Shale fracture modeling platform based on cloud computing

Abstract

The invention relates to the technical field of cloud computing application, and discloses a cloud computing-based shale fracture modeling platform, which comprises a multi-source data acquisition module, a heterogeneous data fusion module, a fracture dynamic modeling module, a parameter optimization module and a cloud verification module. The multi-source data acquisition module acquires various data and uploads the data to the cloud; the heterogeneous data fusion module processes data isomerism to generate a fusion feature matrix; the crack dynamic modeling module constructs a three-dimensional dynamic model; the parameter optimization module optimizes physical property parameters; and the cloud verification module verifies the model in multiple dimensions and manages the version. The platform solves the problems of traditional modeling in the aspects of data acquisition, fusion, modeling algorithm, parameter optimization and model verification by means of the advantages of cloud computing, can efficiently process multi-source data, accurately construct a crack model, improve the precision and reliability of the model, realize comprehensive verification and effective management of the model, and provide powerful support for efficient exploitation of shale gas.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing applications, and in particular to a shale fracture modeling platform based on cloud computing. Background Art

[0002] As global energy demand continues to grow, shale gas, as an important unconventional natural gas resource, has attracted much attention for its efficient development. The effectiveness of shale gas extraction depends largely on the accurate understanding and modeling analysis of shale fractures, which is directly related to reserve assessment, mining plan design, and recovery rate improvement. However, there are currently many challenges in the field of shale fracture modeling.

[0003] The data required for shale fracture modeling comes from a wide range of sources, including seismic wave data, core physical property data, drilling trajectory data, microseismic monitoring data, and historical fracturing construction data. Traditional data acquisition methods are not only inefficient, but also due to the limitations of data acquisition equipment and the complexity of the acquisition environment, the acquired data often have problems such as low accuracy, missing data, or errors. For example, in remote shale gas mining areas, the signals of sensors are easily interfered with, resulting in noise in the collected seismic wave data, affecting data quality. At the same time, the formats and structures of these multi-source data vary greatly, which brings great difficulties to the centralized storage and unified management of data. Different types of data are stored in different systems or media, and data integration and call are extremely inconvenient, which seriously restricts the subsequent modeling work.

[0004] Since the various types of data are collected at different times, locations, and methods, there are obvious spatiotemporal differences and data characteristics differences between them, that is, heterogeneity. In the time dimension, the acquisition frequencies of seismic wave data and microseismic monitoring data are different, which makes it difficult to align the time series; in the spatial dimension, the sampling points of drilling trajectory data and core physical property data are unevenly distributed, and direct fusion analysis cannot be performed. In addition, the characteristic dimensions of the data are also different. For example, core physical property data contains complex information such as a variety of mineral components and mechanical parameters, and has a higher dimension, while some monitoring data has a relatively low dimension. Existing data fusion technology is difficult to effectively solve these problems and cannot fully explore the potential connections between multi-source data, so that the fused data cannot accurately reflect the true characteristics of the shale formation, affecting the accuracy of modeling.

[0005] Traditional shale fracture modeling algorithms have many shortcomings when dealing with complex geological conditions. Methods based on empirical formulas or simple physical models oversimplify geological processes and cannot accurately describe the complex topological structure and dynamic evolution of shale fractures. For example, these methods usually assume that fractures expand uniformly in homogeneous media, ignoring the impact of formation heterogeneity on fracture expansion. Although methods based on numerical simulation can consider more geological factors to a certain extent, they are computationally expensive and highly dependent on model parameters. When model parameters are inaccurate, the reliability of simulation results will be greatly reduced. In recent years, machine learning algorithms have been applied in shale fracture modeling, but they still face problems such as poor model generalization ability and difficulty in processing large-scale data. For example, deep learning models are prone to overfitting during training and cannot accurately predict fracture morphology and distribution when faced with new geological data.

[0006] The shale fracture model contains many physical parameters, such as fracture permeability, porosity, and stress field distribution parameters. The accuracy of these parameters directly affects the accuracy of the model. However, due to the uncertainty of geological conditions and data limitations, it is very difficult to obtain accurate model parameters. Traditional parameter optimization methods, such as trial and error or simple optimization algorithms, are inefficient and prone to local optimal solutions. These methods cannot fully utilize the information in multi-source data, and it is difficult to achieve coordinated optimization of multiple parameters, resulting in a large deviation between the model and the actual geological conditions.

[0007] In terms of model verification, most of the existing verification methods only test some features of the model and lack multi-dimensional and systematic verification. For example, some methods only focus on whether the geometric morphology of the fracture is consistent with the actual observation, while ignoring the rationality of the physical parameters; other methods, although considering some physical parameters, lack quantitative analysis of model uncertainty. In addition, due to the lack of an effective data sharing mechanism and unified verification standards, model verification results in different research or mining projects are difficult to compare and evaluate, and the reliability and versatility of the model cannot be guaranteed. Summary of the invention

[0008] The purpose of the present invention is to provide a shale fracture modeling platform based on cloud computing to solve the problems raised in the above background technology.

