A method and device for formation pressure prediction for high water injection rate conditions

Abstract

The application provides a formation pressure prediction method and device for high water injection rate conditions, and is suitable for the technical field of data processing. The method comprises the following steps: obtaining water injection speed information, water injection time information and suspended particle concentration information; determining water injection rate and formation permeability according to the water injection speed information, the water injection time information and the suspended particle concentration information; performing multi-dimensional calculation on the formation pressure according to the water injection rate and the formation permeability, to obtain initial formation pressure values in multiple dimensions; generating a plurality of formation pressure matrices according to the initial formation pressure values in multiple dimensions; and performing prediction calculation on the formation pressure matrices based on a preset weight vector, a nonlinear mapping function and a gating vector, to obtain a target formation pressure value. In the case of changes in the formation permeability, the application predicts the formation pressure to evaluate the formation stability during the water injection process, avoids adverse effects of water injection work on drilling, and improves the safety of oil collection work.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and in particular relates to a method and apparatus for predicting formation pressure under high water injection conditions. Background Technology

[0002] Produced water reinjection into the formation is becoming a popular engineering project in offshore oil development. In well operations with a single injection volume of 15,000 to 30,000 cubic meters, production water reinjection and drilling operations are carried out simultaneously. As the injection time and volume increase, the pressure in the production water reinjection well continuously rises, posing significant high-pressure safety hazards to the injection pipeline, wellbore, and formation. Whether long-term, high-volume production water reinjection will cause the injection layer to become an abnormally high-pressure layer, leading to subsequent drilling and development risks, blowouts, and other accidents, urgently needs clarification. Long-term reinjection continuously expands the underground diffusion range, necessitating a clear understanding of the dynamic range of the production water. Furthermore, long-term reinjection causes drastic changes in the formation pressure system and geostress, easily leading to formation instability and other problems, requiring effective safety assessments through formation pressure prediction.

[0003] In existing technologies, formation pressure prediction methods are mainly used in applications where formation pressure is detected before drilling. Based on the temporal characteristics of the detection data and the sedimentary characteristics of the formation, a formation pressure prediction model is constructed to analyze the formation characteristics at different locations and obtain the causes of formation pressure formation.

[0004] However, existing technologies are only used for formation surveys before drilling when formation characteristics are stable. They cannot be used to predict formation pressure when formation characteristics change over time after drilling operations begin or during drilling development. Summary of the Invention

[0005] In view of this, the present application provides a method and apparatus for predicting formation pressure under high water injection conditions. It can predict formation pressure during drilling in multiple dimensions by considering changes in formation permeability. This solves the problem in the prior art that prediction is only performed when the formation characteristics are stable and cannot be performed when the formation characteristics change. This ensures that the reinjection of production water will not cause safety hazards to drilling and development operations during the operation of a 10,000 cubic meter water injection well.

[0006] The first aspect of this application provides a method for predicting formation pressure under high water injection conditions, including:

[0007] Acquire information on water injection rate, water injection time, and suspended particulate concentration;

[0008] Based on the water injection rate information, water injection time information, and suspended particle concentration information, the water injection volume and formation permeability are determined.

[0009] Based on the injected water volume and formation permeability, the formation pressure is calculated in multiple dimensions to obtain initial formation pressure values ​​in multiple dimensions.

[0010] Based on the initial formation pressure values ​​of the multiple dimensions, multiple formation pressure matrices are generated;

[0011] Based on a preset weight vector, nonlinear mapping function, and gating vector, the formation pressure matrix is ​​predicted and calculated to obtain the target formation pressure value.

[0012] A second aspect of this application provides a formation pressure prediction device for high water injection conditions, comprising:

[0013] The information acquisition module is used to acquire information on water injection rate, water injection time, and suspended particle concentration.

[0014] The water injection volume and formation permeability determination module is used to determine the water injection volume and formation permeability based on the water injection rate information, water injection time information and suspended particle concentration information.

[0015] The initial formation pressure value calculation module is used to perform multi-dimensional calculations of formation pressure based on the injection volume and formation permeability to obtain initial formation pressure values ​​in multiple dimensions.

[0016] The formation pressure matrix generation module is used to generate multiple formation pressure matrices based on the initial formation pressure values ​​of the multiple dimensions.

