatial continuity; secondly, designing a coal seam structure-physical parameter intelligent sensing model, and performing joint prediction on coal seam image semantic recognition and physical property estimation; then, performing coal seam three-dimensional structure reconstruction and regional layering modeling by utilizing a prediction output result, and performing finite element mesh division and material attribute accurate mapping on the basis to construct a coal seam three-dimensional finite element model; and finally, a coal seam hydraulic fracturing parameter multi-objective optimization problem is constructed, a coal seam physical information embedded agent model is designed, the optimization problem is dynamically solved in combination with a multi-objective evolutionary algorithm, and optimal hydraulic fracturing parameters are obtained. On the basis of deep learning, multi-source information, structural parameters and material attributes are fused, and intelligent development of coal seam hydraulic fracturing design is promoted.

Description

Technical Field

[0001] The present invention belongs to the technical field of coal seam parameter optimization based on computer data processing, and in particular relates to a method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning. Background Art

[0002] With the continuous advancement of coalbed methane resource development, coalbed hydraulic fracturing, as a key means to improve coalbed permeability and enhance coalbed methane production capacity, has become one of the core technologies in coal mining and coalbed methane development. Hydraulic fracturing forms a fracture network in the coal seam by injecting fracture fluid at high pressure, thereby improving the conductivity of the coal seam. The setting of fracturing parameters, such as injection pressure, displacement, proppant concentration and injection time, directly determines the fracturing effect and economic benefits. However, due to the complex structure of coal seams and the changeable physical parameters, traditional fracturing design usually relies on experience or simplified models, which is difficult to dynamically respond to on-site geological changes, affecting the fracture development effect and gas production stability. In recent years, the development of artificial intelligence and deep learning technology has provided new possibilities for the intelligent optimization of coalbed hydraulic fracturing parameters. Through in-depth analysis of on-site images and material properties, more accurate prediction and control of coal seam fracturing behavior can be achieved, providing a basis for formulating more scientific fracturing parameters.

[0003] The existing methods for formulating coal seam hydraulic fracturing parameters mainly include the following three methods: Fracturing design method based on field experience and static geological parameters: This method is usually used by engineers to manually formulate fracturing parameter plans based on geological exploration data (such as coal seam thickness, ground stress, fracture density, etc.) and past experience. Common processes include geological parameter surveys, empirical template selection, manual parameter adjustment, etc. This type of method is relatively common in early coal seam fracturing projects and has certain practicality. However, this method relies heavily on the experience of construction personnel and is difficult to adapt to complex or atypical geological conditions. In areas with strong coal seam structural heterogeneity, this method often cannot accurately reflect the impact of microstructure on fracturing response, and there are problems such as large deviations in parameter settings and unstable fracturing effects; Numerical simulation method based on single physical property input: This method constructs a finite element model through structural parameters such as ground stress, porosity, and permeability, uses fracturing simulation software (such as ABAQUS and FLAC3D) to predict the crack propagation path and impact range, and iteratively optimizes the fracturing parameters based on the simulation. However, model construction relies on simplified expressions of geological parameters, and the input data dimension is low, making it difficult to fully describe the heterogeneity and detailed characteristics of coal seams. In addition, this method usually ignores the influence of actual material morphology (such as crack distribution, interlayers, etc.) on the simulation results, resulting in limited simulation accuracy and a certain deviation in the actual construction guidance role; Fracturing effect prediction method based on machine learning model: This method collects historical fracturing construction data (including fracturing fluid dosage, injection rate, proppant strength, etc.) and corresponding coal seam characteristics, and uses traditional machine learning algorithms (such as decision trees, support vector machines, etc.) to establish a regression or classification model between fracturing parameters and gas production effects, so as to achieve prediction of fracturing effects and parameter adjustment suggestions. However, this type of method is highly dependent on data quality and sample size, and the model training process lacks an explanation of the physical mechanism. In addition, since the model input is mainly based on numerical variables and lacks the ability to model non-numerical features such as images and structures, its generalization performance is poor when facing new well areas or complex structural conditions.

[0004] Therefore, the existing methods have limitations to varying degrees in the process of fracturing parameter optimization, making it difficult to achieve accurate perception and dynamic response to complex coal seam structures. Summary of the invention

[0005] In view of the above problems, the present invention proposes a method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning, comprising the following steps: S1, collect CT images of coal seam slices with spatial continuity; S2, input the CT image into the constructed and trained coal seam structure-physical parameter intelligent perception model, which includes a coal seam multi-scale spatial attention feature encoder and a bedding context perception decoder, which is used to output the coal seam semantic segmentation results and the corresponding material physical parameters; S3, performing regional stratification and meshing according to the semantic segmentation results, and then mapping the corresponding material physical parameters to each unit of the finite element model according to the semantic region to obtain a three-dimensional finite element model of the coal seam; S4, taking maximizing the transformation volume, minimizing the comprehensive cost and maximizing the fracture conductivity as the optimization objectives, and taking the coal seam hydraulic fracturing parameters to be solved as the decision variables, a multi-objective optimization problem of coal seam hydraulic fracturing parameters is established; the multi-objective optimization problem is solved based on the agent-assisted multi-objective evolutionary algorithm to obtain the optimal coal seam hydraulic fracturing parameters; The agent-assisted multi-objective evolutionary algorithm generates training data and test data based on the constructed three-dimensional finite element model of the coal seam to train and test the agent model embedded with the physical information of the coal seam to obtain a global agent model, and combines the global agent model with the multi-objective evolutionary algorithm to dynamically solve the multi-objective optimization problem.

[0006] Preferably, based on the model training task, the coal seam semantic segmentation and material calibration dataset is collected and constructed, and the specific process is as follows: First, for the coal seam to be collected, a slice CT image sequence with spatial continuity is obtained: the coal seam is evenly divided into slices, each set of CT image data contains The slice images collected at equidistant intervals are ,The acquisition process adopts a unified coordinate system calibration; Then The key coal seam geological structures in the slice images are semantically annotated, including 8 categories: background, main coal seam, interlayer gangue, original cracks, faults / joints, mineralized zones, weak interlayers, and aquifers. Semantic segmentation map of slice images ; Finally, based on the semantic annotation, the corresponding material physical parameters of each structural category area are calibrated. The calibrated parameters include elastic modulus , Poisson's ratio , Permeability , tensile strength , Shear Strength , fracture toughness , fracture-matrix coupling ratio , Saturation , Fracturing-induced fracture threshold , Fracturing Priority ;against slice images, a total of Group material physical parameter diagram ,The material physical parameter map contains the material physical parameters corresponding to each coal seam semantic area.