[0009] To achieve the above object, the present invention provides the following technical solutions: a shale fracture modeling platform based on cloud computing, the platform comprising a multi-source data acquisition module, a heterogeneous data fusion module, a fracture dynamic modeling module, a parameter optimization module and a cloud verification module; The multi-source data acquisition module is used to acquire the original multi-source data set required for shale fracture modeling, specifically by deploying a distributed sensor network and geological exploration equipment to collect seismic wave data of shale formations, core physical property data, drilling trajectory data, microseismic monitoring data and historical fracturing construction data, and upload the original multi-source data set to a cloud storage cluster; The heterogeneous data fusion module is used to perform cross-domain fusion and feature enhancement on the original multi-source data, specifically to eliminate the difference in data acquisition delay through the time-space alignment algorithm, and to perform noise reduction on the seismic wave data using a multi-scale decomposition method based on wavelet transform, and to perform feature dimension reduction on the core physical property data in combination with the principal component analysis algorithm, to generate a fusion feature matrix, and to transmit the fusion feature matrix to the fracture dynamic modeling module; The fracture dynamic modeling module is used to construct a three-dimensional dynamic model of shale fractures. Specifically, a fracture topology generation algorithm based on a graph convolutional network is used to perform spatial structure modeling on the fusion feature matrix to generate an initial fracture network graph, and a generative adversarial network is introduced to dynamically optimize the fracture distribution to obtain a shale fracture dynamic model, wherein the discriminator of the generative adversarial network uses an adaptive attention mechanism to optimize weight distribution; The parameter optimization module is used to iteratively optimize the physical property parameters of the fracture model, specifically using a hybrid particle swarm optimization algorithm combined with a Bayesian probability model to perform multi-objective optimization on fracture permeability, porosity and stress field distribution parameters, generate an optimized parameter set, and feed the optimized parameter set back to the fracture dynamic modeling module for model update; The cloud verification module is used to perform multi-dimensional verification on the optimized crack model. Specifically, it constructs a reconstruction error evaluation model based on a variational autoencoder to perform consistency verification on the crack geometry and physical property parameters. At the same time, it combines the Monte Carlo simulation method to quantify the model uncertainty, output the verification results and store them in the cloud database.

[0010] Preferably, in the multi-source data acquisition module, the seismic wave data include the ratio of longitudinal and transverse wave velocities, a reflection coefficient matrix and a dispersion curve; the core physical property data include the percentage of mineral composition, Young's modulus and fracture toughness value; and the historical fracturing construction data include the pumping pressure curve, proppant concentration distribution and fracturing fluid viscosity time series data.

[0011] Preferably, in the heterogeneous data fusion module, the time-space alignment algorithm specifically includes the following steps: performing time series matching on drilling trajectory data and microseismic monitoring data based on a dynamic time warping algorithm, using the Kriging interpolation method to complete the missing spatial data, and extracting the frequency domain characteristics of the seismic wave data through Fourier transform to construct a multi-source data tensor in a unified time-space coordinate system.

[0012] Preferably, in the fracture dynamic modeling module, the fracture topology generation algorithm of the graph convolutional network specifically includes: discretizing the shale formation into a node graph structure, wherein the node attributes include lithological characteristics and stress state, and the edge weight is calculated by the distance between nodes and the physical property difference; using a multi-layer graph convolution operator to extract local fracture connectivity characteristics, and fusing multi-scale topological information through a jump connection mechanism to generate a weighted fracture network graph.

[0013] Preferably, the dynamic optimization process of the generative adversarial network includes: the generator adopts a sequence generation model with a gated recurrent unit to predict the crack propagation path according to the node status of the crack network graph; the discriminator uses a multi-head self-attention mechanism to calculate the rationality score of the crack distribution, and iteratively optimizes the generator parameters through an adversarial loss function.

[0014] Preferably, in the hybrid particle swarm optimization algorithm, the particle position update formula introduces a random perturbation term of the simulated annealing mechanism, specifically: in, Indicates At the iteration, the particle speed; Indicates At the iteration, the particle speed; Represents the inertia weight, which is used to balance the global search and local search capabilities of the algorithm; , represents the learning factor, which is used to adjust the step size of the particle moving to the individual optimal position and the global optimal position; , Represents two random numbers between [0, 1]; Represents particles The best historical position; Indicates At the iteration, the particle location; represents the global optimal position; represents the annealing coefficient, which is used to control the amplitude of random perturbations in the simulated annealing mechanism; represents the cooling rate, which determines how fast the random perturbation weakens as the number of iterations increases; Indicates the number of iterations; Represents a random vector that follows a Gaussian distribution, which is used to enhance the global search capability of the algorithm.