[0017] The target formation pressure value calculation module is used to predict and calculate the formation pressure matrix based on a preset weight vector, a nonlinear mapping function, and a gating vector to obtain the target formation pressure value.

[0018] A third aspect of this application provides a terminal device, the terminal device including a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the computer program to implement the steps of the formation pressure prediction method for high water injection conditions as described in any of the first aspects above.

[0019] A fourth aspect of this application provides a computer-readable storage medium, comprising: storing a computer program, which, when executed by a processor, implements the steps of the formation pressure prediction method for high water injection conditions as described in any of the first aspects above.

[0020] The beneficial effects of this application embodiment compared with the prior art are as follows: This application calculates the formation pressure under the condition of formation permeability change, and then maps the calculated formation pressure value to a multidimensional nonlinear space through a preset nonlinear mapping function. The formation pressure value in the nonlinear space is further analyzed and calculated through preset weight vectors and preset gate vectors. This is used for nonlinear prediction of formation pressure value under the condition of formation permeability change, thereby ensuring that the formation pressure change caused by the reinjection of production water will not affect drilling operations during the offshore oil development process, so as to improve the operational safety during the offshore oil development process. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram illustrating the implementation process of the formation pressure prediction method for high water injection conditions provided in the embodiments of this application.

[0023] Figure 2 This is a schematic diagram illustrating the implementation process of the formation pressure prediction method for high water injection conditions provided in the embodiments of this application.

[0024] Figure 3 This is a schematic diagram illustrating the implementation process of the formation pressure prediction method for high water injection conditions provided in the embodiments of this application.

[0025] Figure 4 This is a schematic diagram illustrating the implementation process of the formation pressure prediction method for high water injection conditions provided in the embodiments of this application.

[0026] Figure 5 This is a schematic diagram illustrating the implementation process of the formation pressure prediction method for high water injection conditions provided in the embodiments of this application.

[0027] Figure 6 This is a schematic diagram illustrating the implementation process of the formation pressure prediction method for high water injection conditions provided in the embodiments of this application.

[0028] Figure 7 This is a schematic diagram illustrating the implementation process of the formation pressure prediction method for high water injection conditions provided in the embodiments of this application.

[0029] Figure 8 This is a schematic diagram of the formation pressure prediction device for high water injection conditions provided in the embodiments of this application;

[0030] Figure 9 This is a schematic diagram of the terminal device provided in the embodiments of this application. Detailed Implementation

[0031] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0032] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0033] Figure 1 The implementation flowchart of the formation pressure prediction method for high water injection conditions provided in Embodiment 1 of this application is shown, and is described in detail below:

[0034] Step S101: Obtain water injection speed information, water injection time information, and suspended particle concentration information.

[0035] In this embodiment, the water injection rate information can refer to the water injection speed, which is the amount of water injected into the formation per unit time, typically expressed in cubic meters per day (m³ / day). 3 / d), cubic meters / month (m 3 / mon) or cubic meters / year (10 4 m 3 / a) represents the time taken to inject water into the formation at a certain rate. Suspended particle concentration information refers to the concentration of suspended particles in the reinjected water, i.e., the quantity or mass concentration of suspended particles in the reinjected liquid. Suspended particles refer to particles suspended in the reinjected liquid with diameters between 0.1 μm and 1000 μm, including silt, clay, protozoa, algae, bacteria, viruses, and high-molecular-weight organic matter. The injection rate information can be measured by specific sensors and manually input into a computer for subsequent analysis. The injection time information can be obtained by measuring the time during the injection operation using a timing device. The suspended particle concentration information can be obtained by collecting production water samples from the oilfield's downstream shore, i.e., reinjected water samples, and analyzing the production water samples using the GB / T14848-9 groundwater quality analysis standard. The analyzed production water information shows a CaCl2 water type, a total mineralization of 3.21%, and a total calcium and magnesium ion content of 973 mg / L, belonging to medium-mineralized water quality. The production water has a pH of 6.5, is weakly acidic, and has a density of 1.0457 g / cm³. 3The viscosity is 0.4 cP (40℃), the mass concentration of suspended particles in water is 1800 mg / L, the median particle size is >0.45 μm, and it exhibits a flocculent, sheet-like structure that agglomerates upon static standing. The analysis results include the concentration of suspended particles, which can be manually input into a computer for subsequent calculations.

[0036] Step S102: Determine the injection volume and formation permeability based on the injection rate information, injection time information, and suspended particle concentration information.