[0007] Preferably, the specific processing process of the layer structure-physical parameter intelligent perception model is: S21, for Each slice of CT image to be segmented ,in ; Firstly, two 3*3 convolution blocks are used to extract preliminary features of CT images. Each 3*3 convolution block consists of a convolution layer with a convolution kernel of 3*3, a batch normalization layer, and a ReLU activation function to extract local texture information and basic spatial features; then, three maximum pooling layers are introduced in turn for multi-scale downsampling to obtain coal seam structural features at different spatial scales, and the feature outputs from shallow to deep are as follows: , corresponding to the front-layer features, middle-layer features and deep-layer features respectively; S22, multi-scale features Input coal seam multi-scale spatial attention feature encoder to realize deep feature extraction of coal seam structure and fusion of bedding context information: For deep features After upsampling, compared with the middle-level features The fusion feature is obtained by fusing through the second mixed feature encoding block ; then continue to Upsample and combine with shallow features Further fusion is performed through the first hybrid feature encoding block to obtain a shallow fusion result ; Shallow fusion results After average pooling, combined with the fusion feature The third hybrid feature encoding block is input together to enhance the perception of mid-scale features and obtain fused features ; then further Perform average pooling and combine with deep features Input to the fourth hybrid feature encoding block and output the fusion result ; The fusion features of three different scales , , Channel splicing is performed to construct a multi-scale feature set; finally, the spatial distribution characteristics of the key structures in the image are enhanced through the spatial attention module, and the final output represents the Layer structure feature map of the slice image ; S23, based on the bedding context-aware decoder, performs image semantic reconstruction by combining the bedding information of adjacent slices: For the Bedding structure characteristics of each slice , combined with the bedding structure characteristics of the two adjacent slices and Perform channel splicing to form context fusion features ; Afterwards, the context fusion feature It is input into a 3*3 deconvolution block for preliminary decoding and upsampling, and a multi-head attention mechanism is introduced to explicitly model and enhance the key structures in the layered context to obtain enhanced features. ; Then, After decoding through four parallel feature reconstruction channels, each channel extracts information from different receptive fields and modeling strategies, and the outputs of the four channels are spliced ​​to obtain the Slice image layer structure feature map The decoder output features ; S24, and finally through the output layer Perform feature compression and map them separately The estimated result of the final semantic segmentation map of the slice CT image , and the estimated results of the material physical parameter map .

[0008] Preferably, the first mixed feature coding block, the second mixed feature coding block, the third mixed feature coding block, and the fourth mixed feature coding block are specifically: For the two sets of input features to be fused, they are first spliced ​​in the channel dimension to form a joint feature representation; then, the joint feature is simultaneously input into three parallel paths to extract spatial and semantic information from different angles: Path 1 adopts a residual connection structure with batch normalization to retain the original feature trend and boundary information; Path 2 first compresses the channel through a 1*1 convolution block, and then extracts local structural features through a maximum pooling operation to enhance the model's response to the key areas of the main coal seam boundary and interbedded gangue blocks; Path 3 performs channel mapping through a 1*1 convolution block, and then introduces the RepVGG layer with deep modeling capabilities to obtain a stronger semantic representation, and finally uses a 3*3 convolution layer to further improve the perception accuracy of spatial bedding changes; finally, the features extracted by the three paths are spliced ​​through the channels to form a fused feature representation.

[0009] Preferably, the Decoding is performed through four parallel feature reconstruction paths, specifically: Among them, the first pathway uses a combination of 3*1 and 1*3 deconvolution blocks to enhance the directional perception of the horizontal and vertical boundary structures of the bedding; the second pathway uses 3*3 and 1*1 deconvolution blocks to fuse local details with global contour information; the third pathway uses a combination of 5*5 and 1*1 deconvolution blocks to expand the receptive field and capture larger-scale bedding change characteristics; the fourth pathway effectively retains the original semantic features and strengthens the deep structural information through the residual connection mechanism.

[0010] Preferably, the specific process of S3 is: S31, 3D structure reconstruction and regional hierarchical modeling: First, the semantic segmentation map of the serial CT slice images is used to estimate the results. Reconstruct the three-dimensional structure; obtain the three-dimensional semantic structure distribution of the coal seam by arranging the equidistant slice images in sequence in a unified coordinate system, construct corresponding structural voxel models for different semantic categories, and perform regional stratification to obtain a hierarchical semantic structure model; S32, Meshing and structural discretization: Based on the hierarchical semantic structural model, the three-dimensional finite element method is used for structural discretization, and an unstructured meshing strategy is adopted for complex geological structural morphology. The unit density is adaptively adjusted in combination with the geometric characteristics of different regions, and finally a discretized finite element model is obtained; S33, Material physical parameter mapping: Mapping the material physical parameter estimation results The various parameters in the model are accurately mapped to the finite element grid units according to the semantic area; the mapping process supports the parameter correspondence at the pixel level, so that the model can carefully reflect the heterogeneity and spatial variability of the physical property distribution inside the coal seam, and obtain the three-dimensional finite element model of the coal seam. , Represents the physical parameters of the material.

[0011] Preferably, the multi-objective optimization problem of coal seam hydraulic fracturing parameters established in S4 is specifically: Determine the decision variables: 9 key parameters were selected as decision variables, including 6 fracturing construction parameters and 3 engineering control parameters; the fracturing construction parameters include displacement , its engineering significance is to control the crack expansion speed; sand ratio , its engineering significance is to affect the proppant delivery efficiency; fracturing fluid viscosity , its engineering significance is to affect the sand carrying capacity and filtration loss; injection time , its engineering significance is to control the duration of crack extension; stage volume , used to control the scale of single-stage fracturing; pumping rate , its engineering significance is to affect the construction continuity; engineering control parameters include perforation cluster spacing , its engineering significance is to control the crack initiation position; proppant particle size , its engineering significance is to balance the flow conductivity and sedimentation; the concentration of temporary plugging agent , its engineering significance is the balance of image multi-cluster expansion; Design objective function: ; in To transform the volume objective function, the objective function is minimized to improve the utilization rate of natural fractures in coal seams. is the transformation volume, which is calculated from the three-dimensional finite element model of the coal seam; is a comprehensive cost objective function, which covers the cost of sand material, fracturing fluid and temporary plugging agent consumption. Minimizing this objective function is used to reduce the cost of coal seam hydraulic fracturing. ， ， and They are fixed cost coefficient of displacement, fixed cost coefficient of sand ratio, fixed cost coefficient of stage volume and fixed cost coefficient of temporary plugging agent concentration; is the fracture conductivity objective function, and minimizing this objective function is used to improve the fracture conductivity of the coal seam after hydraulic fracturing. is the conductivity coefficient, is the fracture permeability, which is calculated by the three-dimensional finite element model of the coal seam.