[0015] Preferably, the Bayesian probability model uses a non-parametric kernel density estimation method to construct a joint probability distribution of fracture parameters, specifically by smoothing the parameter association relationship in the historical fracturing data through a kernel function, generating a conditional probability table, and using a Markov chain Monte Carlo sampling method to obtain a parameter optimization direction.

[0016] Preferably, in the reconstruction error evaluation model of the variational autoencoder, the encoder uses a deep residual network to extract latent variables of crack morphology, the decoder uses a deconvolution network to reconstruct the three-dimensional crack geometry, and the loss function includes a weighted sum of a KL divergence term and a mean square error term, which is used to quantify the model reconstruction accuracy.

[0017] Preferably, in the Monte Carlo simulation method, a parameter space sample set is generated by Latin hypercube sampling, and a mapping relationship between crack extension probability and stress field disturbance is constructed based on a response surface model to calculate the confidence interval and sensitivity index of the model output.

[0018] Preferably, the cloud verification module also includes a model version management function, specifically generating a unique identifier for each optimized crack model through a hash algorithm, and recording the parameter optimization path and verification results in the blockchain evidence storage system.

[0019] Compared with the prior art, the present invention has the following beneficial effects: With the help of distributed sensor networks and geological exploration equipment, the multi-source data acquisition module can comprehensively collect various key data, greatly improving the efficiency and accuracy of data acquisition. Compared with traditional acquisition methods, this module can obtain seismic wave data, core physical property data, etc. in real time, reducing data loss and errors. For example, in shale gas mining areas with complex terrain, distributed sensor networks can collect data more stably and reduce the impact of environmental interference. After acquisition, the original multi-source data set is uploaded to the cloud storage cluster, and the powerful storage capacity of cloud computing is used to achieve centralized management and convenient call of data, solving the problem of data dispersion and difficulty in integration in traditional storage methods, and providing a reliable data foundation for subsequent modeling work.

[0020] The heterogeneous data fusion module effectively solves the heterogeneity problem of multi-source data through technologies such as time-space alignment algorithm, wavelet transform noise reduction and principal component analysis feature dimensionality reduction. The time-space alignment algorithm can accurately eliminate the difference in data acquisition delay, such as matching the time series of drilling trajectory and microseismic monitoring data based on the dynamic time warping algorithm, and using Kriging interpolation to fill in the missing data in space to ensure the consistency of data in time and space. Wavelet transform reduces the noise of seismic wave data, improves data quality, and highlights effective signals; principal component analysis reduces the dimensionality of core physical property data, while retaining key information, reducing data redundancy and improving subsequent processing efficiency. The generated fusion feature matrix more accurately reflects the true characteristics of shale formations, laying the foundation for building a high-precision fracture model.

[0021] The fracture dynamic modeling module adopts a fracture topology generation algorithm based on graph convolutional networks and a generative adversarial network with an adaptive attention mechanism, making a major breakthrough in shale fracture modeling. The graph convolutional network discretizes the shale formation into a node graph structure, comprehensively considers the lithological characteristics, stress state and the relationship between nodes, effectively extracts the local fracture connectivity characteristics, and fuses multi-scale topological information through jump connections to generate a more realistic initial fracture network graph. During the dynamic optimization process of the generative adversarial network, the generator uses gated recurrent units to predict the fracture extension path, and the discriminator accurately calculates the rationality score of the fracture distribution through a multi-head self-attention mechanism. The two are trained against each other to make the fracture distribution more reasonable. The constructed three-dimensional dynamic model can more accurately simulate the complex structure and dynamic evolution process of shale fractures, providing a more accurate geological model reference for shale gas extraction.

[0022] The parameter optimization module combines the hybrid particle swarm optimization algorithm with the Bayesian probability model to achieve efficient iterative optimization of the physical parameters of the fracture model. The hybrid particle swarm optimization algorithm introduces the random perturbation term of the simulated annealing mechanism to avoid the algorithm from falling into the local optimal solution, improve the global search capability, and explore the parameter space more comprehensively. For example, when optimizing the fracture permeability, porosity and stress field distribution parameters, the algorithm can dynamically adjust the search strategy according to the actual situation. The Bayesian probability model constructs the joint probability distribution of fracture parameters through non-parametric kernel density estimation, uses Markov chain Monte Carlo sampling to obtain the optimization direction, and makes full use of the information in the historical fracturing data to make the optimized parameters more in line with the actual geological conditions, significantly improving the accuracy and reliability of the model.