[0037] In this embodiment, samples of production water from the offshore oilfield, i.e., reinjection water, can be collected first. The injection volume can be calculated using the injection rate and time information. Permeability can be determined through core displacement experiments. Specifically, the indoor core displacement experiment can adopt the industry standard SY / T5358-2010 "Evaluation Method for Reservoir Sensitivity Flow Experiment". In the core displacement experiment, cores with a depth of 1003-1005m can be collected, placed in a core holder, and maintained at a manually set reservoir temperature. Production water is used for displacement until the pressure stabilizes. Then, under the same temperature conditions, reinjection water is injected into the core at different injection rates. The formation permeability under specific injection volume conditions is measured, and the measured formation permeability is then manually input into a computer for analysis and calculation.

[0038] Step S103: Based on the water injection volume and formation permeability, perform multi-dimensional calculations on the formation pressure to obtain initial formation pressure values ​​in multiple dimensions.

[0039] In this embodiment, the initial formation pressure value can be calculated using the Python module of Petrel RE. Specifically, the injection volume and its corresponding time period are input into the Python module of Petrel RE to calculate the formation permeability at different times, thereby obtaining the time-varying pattern of formation permeability. Then, the Python module of Petrel RE further analyzes and calculates the pressure distribution characteristics of each block of rock using the time-varying pattern of formation permeability. Extracting the pressure distribution characteristics yields the initial formation pressure value in multiple dimensions. These dimensions can be horizontal, vertical, and depth. The horizontal dimension can be represented by the x-axis, the vertical dimension by the y-axis, and the depth dimension by the z-axis.

[0040] Step S104: Generate multiple formation pressure matrices based on the initial formation pressure values ​​of the multiple dimensions.

[0041] In this embodiment, the initial formation pressure values ​​of multiple dimensions can be first converted into a multi-dimensional vector map, and then the multi-dimensional vector map can be converted into a formation pressure matrix using methods such as One-hot encoding, TF-IDF, Word2Vec, Doc2Vec, BERT, ELMo, GPT, FastText, and GloVe. Alternatively, the initial formation pressure values ​​of multiple dimensions can be converted into a one-dimensional array, and then the one-dimensional array can be converted into a matrix to generate a formation pressure matrix for subsequent predictive analysis calculations.

[0042] Step S105: Based on the preset weight vector, nonlinear mapping function and gate vector, the formation pressure matrix is ​​predicted and calculated to obtain the target formation pressure value.

[0043] In this embodiment, the preset weight vector can be manually set to highlight the important values ​​in the initial formation pressure value through different weights, thereby improving the accuracy of the prediction results. The preset nonlinear mapping function can be a Sigmoid function, a Tanh function, a ReLU function, a Leaky ReLU function, or a Softmax function, used to introduce nonlinear factors into the prediction of formation pressure values. It is understood that the variation of formation pressure values ​​is not linear; therefore, the initial formation pressure value needs to be mapped to a nonlinear space for analysis to achieve prediction calculation. The value obtained after prediction calculation is the target formation pressure value, used to characterize whether the current water injection operation may lead to formation fracturing, thereby guiding the staff to take corresponding measures to avoid the negative impact of continuous water injection on drilling operations.

[0044] The formation pressure prediction method for high water injection conditions provided in this application calculates the formation pressure under varying formation permeability. Then, it maps the calculated formation pressure value to a multidimensional nonlinear space using a preset nonlinear mapping function. Further analysis and calculation of the formation pressure value in the nonlinear space are performed using preset weight vectors and preset gating vectors. This method is used for nonlinear prediction of formation pressure under varying formation permeability, thereby ensuring that formation pressure changes caused by production water reinjection do not affect drilling operations during offshore oil development, thus improving operational safety during offshore oil development.

[0045] Figure 2 The flowchart illustrating the implementation of the formation pressure prediction method for high water injection conditions provided in Embodiment 2 of this application is shown. The difference between this method and Embodiment 1 is that the initial formation pressure value includes formation pressure data values, formation pressure dimension values, and formation pressure dimension axis identifiers; step S104 specifically includes:

[0046] Step S201: Generate an array of formation pressure data values ​​and an array of formation pressure dimension values ​​based on the formation pressure data values ​​and formation pressure dimension values.