[0012] Preferably, the global proxy model includes an encoder, a large-scale physical information extraction layer, a small-scale physical information extraction layer, a physical information aggregation layer and a fully connected layer; data Enter the global proxy model to obtain and Prediction value, the specific steps are as follows: Input data into the encoder to extract the primary features of each decision variable data and obtain the primary feature matrix of the decision variable ; The encoder is composed of a convolution layer and a RELU activation function layer, the convolution layer is used to extract the local features of each decision variable data, and the RELU activation function layer is used to perform nonlinear transformation on the local features of each decision variable data; Input data into the large-scale physical information extraction layer to extract the large-scale correlation features of each decision variable data and obtain the large-scale correlation feature matrix The large-scale physical information extraction layer uses a convolution kernel scale of The graph attention neural network is used to extract the large-scale features of each decision variable in the data and calculate the correlation coefficient of the large-scale features of each decision variable to obtain ; Input data into the small-scale physical information extraction layer to extract the small-scale correlation features of each decision variable data and obtain the small-scale correlation feature matrix The large-scale physical information extraction layer uses a convolution kernel scale of The graph attention neural network is used to extract the small-scale features of each decision variable in the data and calculate the correlation coefficient of the small-scale features of each decision variable to obtain ; Will and After adding Input the physical information aggregation layer together to aggregate the graph features and local features to obtain aggregated features The physical information aggregation layer uses a graph aggregation neural network to combine the characteristics of each decision variable in F with and The features of each decision variable in the added feature graph are sampled and aggregated respectively to obtain the aggregated features that characterize the implicit relationship between the decision variable data. ; Will Input the fully connected layer to reduce the dimension and map the features and Predicted value.

[0013] Compared with the prior art, the present invention has the following beneficial effects: (1) A coal seam structure-physical parameter intelligent perception model integrating multi-scale spatial attention and bedding context modeling is proposed, which can simultaneously realize the semantic segmentation of coal seam images and the joint estimation of regional physical parameters, significantly improving the recognition accuracy and physical property perception ability of coal seam heterogeneous structure; (2) A three-dimensional finite element modeling method for the complex structure and material heterogeneity of coal seams was constructed. Regional stratification and meshing were performed based on the semantic segmentation results, and the material parameters output by the perception model were accurately mapped to the finite element units, achieving high-fidelity simulation modeling of coal seam structure-physical property integration. (3) A fracturing parameter optimization strategy that integrates a global proxy model embedded with coal seam physical information and a multi-objective evolutionary algorithm was designed. This strategy can efficiently approximate the global response law based on finite element simulation and dynamically solve the optimal fracturing scheme that meets multiple constraints such as fracturing effect, construction cost, and conductivity. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of the overall technical route of the present invention.

[0015] Figure 2 This is the overall framework diagram of the coal seam structure-physical parameter intelligent perception model.

[0016] Figure 3 Schematic diagram of the proxy model structure for embedding coal seam physical information.

[0017] Figure 4 This is a graph showing the experimental results of semantic segmentation of coal seam structure in the embodiment.

[0018] Figure 5 It is the super volume convergence curve in the embodiment.

[0019] Figure 6 This is a visualization of the optimal population target value in the embodiment. DETAILED DESCRIPTION

[0020] The present invention proposes a method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning. First, based on the coal seam CT slice image, a semantic segmentation and calibration data set covering the coal seam geological structure and its material physical properties is constructed; secondly, a coal seam structure-physical parameter intelligent perception model is designed to realize the joint prediction of semantic recognition and physical property estimation of coal seam images; then, the output results of the coal seam structure-physical parameter intelligent perception model are used to reconstruct the three-dimensional structure of the coal seam and regional layered modeling, and on this basis, finite element mesh division and material property precise mapping are performed to construct a three-dimensional finite element model of the coal seam that can truly characterize the heterogeneous structure and complex physical properties of the coal seam; finally, a multi-objective optimization problem of coal seam hydraulic fracturing parameters is constructed, and an agent model with coal seam physical information embedded is designed and combined with a multi-objective evolutionary algorithm to dynamically solve the optimization problem, so as to obtain the optimal hydraulic fracturing engineering parameters that meet the balance of transformation effect, construction cost and drainage capacity.

[0021] The overall process of this embodiment is as follows Figure 1 As shown: Coal seam semantic segmentation and material calibration dataset production: First, the coal seam slice CT images are systematically collected in groups, and the key geological structures in the CT images are semantically annotated (including background, main coal seam, interlayer gangue, original cracks, faults / joints, mineralized zones, weak interlayers, and aquifers). In addition, for each semantic area, the corresponding material physical parameters are calibrated (including elastic modulus, Poisson's ratio, permeability, tensile strength, shear strength, fracture toughness, crack-matrix coupling ratio, saturation, fracturing-induced fracture threshold, and fracturing priority), so as to provide data support for the subsequent coal seam structure-physical parameter intelligent perception model training; Constructing a coal seam structure-physical parameter intelligent perception model: This model takes CT images of grouped coal seam slices as input, and can realize the semantic region segmentation of coal seams and the synchronous estimation of corresponding material physical parameters; specifically, the present invention designs two key modules, including a coal seam multi-scale spatial attention feature encoder and a bedding context-aware decoder; Three-dimensional finite element modeling of coal seams: Three-dimensional finite element modeling is performed based on the obtained coal seam semantic segmentation and material physical parameters. Specifically, regional stratification and meshing are first performed according to the semantic segmentation information, and then the material physical parameters are accurately mapped to each unit of the finite element model according to the semantic region to obtain a three-dimensional finite element model of the coal seam. Dynamic optimization of coal seam hydraulic fracturing parameters based on agent assistance: first, a multi-objective optimization problem of coal seam hydraulic fracturing parameters is constructed; secondly, an agent model with coal seam physical information embedded is constructed; thirdly, training data and test data are generated based on the constructed three-dimensional finite element model of the coal seam, and the agent model with coal seam physical information embedded is trained and tested to obtain a global agent model; finally, the global agent model is combined with a multi-objective evolutionary algorithm to dynamically solve the multi-objective optimization problem of coal seam hydraulic fracturing parameters and obtain the optimal coal seam hydraulic fracturing parameters.