[0023] The cloud verification module verifies the optimized fracture model in multiple dimensions by constructing a reconstruction error evaluation model based on a variational autoencoder and a Monte Carlo simulation method. The encoder of the variational autoencoder uses a deep residual network to extract the latent variables of the fracture morphology, and the decoder reconstructs the three-dimensional fracture geometry through a deconvolution network. The loss function comprehensively considers the KL divergence term and the mean square error term, effectively quantifies the model reconstruction accuracy, and verifies the consistency of the fracture geometry and physical parameters. The Monte Carlo simulation method generates a parameter space sample set through Latin hypercube sampling, builds a mapping relationship based on the response surface model, calculates the confidence interval and sensitivity index of the model output, and quantitatively analyzes the model uncertainty. In addition, the model version management function generates a unique identifier through a hash algorithm, and records the parameter optimization path and verification results to the blockchain evidence storage system, realizing effective management of the model version and data traceability, ensuring the stability and traceability of the model quality, and providing a reliable basis for the formulation of shale gas extraction plans.

[0024] The entire platform is built on a cloud computing architecture, making full use of the powerful computing power and flexible scalability of cloud computing. When processing large-scale multi-source data and complex modeling algorithms, cloud computing can quickly complete computing tasks and greatly shorten modeling time. For example, when performing large-scale Monte Carlo simulations, the cloud computing platform can perform parallel calculations, significantly improving computing efficiency. At the same time, the scalability of cloud computing enables the platform to easily cope with the increase in data volume and computing tasks, reduce hardware maintenance costs, improve system stability and reliability, and provide shale gas mining companies with an economical and efficient modeling solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a working principle diagram of the shale fracture modeling platform of the present invention; Figure 2 Flowchart for modeling shale fracture dynamics based on graph convolutional networks and generative adversarial networks; Figure 3 Flowchart of parameter optimization of shale fracture model based on hybrid particle swarm optimization algorithm. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] See also Figure 1-3 The present invention provides a technical solution: a shale fracture modeling platform based on cloud computing, the platform comprising: The multi-source data acquisition module is responsible for acquiring the original multi-source data sets required for shale fracture modeling. By deploying a distributed sensor network and geological exploration equipment, seismic wave data of shale formations, core physical property data, drilling trajectory data, microseismic monitoring data, and historical fracturing construction data are collected. After the data is collected, it will be uploaded to the cloud storage cluster to provide a data basis for subsequent modeling work.

[0028] The heterogeneous data fusion module performs cross-domain fusion and feature enhancement on the original multi-source data. First, the time-space alignment algorithm is used to eliminate the difference in data acquisition delay to ensure the consistency of data in time and space; then, the multi-scale decomposition method based on wavelet transform is used to reduce the noise interference in the data; at the same time, the principal component analysis algorithm is combined to reduce the feature dimension of the core physical property data, reduce the data dimension, improve the data processing efficiency, and finally generate a fusion feature matrix and transmit it to the fracture dynamic modeling module.

[0029] The fracture dynamic modeling module builds a three-dimensional dynamic model of shale fractures based on the fused feature matrix. The fracture topology generation algorithm based on graph convolutional network is used to model the spatial structure of the fused feature matrix and generate the initial fracture network diagram. The generative adversarial network is introduced to dynamically optimize the fracture distribution, and the discriminator of the generative adversarial network uses an adaptive attention mechanism to optimize the weight distribution, thereby obtaining a shale fracture dynamic model.

[0030] The parameter optimization module iteratively optimizes the physical parameters of the fracture model. The hybrid particle swarm optimization algorithm is used in combination with the Bayesian probability model to perform multi-objective optimization of the fracture permeability, porosity and stress field distribution parameters, generate an optimized parameter set, and feed the optimized parameter set back to the fracture dynamic modeling module for model update, so that the model can more accurately reflect the actual situation.

[0031] The cloud verification module verifies the optimized crack model in multiple dimensions. By building a reconstruction error evaluation model based on a variational autoencoder, the consistency of the crack geometry and physical parameters is verified; the model uncertainty is quantitatively analyzed by combining the Monte Carlo simulation method, and finally the verification results are output and stored in the cloud database.

[0032] The present invention will be further described below in conjunction with Examples 1 to 5: Embodiment 1:

[0033] In the multi-source data acquisition module, the various types of data collected are described in detail: Seismic wave data include the ratio of longitudinal and transverse wave velocities, reflection coefficient matrix and dispersion curve. The ratio of longitudinal and transverse wave velocities reflects the difference in elastic properties of shale formations. Different formation structures and lithologies will lead to different ratios of longitudinal and transverse wave velocities. By collecting and analyzing them, we can understand the changes in the lithology of the formation and provide information about the characteristics of the formation medium for fracture modeling. The reflection coefficient matrix contains the characteristic information of seismic waves reflected at different formation interfaces, which can help identify the interface position and property differences of the formation, and is of great significance for determining the area where fractures may exist. The dispersion curve reflects the relationship between the frequency and propagation velocity of seismic waves. Its changing characteristics can reveal the changes in the structure and physical properties of the formation, and provide richer formation information for fracture modeling.