[0047] In this embodiment, it can be understood that the initial formation pressure value is a multidimensional value, used to characterize the pressure experienced by a point in the formation in various directions. The formation pressure data value can refer to the resultant force value of the formation pressure in each dimension. The formation pressure dimension value can be the component force value of the formation pressure in each dimension. The formation pressure dimension axis identifier is used to characterize the dimension to which the component force belongs, and can be represented by first dimension, second dimension, and third dimension, used for subsequent concatenation and merging of arrays formed from multiple component force values. It can be understood that the formation pressure data value and the formation pressure dimension value are respectively formed into one-dimensional arrays or multi-dimensional arrays for subsequent concatenation processing.

[0048] Step S202: Based on the formation pressure dimension axis identifier, the formation pressure data value array and the formation pressure dimension value array are concatenated to obtain multiple one-dimensional arrays of formation pressure values.

[0049] In this embodiment, the formation pressure data value array and the formation pressure dimension value array are concatenated according to the order of the formation pressure dimension axis identifiers. For example, when the formation pressure dimension axis identifier is the first dimension, the array of formation pressure dimension values ​​under the first dimension is concatenated, and then the formation pressure data value array is concatenated with the array of formation pressure dimension values ​​to obtain a one-dimensional array, namely, a one-dimensional array of formation pressure values.

[0050] Step S203: Generate multiple formation pressure matrices based on the one-dimensional array of formation pressure values, the preset number of matrix rows, and the preset number of matrix columns.

[0051] In this embodiment, the `reshape` method from the NumPy library can be used to convert a one-dimensional array into a multi-dimensional array. The preset number of matrix rows and columns are both manually set.

[0052] The formation pressure prediction method for high water injection conditions provided in this application reduces the dimensionality of the initial formation pressure values ​​from multiple dimensions. The one-dimensional matrix after dimensionality reduction is then used to regenerate a formation pressure matrix with a fixed number of rows and columns for subsequent prediction and analysis of formation pressure values. This reduces the computational complexity, improves the timeliness of prediction calculations, and facilitates timely detection of abnormal formation pressure values ​​by operators. Consequently, timely measures can be taken to prevent formation fracturing caused by production water reinjection, ensuring the safety of drilling operations.

[0053] Figure 3The flowchart of the formation pressure prediction method for high water injection conditions provided in Embodiment 3 of this application is shown. The difference between this method and Embodiment 1 is that step S105 specifically includes:

[0054] Step S301: Perform convolution operation on the formation pressure matrix, the preset weight vector, and the preset nonlinear mapping function to obtain the formation pressure feature matrix.

[0055] In this embodiment, the formation pressure matrix and the weight vector can be convolved first. Then, all the values ​​of the result of the convolution operation can be used as the independent variables of the nonlinear mapping function to calculate the function value of the nonlinear mapping function. Then, a matrix can be generated from all the calculated function values ​​of the nonlinear mapping function, that is, a formation pressure feature matrix can be generated to characterize the important features in the formation pressure values, thereby ensuring that the important features in the formation pressure values ​​are considered in the subsequent prediction and analysis.

[0056] Step S302: Generate a global representation feature matrix of formation pressure based on the maximum value of all values ​​in the formation pressure feature matrix.

[0057] In this embodiment, by extracting the maximum value from the formation pressure feature matrix, a one-dimensional vector with the same number of scales as the formation pressure feature matrix is ​​obtained. This reduces the size of the formation pressure feature matrix, thereby reducing computational load and memory consumption. All extracted one-dimensional vectors at different scales are concatenated to generate a vector graph, which serves as the global representation feature of formation pressure, i.e., the global representation feature matrix of formation pressure.

[0058] Step S303: Based on the preset nonlinear mapping function and multiple preset gate vectors, the global representation feature matrix of formation pressure is weighted and summed to obtain multiple formation pressure state weight matrices.

[0059] In this embodiment, the gated vector can be used as the weight value for the weighted summation calculation. Alternatively, the formation pressure global representation feature matrix can be weighted and summed based on one or more gated vectors to further enhance the important numerical features in the formation pressure values. The result of the weighted summation can then be used as the independent variable of the nonlinear mapping function, and the function value of the obtained nonlinear mapping function can be used as the value in the formation pressure state weight matrix, thereby generating the formation pressure state weight matrix.