[0022] The specific implementation process of the present invention is described in detail below in conjunction with specific embodiments.

[0023] 1. Construction of coal seam semantic segmentation and material calibration dataset In order to realize the semantic segmentation of coal seam structure based on coal seam CT images and the intelligent perception of material physical parameters, and to use them for the subsequent construction of coal seam finite element models, the present invention first needs to collect coal seam semantic segmentation and material calibration data sets, including the acquisition of initial coal seam slice CT images, as well as the calibration of semantic categories and material physical parameters; specifically, the following steps are included: S11, for the coal seam to be collected, a slice CT image sequence with spatial continuity is obtained; specifically, the coal seam is evenly divided into slices, so each set of CT image data contains The slice images collected at equidistant intervals are , thus ensuring that the CT image can reflect the three-dimensional coal seam block structure; and, the acquisition process uses a unified coordinate system calibration and ensures consistent image resolution to support subsequent three-dimensional reconstruction and regional matching; S12, based on the grayscale distribution, texture morphology, structural continuity and other characteristics of coal rock in CT images, combined with the existing coal seam lithology data and geological structure interpretation, the LabelMe tool was used to Slice images The semantic annotation of the key coal seam geological structure in the GIS is carried out, including 8 categories: background, main coal seam, interlayer gangue, original fracture, fault / joint, mineralized zone, weak interlayer, and aquifer. Semantic segmentation map of slice images ; S13, based on semantic annotation, calibrate the corresponding material physical parameters for each structural category area. The calibrated parameters cover 10 key indicators closely related to stress propagation, crack induction, seepage characteristics, etc. in the hydraulic fracturing process, including elastic modulus , Poisson's ratio , Permeability , tensile strength , Shear Strength , fracture toughness , fracture-matrix coupling ratio , Saturation , Fracturing-induced fracture threshold , Fracturing Priority Therefore, for slice images, a total of Group material physical parameter diagram ,The material physical parameter map contains the material physical parameters corresponding to each coal seam semantic area; Repeat the above process to collect The data sets together constitute the coal seam semantic segmentation and material calibration dataset, and each data set contains a slice CT image sequence. , semantic segmentation map , and material physical parameter diagrams .

[0024] 2. Construction of intelligent perception model of coal seam structure and physical parameters The coal seam structure-physical parameter intelligent perception model designed by the present invention can intelligently estimate the corresponding coal seam semantic segmentation map and material physical parameter map based on the slice CT image sequence; the model is mainly composed of two key modules, including a coal seam multi-scale spatial attention feature encoder and a bedding context perception decoder; the overall model framework is as follows Figure 2 As shown in the figure, the model building and acquisition process specifically includes the following steps: S21, for Each slice of CT image to be segmented ,in ; Firstly, two 3*3 convolution blocks are used to extract preliminary features of CT images. Each 3*3 convolution block consists of a convolution layer with a convolution kernel of 3*3, a batch normalization layer and a ReLU activation function to extract local texture information and basic spatial features; then, three maximum pooling layers are introduced in turn for multi-scale downsampling to obtain coal seam structural features at different spatial scales, and the feature outputs from shallow to deep are as follows: , corresponding to the front-layer features, middle-layer features and deep-layer features respectively; S22, after completing the preliminary feature extraction, for multi-scale features A coal seam multi-scale spatial attention feature encoder is designed to realize the deep feature extraction of coal seam structure and the fusion of bedding context information. The encoder consists of a hybrid feature encoding block, an upsampling operation, an average pooling layer, and a spatial attention layer. The specific process is as follows: (1) Feature fusion from deep to shallow: deep features After upsampling, compared with the middle-level features The fusion feature is obtained by fusing through the second mixed feature encoding block ; then continue to Upsample and combine with shallow features Further fusion is performed through the first hybrid feature encoding block to obtain a shallow fusion result ; (2) Feature enhancement fusion from shallow to deep: shallow fusion results After average pooling, combined with the fusion feature The third hybrid feature encoding block is input together to enhance the perception of mid-scale features and obtain fused features ; then further Perform average pooling and combine with deep features Input to the fourth hybrid feature encoding block and output the fusion result ; (3) Feature integration and spatial attention modeling: integrating features from three different scales , , Channel splicing is performed to construct a multi-scale feature set; finally, the spatial distribution characteristics of key structures in the image (including coal seam interfaces and crack boundaries) are enhanced through the spatial attention module, and the final output represents the Layer structure feature map of the slice image ; Among them, the hybrid feature encoding block is used to fuse the multi-level features of coal seam images at different scales to enhance the model's perception of complex geological structures. For the two sets of input features to be fused, they are first spliced ​​in the channel dimension to form a joint feature representation; then, the joint feature is simultaneously input into three parallel pathways to extract spatial and semantic information from different angles: Path 1 uses a residual connection structure with batch normalization to retain the original feature trend and boundary information, and enhance the stability and transferability of feature fusion; Path 2 first compresses the channel through a 1*1 convolution block, and then extracts local structural features through a maximum pooling operation to enhance the model's responsiveness to key areas such as the main coal seam boundary and interbedded gangue blocks; Path 3 performs channel mapping through a 1*1 convolution block, and then introduces a RepVGG layer with deep modeling capabilities to obtain a stronger semantic representation, and finally a 3*3 convolution layer is used to further improve the perception accuracy of spatial bedding changes. Finally, the features extracted from the three pathways are spliced ​​through the channels to form a fused feature representation; S23, for Slice CT images , respectively through S1 and S2 to obtain the corresponding The bedding structure characteristics corresponding to the CT images ;Since the coal seam slices are collected using the principle of equidistant interlayer sampling, and the coal seam bedding structure is highly continuous in physical space, there is a significant contextual semantic correlation between adjacent slices; In order to make full use of this interlayer continuity, the present invention designs a bedding context-aware decoder, which achieves more accurate and consistent image semantic reconstruction by combining the bedding information of adjacent slices; Specifically, for the Bedding structure characteristics of each slice , combined with the bedding structure characteristics of the two adjacent slices and Perform channel splicing to form context fusion features (when When selecting and As a structural feature of its adjacent bedding; When selecting and as its adjacent bedding structure feature); then, the context fusion feature It is input into a 3*3 deconvolution block for preliminary decoding upsampling, and further introduces a multi-head attention mechanism to explicitly model and enhance the features of key structures in the bedding context, improve the spatial recognition ability of complex geological structures of coal seams (including faults, interlayers, and cracks), and obtain enhanced features. ; Then, Decoding is performed through four parallel feature reconstruction pathways, each of which extracts information from different receptive fields and modeling strategies. The first pathway uses a combination of 3*1 and 1*3 deconvolution blocks to enhance the directional perception of the horizontal and vertical boundary structures of the bedding; the second pathway uses a combination of 3*3 and 1*1 deconvolution blocks to fuse local details with global contour information; the third pathway uses a combination of 5*5 and 1*1 deconvolution blocks to expand the receptive field and capture larger-scale bedding change features; the fourth pathway effectively retains the original semantic features and strengthens deep structural information through the residual connection mechanism. Finally, the outputs of the four pathways are spliced ​​to obtain the channel-wise prediction for the first pathway. Slice image layer structure feature map The decoder output features ; S24, and finally through the output layer Perform feature compression and map them separately The estimated result of the final semantic segmentation map of the slice CT image , and the estimated results of the material physical parameter map ; S25, based on the constructed coal seam semantic segmentation and material calibration data set, the coal seam structure-physical parameter intelligent perception model is trained. During the training process, the cross entropy loss function is used to calculate the classification error of the semantic segmentation map, and the mean square error (MSE) loss function is used to calculate the regression error of the material physical parameter map; the model training uses the Adam optimization algorithm to update the parameters, and finally the trained coal seam structure-physical parameter intelligent perception model is obtained.