[0034] Core physical property data include mineral composition percentage, Young's modulus and fracture toughness value. The mineral composition percentage directly affects the mechanical properties of shale and the development of cracks. The brittleness, toughness and other properties of shale will vary with the content of different mineral components, thus affecting the generation and expansion of cracks. Young's modulus characterizes the ability of shale to resist elastic deformation. It is an important parameter for describing the mechanical properties of shale and plays a key role in analyzing the deformation and expansion laws of cracks under stress. The fracture toughness value reflects the ability of shale to resist crack expansion and is an important indicator for evaluating the stability of shale cracks.

[0035] Historical fracturing construction data includes pumping pressure curve, proppant concentration distribution and fracturing fluid viscosity time series data. The pumping pressure curve records the change of pressure over time during the fracturing construction process. By analyzing this curve, we can understand the effect of fracturing construction and the response of the formation to pressure, and then infer the expansion of the cracks. The proppant concentration distribution determines the effect of supporting the opening of the cracks. A reasonable proppant concentration distribution can ensure that the cracks maintain a certain opening after fracturing, which is conducive to the extraction of oil and gas. The fracturing fluid viscosity time series data reflects the viscosity change of the fracturing fluid during the construction process. The fracturing fluid viscosity has an important influence on the expansion morphology and efficiency of the cracks. By monitoring its changes over time, we can better understand the fracturing construction process. Embodiment 2:

[0036] In the heterogeneous data fusion module, the spatiotemporal alignment algorithm is implemented in the following steps. First, the drilling trajectory data and microseismic monitoring data are matched in time series based on the dynamic time warping algorithm. The drilling trajectory data records the position and direction information of the wellbore during the drilling process, while the microseismic monitoring data reflects the time and position information of tiny earthquake events caused by underground rock fractures. Since the acquisition time and frequency of these two types of data may be different, it is necessary to find the optimal time correspondence between them through the dynamic time warping algorithm. The basic idea of ​​the dynamic time warping algorithm is to calculate the similarity between the two time series and find a path to align them so that the sum of the distances of the corresponding points on the path is minimized. Suppose there are two time series and , whose distance matrix express and The distance between them is calculated by dynamic programming algorithm. : in, ,Finally, the best matching path is found by backtracking to achieve time series matching of drilling trajectory data and microseismic monitoring data.

[0037] Then, the Kriging interpolation method is used to fill in the missing data. Kriging interpolation is a spatial interpolation method based on regionalized variable theory, which takes into account the spatial autocorrelation of data. Assume that Sampling points Observed value , to estimate the unknown point The value at , the Kriging interpolation formula is: in, is the weight coefficient, which is obtained by solving the equation group. The construction of this equation group is based on the semivariogram. The semivariogram reflects the degree of variation of spatial data. By fitting experimental semivariogram models, such as the spherical model and the exponential model, the parameters of the semivariogram are determined, and then the weight coefficient is calculated to achieve accurate completion of spatial missing data.

[0038] Finally, the frequency domain characteristics of the seismic wave data are extracted through Fourier transform. Fourier transform can convert the seismic wave data in the time domain into frequency domain data, and its formula is: Among them, f(t) is the seismic wave data in the time domain, F(ω) is the frequency domain data, and ω is the frequency. Through Fourier transform, the frequency domain characteristics of the seismic wave data are obtained, and then a multi-source data tensor in a unified time and space coordinate system is constructed, and the data after time series matching and spatial data completion are integrated together to form a unified multi-source data tensor, which facilitates subsequent data fusion and analysis. Embodiment 3:

[0039] In the fracture dynamic modeling module, the fracture topology generation algorithm of the graph convolutional network is carried out in the following steps. First, the shale formation is discretized into a node graph structure, where the node attributes include lithology characteristics and stress state. The lithology characteristics include information such as the mineral composition and porosity of the shale, and the stress state includes information such as the stress magnitude and direction of the formation. The edge weights between nodes are calculated by the distance between nodes and the difference in physical properties. For example, for the node and nodes , its edge weight It can be calculated by the following formula: in, For Node and nodes The distance between and Node and nodes The physical parameters (such as porosity) of the nodes are taken into account in calculating the edge weights, so that the spatial distance and physical property differences between nodes can be comprehensively considered, and the structural relationship of the strata can be more accurately reflected.

[0040] Next, a multi-layer graph convolution operator is used to extract local crack connectivity features. The graph convolution operator can perform convolution operations on the graph structure to extract the neighborhood features of the nodes. Assume that the adjacency matrix of the graph is , the node feature matrix is , No. Output of the layer graph convolution operator It can be calculated by the following formula: in, , is the identity matrix, for The degree matrix of For the The weight matrix of the layer, is an activation function, such as the ReLU function. By stacking multiple layers of graph convolution operators, more advanced local crack connectivity features can be gradually extracted.