[0060] Step S304: Based on the preset nonlinear mapping function and the preset gating vector, perform nonlinear calculation on the global representation feature matrix of formation pressure to obtain the formation pressure offset weight matrix.

[0061] In this embodiment, the nonlinear calculation can be a hyperbolic tangent function. It can be achieved by using gated vectors as weights in a weighted summation calculation, performing a weighted summation on the global formation pressure representation feature matrix based on one or more gated vectors, and using the calculation result as the independent variable of the hyperbolic tangent function. The calculated function value is then used as the numerical value of the formation pressure offset weight matrix, thereby generating the formation pressure offset weight matrix. This matrix is ​​used to quantify the offset of important features in the formation pressure value relative to other features, thus indirectly quantifying the offset between the target formation pressure value and the initial formation pressure value in the predicted result.

[0062] Step S305: Calculate the target formation pressure matrix based on the formation pressure state weight matrix and the formation pressure offset weight matrix.

[0063] In this embodiment, the target formation pressure matrix can be obtained by performing a dot product between the formation pressure state weight matrix and the formation pressure offset weight matrix. It can be understood that by analyzing and calculating the important features in the initial formation pressure values, the offsets of these important features that may occur during continuous water injection are obtained. Then, by adjusting the offsets of the important features in the initial formation pressure numerical characteristics (which can be done through a dot product operation), the important features of the predicted target formation pressure values ​​are obtained. Therefore, the matrix composed of the important features of the calculated target formation pressure values ​​is the target formation pressure matrix.

[0064] Step S306: Obtain the target formation pressure value based on the target formation pressure matrix.

[0065] In this embodiment, the target formation pressure matrix can be converted into a one-dimensional matrix, and this one-dimensional matrix can be transformed into a one-dimensional array to extract the corresponding values ​​as the target formation pressure values. Specifically, the target formation pressure data values, target formation pressure dimension values, and target formation pressure dimension axis identifiers can be extracted, or only the resultant force values ​​of the target formation pressure in multiple dimensions can be extracted.

[0066] The formation pressure prediction method for high water injection conditions provided in this application introduces a nonlinear function to map the initial formation pressure matrix into a nonlinear space for calculation. It extracts and enhances important features in the initial formation pressure matrix using a weight vector to ensure that these features are considered during the prediction calculation. Furthermore, it calculates the possible offsets of these important features in the initial formation pressure matrix using a gating vector to quantify the potential offsets of formation pressure values ​​in the nonlinear space during the production water reinjection process. Finally, it obtains the predicted target formation pressure value by combining the initial formation pressure matrix with the possible offsets during water injection.

[0067] Figure 4 This document illustrates a flowchart of the formation pressure prediction method for high water injection conditions provided in Embodiment 4 of this application. The difference between this method and Embodiment 3 is that the gating vector includes a first gating vector, a second gating vector, and a third gating vector; the formation pressure state weight matrix includes a first formation pressure state weight matrix, a second formation pressure state weight matrix, and a third formation pressure state weight matrix; and step S303 specifically includes:

[0068] Step S401: Based on the first gating vector, the preset nonlinear mapping function, and the preset time state matrix, perform nonlinear calculation on the global representation feature matrix of formation pressure to obtain the first formation pressure state weight matrix.

[0069] In this embodiment, the global representation feature matrix of formation pressure can be represented as X, the first gating vector can be represented as E1, and the preset nonlinear mapping function can be represented as... ( ). The preset time state matrix can be manually set to introduce a time dimension into the prediction results, so that the predicted formation pressure values ​​can exhibit a nonlinear time-varying law in the time domain, making it easier for staff to observe formation pressure changes at different times. The time state matrix can be represented as T. The process of calculating the first formation pressure state weight matrix Y1 can be represented as:

[0070] Y1= (E1 )

[0071] Step S402: Based on the second gating vector, the preset nonlinear mapping function, and the preset time state matrix, perform nonlinear calculation on the global representation feature matrix of formation pressure to obtain the second formation pressure state weight matrix.

[0072] In this embodiment, the second gating vector can be represented as E2. The process of calculating the second formation pressure state weight matrix Y2 can be expressed as:

[0073] Y2= (E2 )

[0074] Step S403: Based on the third gating vector, the preset nonlinear mapping function, and the preset time state matrix, perform nonlinear calculation on the global representation feature matrix of formation pressure to obtain the third formation pressure state weight matrix.