[0025] 3. 3D finite element modeling of coal seams Based on the estimated results of the semantic segmentation map and the material physical parameter map, a three-dimensional finite element model that can truly reflect the coal seam structure distribution and material physical properties is constructed to support the numerical simulation and parameter optimization of the subsequent hydraulic fracturing process; specifically, the following steps are included: S31, 3D structure reconstruction and regional hierarchical modeling: First, the semantic segmentation map of the serial CT slice images is used to estimate the results. Reconstruct the three-dimensional structure; obtain the three-dimensional semantic structure distribution of the coal seam by arranging the equidistant slice images in sequence in a unified coordinate system, construct corresponding structural voxel models for different semantic categories (including main coal seam, interlayer, fault, and fissure), and perform regional stratification to obtain a hierarchical semantic structure model; S32, Meshing and structural discretization: Based on the hierarchical semantic structural model, the three-dimensional finite element method is used for structural discretization, and an unstructured meshing strategy is adopted for complex geological structural forms (including original fractures, faults / joints, mineralized zones, and weak interlayers). The unit density is adaptively adjusted in combination with the geometric characteristics of different regions, and finally a discretized finite element model is obtained; S33, Material physical parameter mapping: Mapping the material physical parameter estimation results Various parameters (such as elastic modulus, permeability, shear strength, etc.) in the model are accurately mapped to the finite element grid unit according to the semantic area; the mapping process supports the parameter correspondence at the pixel level, ensuring that the model can carefully reflect the heterogeneity and spatial variability of the physical property distribution inside the coal seam, and obtain a three-dimensional finite element model of the coal seam. , Represents the physical parameters of the material.