[0041] Then, the multi-scale topological information is fused through the skip connection mechanism. The skip connection mechanism allows the features of different layers to be directly connected, avoiding the prediction of the crack extension path in the deep layer of the network. The gated recurrent unit (GRU) can effectively process time series data. and reset gate The calculation formula is as follows: ; in, and is the weight matrix, is the hidden state at the previous moment, is the input at the current moment, is the sigmoid function. Candidate hidden state The calculation formula is:

[0042] Final hidden state The calculation formula is:

[0043] The generator continuously predicts the crack expansion path through GRU based on the node status of the crack network graph.

[0044] The discriminator uses a multi-head self-attention mechanism to calculate the rationality score of the crack distribution. The multi-head self-attention mechanism can learn features from different representation subspaces. The calculation process is as follows. First, the input feature The query matrix is ​​obtained by linear transformation , key matrix Sum Matrix : ; ;

[0045] Then, calculate the attention score matrix : in, is the key matrix Finally, the results of multiple attention heads are concatenated through the multi-head attention mechanism: in, For the The output of an attention head is is the output weight matrix. The discriminator calculates the rationality score of the fracture distribution through the multi-head self-attention mechanism, and iteratively optimizes the generator parameters through the adversarial loss function to make the fracture distribution more reasonable and obtain a more accurate shale fracture dynamic model. Embodiment 4:

[0046] In the parameter optimization module, the particle position update formula in the hybrid particle swarm optimization algorithm introduces the random perturbation term of the simulated annealing mechanism. The particle position update formula is: in, Indicates At the iteration, the particle speed; Indicates At the iteration, the particle speed; Inertia weight is used to balance the global search and local search capabilities of the algorithm. It usually decreases from a larger value during the iteration process. For example, the initial value is set to 0.9 and decreases linearly to 0.4 as the number of iterations increases. , Represents the learning factor, which is generally between [1, 2]. For example, 1.5 is used to adjust the step size of the particle moving to the individual optimal position and the global optimal position. , Represents two random numbers between [0, 1]; Represents particles The best historical position; Indicates At the iteration, the particle location; represents the global optimal position; It represents the annealing coefficient, which is used to control the amplitude of random perturbations in the simulated annealing mechanism. The initial value can be set to 0.5 and gradually decreases as the number of iterations increases. Represents the cooling rate, which determines the speed at which the random disturbance weakens as the number of iterations increases, such as taking the value of 0.01; Indicates the number of iterations; Represents a random vector that follows a Gaussian distribution, which is used to enhance the global search capability of the algorithm. Its mean is 0 and the variance can be adjusted according to actual conditions.

[0047] The Bayesian probability model uses the nonparametric kernel density estimation method to construct the joint probability distribution of fracture parameters. The kernel function is used to smooth the parameter association relationship in the historical fracturing data. Assuming that the fracture parameter samples in the historical fracturing data are , the kernel function is , then the formula for kernel density estimation is: in, is the bandwidth parameter, which is determined by cross-validation and other methods. The conditional probability table is generated by kernel density estimation. Then the Markov chain Monte Carlo sampling method is used to obtain the parameter optimization direction. The Markov chain Monte Carlo sampling method constructs a Markov chain to make its stable distribution the target distribution (that is, the joint probability distribution obtained by kernel density estimation), and samples on the Markov chain to obtain a series of samples. The parameter optimization direction is determined according to the distribution of these samples, thereby realizing multi-objective optimization of fracture permeability, porosity and stress field distribution parameters. Embodiment 5:

[0048] This embodiment elaborates on the specific implementation of the reconstruction error evaluation model of the variational autoencoder, the Monte Carlo simulation method, and the model version management function in the cloud verification module. Through these methods, the optimized fracture model can be fully and deeply verified, the accuracy and reliability of the model can be effectively evaluated, and the model version can be effectively managed and recorded, which is convenient for the subsequent tracing, comparison and improvement of the model, and provides strong support for the practical application of shale fracture modeling.

[0049] In the cloud verification module, the reconstruction error evaluation model of the variational autoencoder is implemented as follows. The encoder uses a deep residual network to extract the latent variables of crack morphology. The deep residual network solves the gradient vanishing problem in the deep neural network training process by introducing residual blocks, and can better learn the characteristics of the data. Assume that the input crack morphology data is After a series of convolutional layers and residual blocks, the encoder output is the latent variable For example, a simple residual block contains two convolutional layers, and its calculation process is: in, is the input of the residual block, is the output, represents the convolution operation, is the weight of the convolutional layer. Through multiple layers of such residual blocks, the encoder gradually extracts high-level features of crack morphology and obtains latent variables .