[0026] 4. Dynamic optimization process of coal seam hydraulic fracturing parameters based on agent assistance 1. Construct a multi-objective optimization problem for coal seam hydraulic fracturing parameters The present invention takes maximizing the transformation volume, minimizing the comprehensive cost and maximizing the fracture conductivity as the optimization objectives, and takes the engineering parameters in the hydraulic fracturing process as the decision variables to establish a multi-objective optimization problem of coal seam hydraulic fracturing parameters, which is specifically as follows: Determine decision variables: In order to optimize the key parameters of the entire process of coal seam hydraulic fracturing, the present invention selects 9 key parameters as decision variables, including 6 fracturing construction parameters and 3 engineering control parameters. The fracturing construction parameters include displacement , its engineering significance is to control the crack expansion speed; sand ratio , its engineering significance is to affect the proppant delivery efficiency; fracturing fluid viscosity Its engineering significance is to affect the sand carrying capacity and filtration loss. Injection time , its engineering significance is to control the duration of crack extension; stage volume , used to control the scale of single-stage fracturing; pumping rate , its engineering significance is to affect the continuity of construction. Engineering control parameters include perforation cluster spacing , its engineering significance is to control the crack initiation position; proppant particle size , its engineering significance is to balance the flow conductivity and sedimentation; the concentration of temporary plugging agent , its engineering significance is the balance of image multi-cluster expansion; Design objective function: The objective function is as follows: ; in To transform the volume objective function, minimizing this objective function helps to improve the utilization rate of natural fractures in coal seams. is the transformation volume, which is calculated from the three-dimensional finite element model of the coal seam; is a comprehensive cost objective function, which covers the cost of sand material, fracturing fluid and temporary plugging agent consumption. Minimizing this objective function helps to reduce the cost of coal seam hydraulic fracturing. ， ， and They are fixed cost coefficient of displacement, fixed cost coefficient of sand ratio, fixed cost coefficient of stage volume and fixed cost coefficient of temporary plugging agent concentration; is the fracture conductivity objective function. Minimizing this objective function helps to improve the fracture conductivity of coal seams after hydraulic fracturing. is the conductivity coefficient, is the fracture permeability, calculated from the three-dimensional finite element model of the S coal seam; 2. Constructing an agent model with embedded coal seam physical information Calculation using a three-dimensional finite element model of the coal seam and It takes a long time for simulation. In addition, since there is no time dependency and space dependency between the nine decision variables to be optimized in the present invention, it is difficult for traditional time series models and space models to fully learn the potential correlation between the nine decision variables and the effects of the decision variables on the decision variables. and In order to reduce the simulation time cost of the finite element model and improve the optimization efficiency, the present invention constructs a proxy model with coal seam physical information embedded in it to learn the decision variables. and The mapping relationship between them is as follows: Figure 3 The proxy model for coal seam physical information embedding includes an encoder, a large-scale physical information extraction layer, a small-scale physical information extraction layer, a physical information aggregation layer, and a fully connected layer. data= [ ; ; ; ; ; ; ; ; ] Input coal seam physical information embedded in the proxy model to obtain and Prediction value, the specific steps are as follows: 1) Input data into the encoder to extract the primary features of each decision variable data and obtain the primary feature matrix of the decision variable ; The encoder is composed of a convolution layer and a RELU activation function layer. The convolution layer is used to extract the local features of each decision variable data, and the RELU activation function layer is used to perform nonlinear transformation on the local features of each decision variable data to enhance the generalization ability of the model; 2) Input data into the large-scale physical information extraction layer to extract the large-scale correlation features of each decision variable data and obtain the large-scale correlation feature matrix The large-scale physical information extraction layer uses a convolution kernel scale of The graph attention neural network is used to extract the large-scale features of each decision variable in the data and calculate the correlation coefficient of the large-scale features of each decision variable to obtain ; 3) Input the data into the small-scale physical information extraction layer to extract the small-scale correlation features of each decision variable data and obtain the small-scale correlation feature matrix The large-scale physical information extraction layer uses a convolution kernel scale of The graph attention neural network is used to extract the small-scale features of each decision variable in the data and calculate the correlation coefficient of the small-scale features of each decision variable to obtain ; 4) and After adding Input the physical information aggregation layer together to aggregate the graph features and local features to obtain aggregated features The physical information aggregation layer uses a graph aggregation neural network to combine the characteristics of each decision variable in F with and The features of each decision variable in the added feature graph are sampled and aggregated respectively to obtain the aggregated features that characterize the implicit relationship between the decision variable data. ; 5) Input the fully connected layer to reduce the dimension and map the features and Predicted value; 3. Initialize the proxy model The Latin hypercube sampling method is used to 、 、 、 、 、 , 、 and within its value range Sampling, get Group engineering parameter data; The engineering parameter data of the group are used as the parameters of coal seam hydraulic fracturing simulation in turn, and the established coal seam three-dimensional finite element model is simulated in the time domain to obtain The reconstruction volume and fracture permeability corresponding to the engineering parameter data are obtained. The engineering parameter data and their corresponding transformation volume and fracture permeability are combined and divided into training set and test set in a ratio of 8:2. The proxy model embedded with coal seam physical information is trained and tested using the training set and test set respectively, and the optimal model structure is saved to obtain the global proxy model; 4. Dynamic solution of multi-objective optimization problem of coal seam hydraulic fracturing parameters The present invention combines the obtained global proxy model with a multi-objective evolutionary algorithm to dynamically solve the multi-objective optimization problem of coal seam hydraulic fracturing parameters. The specific steps are as follows: 1) First, Latin hypercube sampling is used to select 、 、 、 、 、 、 、 and Sampling within its value range Obtained The first step is to use the decision variables of each group as the parameters of coal seam hydraulic fracturing simulation, and then perform time domain simulation on the three-dimensional finite element model of the coal seam to calculate the transformation volume and fracture permeability corresponding to each group of decision variables. The second step is to use the decision variables of each group and their corresponding transformation volume and fracture permeability into the objective function to calculate the target value of each group of decision variables. Finally, the decision variables and the target value are combined to obtain the population. , among which Individuals ,in For the The target value vector of each individual in the population ; 2) Initialize the knowledge base and Store in knowledge base; 3) Initialize the number of evaluations and set the maximum number of evaluations; 4) Yes Perform binary tournament selection to obtain paired populations ; 5) Yes Perform crossover and mutation operations to obtain the offspring population ; 6) Using the global proxy model prediction obtained in S5-3 The transformation volume and fracture permeability of all population individuals in , the decision variables of the population individuals and their corresponding transformation volume and fracture permeability are brought into the objective function described in S5-1 to calculate the target value of the population individuals; 7) Yes Perform non-dominated sorting and crowding distance calculation to filter out the population on the first Pareto front ;Will The decision variables of each individual in the population are used as the parameters of the coal seam hydraulic fracturing simulation in turn, and the time domain simulation is performed on the three-dimensional finite element model of the coal seam to calculate the transformation volume and fracture permeability corresponding to each group of decision variables; the decision variables of the individual in the population and their corresponding transformation volume and fracture permeability are again brought into the objective function to calculate the target value of the individual in the population; 8) Update the number of evaluations and compare the number of evaluations with Add up the size of 9) Stored in the knowledge base; when the number of individuals in the knowledge base is greater than When , firstly, the population in the knowledge base is non-dominated and the cosine similarity is calculated, and then the population individuals are paired using the cosine similarity; finally, the population individual with a larger non-dominated ranking in each pair of population individuals is removed from the knowledge base; 10) and Merge to obtain the merged population, perform non-dominated sorting and crowding distance calculation on the merged population, and select the sorting as The next generation of population individuals is updated ; 11) Determine whether the number of evaluations is greater than the maximum number of evaluations. If so, output , otherwise repeat steps 4)-11); Step 11) Output That is the optimal Pareto solution set of the multi-objective optimization problem of coal seam hydraulic fracturing parameters. The decision variable of the population individual with the smallest L2 norm between the target value and the origin is selected as the final optimization parameter.

[0027] 5. Simulation Experiment 1. Verify the performance of the coal seam structure-physical parameter intelligent perception model In order to verify the effect of the coal seam structure-physical parameter intelligent perception model in the semantic segmentation of coal seam structure, the present invention selects multiple groups of representative coal seam slice CT images as input and outputs the corresponding structural semantic segmentation results; in addition, since the material physical parameter map is a further mapping based on semantic region recognition, the experiment focuses on reflecting the model's ability to identify coal seam structure through semantic segmentation results; the experimental results are shown in Figure 2. Figure 4 As shown, pink is the main coal seam, blue-green is interlayer gangue, yellow is weak interlayer, and black is original fracture; Since the multi-scale spatial attention feature encoder can capture the cross-scale geological structure differences in coal seam images and strengthen the representation of key structures (such as main coal seams and interlayers, weak interlayers, etc.) through the spatial attention mechanism, the main coal seam (pink) area in the segmentation result has clear contours and coherent morphology, and interlayer interlayers (blue-green) and weak interlayers (yellow) can be accurately distinguished in the structural transition zone. This result verifies the advantages of multi-scale modeling in fine-grained structural division; in addition, the bedding context-aware decoder introduces contextual bedding information, so that the model can still maintain accurate judgment of structural boundaries under complex textures, thereby ensuring that the model still has high-quality detection effects on complex textures such as original fractures (black). Therefore, the experiment shows that the coal seam structure-physical parameter intelligent perception model shows good segmentation accuracy and structural distinction ability in multi-structure complex coal seam images; 2. Verify the performance of the coal seam physical information embedded proxy model proposed in the present invention combined with the multi-objective evolutionary algorithm to solve the multi-objective optimization problem of coal seam hydraulic fracturing parameters proposed in the present invention According to the data of a coalfield in Luxi area of ​​Shandong Province, a multi-objective optimization problem of coal seam hydraulic fracturing parameters is established. The multi-objective optimization problem of coal seam hydraulic fracturing parameters is solved by using the method of the present invention (PINN-MOEAS), NSGA-II based on Kriging model (Kriging-NSGA-II) and MOPSO based on Kriging model (Kriging-MOPSO). The hypervolume convergence curve is as follows: Figure 5 shown.