[0050] The decoder uses a deconvolution network to reconstruct the 3D crack geometry. The deconvolution network is the inverse process of the convolution network, which maps the low-dimensional latent variables back to the high-dimensional image space through the transposed convolution operation. Assuming the latent variable After a series of deconvolution layers and activation functions, the reconstructed crack geometry is obtained. Taking a simple deconvolution layer as an example, its calculation formula is: in, are the weights of the deconvolution layer.

[0051] The loss function contains the weighted sum of the KL divergence term and the mean square error term, which is used to quantify the model reconstruction accuracy. The calculation formula is: in, is the weight coefficient, which takes a value between [0, 1], for example, 0.5; is the KL divergence term, which represents the approximate posterior distribution With prior distribution The difference between them is calculated as:

[0052] In actual calculations, reparameterization techniques are usually used for approximate calculations. is the mean square error term, which is used to measure the original crack morphology Reconstruction of crack morphology The difference between them is calculated as: in, is the number of data samples. By minimizing the loss function , and continuously optimize the parameters of the variational autoencoder to make the reconstructed crack geometry as close to the original structure as possible, thereby performing consistency verification on the crack geometry and physical property parameters.

[0053] In the Monte Carlo simulation method, Latin hypercube sampling is used to generate parameter space sample sets. Latin hypercube sampling is a stratified sampling method that divides the parameter space into several intervals and randomly selects sample points in each interval. This ensures that the samples are evenly distributed in the parameter space and improves sampling efficiency. parameters are sampled, each parameter has intervals, a total of For example, for fracture permeability, porosity and stress field distribution parameters, parameters, and divide the value range of each parameter into intervals, randomly select a sample point from each interval, and obtain a sample set containing 100 sample points.

[0054] The mapping relationship between crack extension probability and stress field disturbance is constructed based on the response surface model. The response surface model is a model that establishes the relationship between variables by fitting data. For example, a quadratic polynomial is used as a response surface model: in, is the crack extension probability, are input parameters such as stress field disturbance, , , , are model parameters, estimated by the least squares method, is the error term. The confidence interval and sensitivity index of the model output are calculated through the response surface model. The confidence interval is used to measure the uncertainty range of the model prediction results, such as calculating the 95% confidence interval of the probability of crack extension. The sensitivity index is used to evaluate the influence of each input parameter on the model output, such as using the variance decomposition method to calculate the sensitivity index : in, Indicates that when other parameters are fixed, the parameter The variance of the mean of the model output caused by the change of is the total variance of the model output.

[0055] The cloud verification module also includes a model version management function. A unique identifier is generated for each optimized crack model through a hash algorithm. The hash algorithm maps the model data to a hash value of a fixed length. For example, the SHA-256 hash algorithm is used, and its calculation formula is: in, is the optimized crack model data, is the generated hash value. This hash value is the unique identifier of the model and is unique and irreversible. At the same time, the parameter optimization path and verification results are recorded in the blockchain evidence system. The blockchain evidence system has the characteristics of decentralization and non-tamperability. The parameter optimization path and verification results of the model are recorded on the blockchain to ensure the security and traceability of the data. For example, the parameter changes, verification time, verification results and other information of each optimization are recorded in the form of blockchain transactions. Each transaction contains information such as transaction timestamp, transaction data, and hash value of the previous transaction, forming a chain structure, which is convenient for subsequent management and backtracking of model versions.

[0056] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0057] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A shale fracture modeling platform based on cloud computing, characterized in that: It includes multi-source data acquisition module, heterogeneous data fusion module, fracture dynamic modeling module, parameter optimization module and cloud verification module; The multi-source data acquisition module is used to acquire the original multi-source data set required for shale fracture modeling, specifically by deploying a distributed sensor network and geological exploration equipment to collect seismic wave data of shale formations, core physical property data, drilling trajectory data, microseismic monitoring data and historical fracturing construction data, and upload the original multi-source data set to a cloud storage cluster; The heterogeneous data fusion module is used to perform cross-domain fusion and feature enhancement on the original multi-source data, specifically to eliminate the difference in data acquisition delay through the time-space alignment algorithm, and to perform noise reduction on the seismic wave data using a multi-scale decomposition method based on wavelet transform, and to perform feature dimension reduction on the core physical property data in combination with the principal component analysis algorithm, to generate a fusion feature matrix, and to transmit the fusion feature matrix to the fracture dynamic modeling module; The fracture dynamic modeling module is used to construct a three-dimensional dynamic model of shale fractures. Specifically, a fracture topology generation algorithm based on a graph convolutional network is used to perform spatial structure modeling on the fusion feature matrix to generate an initial fracture network graph, and a generative adversarial network is introduced to dynamically optimize the fracture distribution to obtain a shale fracture dynamic model, wherein the discriminator of the generative adversarial network uses an adaptive attention mechanism to optimize weight distribution; The parameter optimization module is used to iteratively optimize the physical property parameters of the fracture model, specifically using a hybrid particle swarm optimization algorithm combined with a Bayesian probability model to perform multi-objective optimization on fracture permeability, porosity and stress field distribution parameters, generate an optimized parameter set, and feed the optimized parameter set back to the fracture dynamic modeling module for model update; The cloud verification module is used to perform multi-dimensional verification on the optimized crack model. Specifically, it constructs a reconstruction error evaluation model based on a variational autoencoder to perform consistency verification on the crack geometry and physical property parameters. At the same time, it combines the Monte Carlo simulation method to quantify the model uncertainty, output the verification results and store them in the cloud database.

2. The cloud computing-based shale fracture modeling platform according to claim 1, characterized in that: In the multi-source data acquisition module, the seismic wave data include the ratio of longitudinal and transverse wave velocities, the reflection coefficient matrix and the dispersion curve; the core physical property data include the mineral composition percentage, Young's modulus and fracture toughness value; the historical fracturing construction data include the pumping pressure curve, the proppant concentration distribution and the fracturing fluid viscosity time series data.

3. The cloud computing-based shale fracture modeling platform according to claim 2, characterized in that: In the heterogeneous data fusion module, the spatiotemporal alignment algorithm specifically includes the following steps: time series matching of drilling trajectory data and microseismic monitoring data is performed based on a dynamic time warping algorithm, the Kriging interpolation method is used to complete the missing spatial data, and the frequency domain characteristics of the seismic wave data are extracted through Fourier transform to construct a multi-source data tensor in a unified spatiotemporal coordinate system.

4. The cloud computing-based shale fracture modeling platform according to claim 3, characterized in that: In the fracture dynamic modeling module, the fracture topology generation algorithm of the graph convolutional network specifically includes: discretizing the shale formation into a node graph structure, in which the node attributes include lithological characteristics and stress state, and the edge weight is calculated by the distance between nodes and the physical property difference; using a multi-layer graph convolution operator to extract local fracture connectivity characteristics, and fusing multi-scale topological information through a jump connection mechanism to generate a weighted fracture network graph.

5. The cloud computing-based shale fracture modeling platform according to claim 4, characterized in that: The dynamic optimization process of the generative adversarial network includes: the generator adopts a sequence generation model with a gated recurrent unit to predict the crack extension path according to the node status of the crack network graph; the discriminator uses a multi-head self-attention mechanism to calculate the rationality score of the crack distribution, and iteratively optimizes the generator parameters through an adversarial loss function.

6. The cloud computing-based shale fracture modeling platform according to claim 5, characterized in that: In the hybrid particle swarm optimization algorithm, the particle position update formula introduces the random perturbation term of the simulated annealing mechanism, specifically: in, Indicates At the iteration, the particle speed; Indicates At the iteration, the particle speed; Represents the inertia weight, which is used to balance the global search and local search capabilities of the algorithm; , represents the learning factor, which is used to adjust the step size of the particle moving to the individual optimal position and the global optimal position; , Represents two random numbers between [0, 1]; Represents particles The best historical position; Indicates At the iteration, the particle location; represents the global optimal position; represents the annealing coefficient, which is used to control the amplitude of random perturbations in the simulated annealing mechanism; represents the cooling rate, which determines how fast the random perturbation weakens as the number of iterations increases; Indicates the number of iterations; Represents a random vector that follows a Gaussian distribution, which is used to enhance the global search capability of the algorithm.

7. The cloud computing-based shale fracture modeling platform according to claim 1, characterized in that: The Bayesian probability model uses a non-parametric kernel density estimation method to construct a joint probability distribution of fracture parameters, specifically, by smoothing the parameter association relationship in historical fracturing data through a kernel function, generating a conditional probability table, and using a Markov chain Monte Carlo sampling method to obtain a parameter optimization direction.

8. The cloud computing-based shale fracture modeling platform according to claim 1, characterized in that: In the reconstruction error evaluation model of the variational autoencoder, the encoder uses a deep residual network to extract latent variables of crack morphology, the decoder uses a deconvolution network to reconstruct the three-dimensional crack geometric structure, and the loss function includes a weighted sum of a KL divergence term and a mean square error term, which is used to quantify the model reconstruction accuracy.

9. The cloud computing-based shale fracture modeling platform according to claim 1, characterized in that: In the Monte Carlo simulation method, a parameter space sample set is generated by Latin hypercube sampling, and a mapping relationship between crack extension probability and stress field disturbance is constructed based on a response surface model to calculate the confidence interval and sensitivity index of the model output.

10. The cloud computing-based shale fracture modeling platform according to claim 1, characterized in that: The cloud verification module also includes a model version management function, which specifically generates a unique identifier for each optimized crack model through a hash algorithm, and records the parameter optimization path and verification results to the blockchain evidence storage system.

Images