[0028] Depend on Figure 5 It can be seen that when the function evaluation times (FEs) reaches 600, the hypervolume value (HV) of the method of the present invention is greater than that of NSGA-II based on the Kriging model and MOPSO based on the Kriging model, which proves that the method of the present invention can improve the convergence speed of the population and the efficiency of solving the multi-objective optimization problem of coal seam hydraulic fracturing parameters. It proves that the proxy model embedded with coal seam physical information used in the present invention has high prediction accuracy and can accurately guide the population to search. The two-dimensional visualization and three-dimensional visualization of the best population obtained by the method of the present invention on the three optimization objectives are shown in Figure 2. Figure 6 As shown. Figure (a) shows the population target value 1 ( ) and the population target value 2 ( ), and Figure (b) is a two-dimensional visualization scatter plot of population target value 1 and population target value 3 ( ), and Figure (c) is a two-dimensional visualization scatter plot of population target value 1, population target value 2, and population target value 3. It can be seen that the solution set obtained by the method of the present invention is evenly distributed in space whether in the two-dimensional scatter plot or the three-dimensional scatter plot, which proves that the method of the present invention can effectively maintain the diversity of the solution set.

[0029] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0030] Although the above describes the specific implementation methods of the present invention, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning, characterized in that: The following steps are involved: S1, collect CT images of coal seam slices with spatial continuity; S2, input the CT image into the constructed and trained coal seam structure-physical parameter intelligent perception model, which includes a coal seam multi-scale spatial attention feature encoder and a bedding context perception decoder, which is used to output the coal seam semantic segmentation results and the corresponding material physical parameters; S3, performing regional stratification and meshing according to the semantic segmentation results, and then mapping the corresponding material physical parameters to each unit of the finite element model according to the semantic region to obtain a three-dimensional finite element model of the coal seam; S4, taking maximizing the transformation volume, minimizing the comprehensive cost and maximizing the fracture conductivity as the optimization objectives, and taking the coal seam hydraulic fracturing parameters to be solved as the decision variables, a multi-objective optimization problem of coal seam hydraulic fracturing parameters is established; the multi-objective optimization problem is solved based on the agent-assisted multi-objective evolutionary algorithm to obtain the optimal coal seam hydraulic fracturing parameters; The agent-assisted multi-objective evolutionary algorithm generates training data and test data based on the constructed three-dimensional finite element model of the coal seam to train and test the agent model embedded with the physical information of the coal seam to obtain a global agent model, and combines the global agent model with the multi-objective evolutionary algorithm to dynamically solve the multi-objective optimization problem.

2. A method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning as claimed in claim 1, characterized in that: Based on the model training task, the coal seam semantic segmentation and material calibration datasets are collected and constructed. The specific process is as follows: First, for the coal seam to be collected, a slice CT image sequence with spatial continuity is obtained: the coal seam is evenly divided into slices, each set of CT image data contains The slice images collected at equidistant intervals are ,The acquisition process adopts a unified coordinate system calibration; Then The key coal seam geological structures in the slice images are semantically annotated, including 8 categories: background, main coal seam, interlayer gangue, original cracks, faults / joints, mineralized zones, weak interlayers, and aquifers. Semantic segmentation map of slice images ; Finally, based on the semantic annotation, the corresponding material physical parameters of each structural category area are calibrated. The calibrated parameters include elastic modulus , Poisson's ratio , Permeability , tensile strength , Shear Strength , fracture toughness , fracture-matrix coupling ratio , Saturation , Fracturing-induced fracture threshold , Fracturing Priority ;against slice images, a total of Group material physical parameter diagram ,The material physical parameter map contains the material physical parameters corresponding to each coal seam semantic area.

3. The method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning according to claim 1, characterized in that: The specific processing process of the layer structure-physical parameter intelligent perception model is as follows: S21, for Each slice of CT image to be segmented ,in ; First, two 3*3 convolution blocks are used to perform preliminary feature extraction on the CT image. Each 3*3 convolution block consists of a convolution layer with a convolution kernel of 3*3, a batch normalization layer, and a ReLU activation function to extract local texture information and basic spatial features; Subsequently, three maximum pooling layers are introduced in turn for multi-scale downsampling to obtain the coal seam structural characteristics at different spatial scales, and the feature outputs from shallow to deep are: , corresponding to the front-layer features, middle-layer features and deep-layer features respectively; S22, multi-scale features Input the coal seam multi-scale spatial attention feature encoder to realize the deep feature extraction of coal seam structure and the fusion of bedding context information: For deep features After upsampling, compared with the middle-level features The fusion feature is obtained by fusing through the second mixed feature encoding block ; then continue to Upsample and combine with shallow features Further fusion is performed through the first mixed feature encoding block to obtain a shallow fusion result ; Shallow fusion results After average pooling, combined with the fusion feature The third hybrid feature encoding block is input together to enhance the perception of mid-scale features and obtain fused features ; then further Perform average pooling and combine with deep features Input to the fourth hybrid feature encoding block and output the fusion result ; The fusion features of three different scales , , Perform channel splicing and construct a multi-scale feature set; Finally, the spatial attention module is used to enhance the spatial distribution characteristics of the key structures in the image. The final output represents Layer structure feature map of the slice image ; S23, based on the bedding context-aware decoder, performs image semantic reconstruction by combining the bedding information of adjacent slices: For the Bedding structure characteristics of each slice , combined with the bedding structure characteristics of the two adjacent slices and Perform channel splicing to form context fusion features ; Afterwards, the context fusion feature It is input into a 3*3 deconvolution block for preliminary decoding and upsampling, and a multi-head attention mechanism is introduced to explicitly model and enhance the key structures in the layered context to obtain enhanced features. ; Then, After decoding through four parallel feature reconstruction channels, each channel extracts information from different receptive fields and modeling strategies, and the outputs of the four channels are spliced ​​to obtain the Slice image layer structure feature map The decoder output features ; S24, and finally through the output layer Perform feature compression and map them separately The estimated result of the final semantic segmentation map of the slice CT image , and the estimated results of the material physical parameter map .

4. A method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning as claimed in claim 3, characterized in that: The first mixed feature coding block, the second mixed feature coding block, the third mixed feature coding block, and the fourth mixed feature coding block are specifically: For the two sets of input features to be fused, they are first concatenated in the channel dimension to form a joint feature representation; then, the joint feature is simultaneously input into three parallel pathways to extract spatial and semantic information from different angles: Path 1 uses a residual connection structure with batch normalization to preserve the original feature trend and boundary information; Path 2 first compresses the channel through a 1*1 convolution block, and then extracts local structural features through a maximum pooling operation to enhance the model's responsiveness to the key areas of the main coal seam boundary and interbedded gangue blocks; Pathway 3 performs channel mapping through a 1*1 convolution block, then introduces the RepVGG layer with deep modeling capabilities to obtain stronger semantic representation, and finally uses a 3*3 convolution layer to further improve the perception accuracy of spatial layer changes; finally, the features extracted from the three paths are channel-joined to form a fused feature representation.

5. The method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning according to claim 3, characterized in that: Said Decoding is performed through four parallel feature reconstruction paths, specifically: Among them, the first pathway uses a combination of 3*1 and 1*3 deconvolution blocks to enhance the directional perception of the horizontal and vertical boundary structures of the bedding; the second pathway uses 3*3 and 1*1 deconvolution blocks to fuse local details with global contour information; the third pathway uses a combination of 5*5 and 1*1 deconvolution blocks to expand the receptive field and capture larger-scale bedding change characteristics; the fourth pathway effectively retains the original semantic features and strengthens the deep structural information through the residual connection mechanism.

6. A method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning as claimed in claim 1, characterized in that: The specific process of S3 is as follows: S31, 3D structure reconstruction and regional hierarchical modeling: First, the semantic segmentation map of the serial CT slice images is used to estimate the results. Reconstruct the three-dimensional structure; obtain the three-dimensional semantic structure distribution of the coal seam by arranging the equidistant slice images in sequence in a unified coordinate system, construct corresponding structural voxel models for different semantic categories, and perform regional stratification to obtain a hierarchical semantic structure model; S32, Meshing and structural discretization: Based on the hierarchical semantic structural model, the three-dimensional finite element method is used for structural discretization, and an unstructured meshing strategy is adopted for complex geological structural morphology. The unit density is adaptively adjusted in combination with the geometric characteristics of different regions, and finally a discretized finite element model is obtained; S33, Material physical parameter mapping: Mapping the material physical parameter estimation results Various parameters in are accurately mapped to the finite element mesh units according to semantic regions; The mapping process supports pixel-by-pixel parameter correspondence, so that the model can carefully reflect the heterogeneity and spatial variability of the physical property distribution inside the coal seam, and obtain a three-dimensional finite element model of the coal seam. , Represents the physical parameters of the material.

7. A method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning as claimed in claim 1, characterized in that: The multi-objective optimization problem of coal seam hydraulic fracturing parameters established in S4 is specifically: Determine the decision variables: 9 key parameters were selected as decision variables, including 6 fracturing construction parameters and 3 engineering control parameters; the fracturing construction parameters include displacement , its engineering significance is to control the crack expansion speed; sand ratio , its engineering significance is to affect the proppant delivery efficiency; fracturing fluid viscosity , its engineering significance is to affect the sand carrying capacity and filtration loss; Injection time , its engineering significance is to control the extension time of cracks; Stage Volume , used to control the scale of single-stage fracturing; pumping rate , its engineering significance is to affect the construction continuity; engineering control parameters include perforation cluster spacing , its engineering significance is to control the crack initiation position; proppant particle size , its engineering significance is to balance the flow conductivity and sedimentation; the concentration of temporary plugging agent , its engineering significance is the balance of image multi-cluster expansion; Design objective function: ； in To transform the volume objective function, the objective function is minimized to improve the utilization rate of natural fractures in coal seams. is the transformation volume, which is calculated from the three-dimensional finite element model of the coal seam; is a comprehensive cost objective function, which covers the cost of sand material, fracturing fluid and temporary plugging agent consumption. Minimizing this objective function is used to reduce the cost of coal seam hydraulic fracturing. ， ， and They are fixed cost coefficient of displacement, fixed cost coefficient of sand ratio, fixed cost coefficient of stage volume and fixed cost coefficient of temporary plugging agent concentration; is the fracture conductivity objective function, and minimizing this objective function is used to improve the fracture conductivity of the coal seam after hydraulic fracturing. is the conductivity coefficient, is the fracture permeability, which is calculated by the three-dimensional finite element model of the coal seam.

8. A method for dynamic optimization of coal seam hydraulic fracturing parameters based on deep learning as claimed in claim 7, characterized in that: The global proxy model includes an encoder, a large-scale physical information extraction layer, a small-scale physical information extraction layer, a physical information aggregation layer and a fully connected layer; data Enter the global proxy model to obtain and Prediction value, the specific steps are as follows: Input data into the encoder to extract the primary features of each decision variable data and obtain the primary feature matrix of the decision variable ; The encoder is composed of a convolution layer and a RELU activation function layer, the convolution layer is used to extract the local features of each decision variable data, and the RELU activation function layer is used to perform nonlinear transformation on the local features of each decision variable data; Input data into the large-scale physical information extraction layer to extract the large-scale correlation features of each decision variable data and obtain the large-scale correlation feature matrix The large-scale physical information extraction layer uses a convolution kernel scale of The graph attention neural network is used to extract the large-scale features of each decision variable in the data and calculate the correlation coefficient of the large-scale features of each decision variable to obtain ; Input data into the small-scale physical information extraction layer to extract the small-scale correlation features of each decision variable data and obtain the small-scale correlation feature matrix The large-scale physical information extraction layer uses a convolution kernel scale of The graph attention neural network is used to extract the small-scale features of each decision variable in the data and calculate the correlation coefficient of the small-scale features of each decision variable to obtain ; Will and After adding Input the physical information aggregation layer together to aggregate the graph features and local features to obtain aggregated features The physical information aggregation layer uses a graph aggregation neural network to combine the characteristics of each decision variable in F with and The features of each decision variable in the added feature graph are sampled and aggregated respectively to obtain the aggregated features that characterize the implicit relationship between the decision variable data. ; Will Input the fully connected layer to reduce the dimension and map the features and Predicted value.

Images