[0075] In this embodiment, the third gating vector can be represented as E3, and the process of calculating the third formation pressure state weight matrix Y3 can be expressed as:

[0076] Y3= (E3 )

[0077] The formation pressure prediction method for high water injection conditions provided in this application introduces different gating vectors to calculate the global representation feature matrix of formation pressure. This is used to retain features of different scales or dimensions in the initial formation pressure matrix, thereby determining which scales or dimensions of features should be retained or enhanced in subsequent analysis and calculations to ensure the accuracy of formation pressure value prediction. By introducing a time state matrix, a time feature is introduced into the formation pressure prediction, allowing the predicted values ​​to be presented in the time dimension. This facilitates the visualization of the predicted formation pressure values ​​by the operators, enabling timely observation of formation pressure changes during the production water reinjection process. When the formation may or is about to rupture, timely measures can be taken to avoid affecting drilling operations, thereby ensuring the safety and reliability of offshore oil extraction.

[0078] Figure 5 The flowchart illustrating the implementation of the formation pressure prediction method for high water injection conditions provided in Embodiment 5 of this application is shown. The difference between this method and Embodiment 4 is that step S305 specifically includes:

[0079] Step S501: Perform a dot product operation on the first formation pressure state weight matrix and the formation pressure offset weight matrix to obtain the first formation pressure state variable matrix.

[0080] In this embodiment, the formation pressure offset weight matrix can be represented as D, the first formation pressure state variable matrix can be represented as Z1, and the process of calculating the first formation pressure state variable matrix Z1 can be expressed as:

[0081] Z1= D

[0082] in, This represents the dot product operation of matrices.

[0083] Step S502: Perform a dot product operation on the second formation pressure state weight matrix and the preset time state matrix to obtain the second formation pressure state variable matrix.

[0084] In this embodiment, the second formation pressure state variable matrix Z2 can be represented as:

[0085] Z2= D

[0086] Step S503: Summate the first formation pressure state variable matrix and the second formation pressure state variable matrix to obtain the global formation pressure state variable matrix.

[0087] In this embodiment, the calculation process of the global variable matrix G of formation pressure state can be expressed as follows:

[0088] G = Z1 + Z2

[0089] Step S504: Perform a nonlinear transformation on the global variable matrix of formation pressure state to obtain the global mapping matrix of formation pressure state.

[0090] In this embodiment, the calculation process of the global mapping matrix F of formation pressure state can be expressed as follows:

[0091] F = tanh(G)

[0092] Here, tanh() represents the hyperbolic tangent function, used to implement nonlinear transformations.

[0093] Step S505: Perform a dot product operation on the third formation pressure state weight matrix and the formation pressure state global mapping matrix to obtain the target formation pressure matrix.

[0094] In this embodiment, the calculation process of the target formation pressure matrix R can be expressed as follows:

[0095] R=Y3 F

[0096] The formation pressure prediction method for high water injection conditions provided in this application strengthens important features in the initial formation pressure matrix through a weight vector, ensuring that important features in the initial formation pressure values ​​are considered in the prediction calculation. It also weights possible offsets in the formation pressure values ​​using a formation pressure offset weight matrix to highlight the most likely offsets, thereby improving the accuracy of formation pressure prediction for high water injection conditions. This ensures that personnel can effectively judge the water injection situation based on the predicted formation pressure values. Furthermore, by introducing a time state matrix, the predicted formation pressure values ​​can be combined with different time steps, facilitating visualization of formation pressure changes over time and improving the efficiency of taking countermeasures when abnormal changes in formation pressure occur.

[0097] Figure 6 The following is a flowchart illustrating the implementation of the formation pressure prediction method for high water injection conditions provided in Embodiment Six of this application. The difference between this method and Embodiment One is that, after step S105, the method further includes:

[0098] Step S601: Based on the preset geostress calculation model, calculate the geostress value according to the target formation pressure value.

[0099] In this embodiment, the preset geostress calculation model can be the Mohr-Coulomb constitutive model in the Visitage simulator. The formation seepage model is coupled with the geomechanical model. By simulating the formation seepage process, the geostress value that changes continuously during the production water reinjection process is calculated based on the target formation pressure value.

[0100] Step S602: Generate the stress distribution variation law based on the stress value and the preset time step information.

[0101] In this embodiment, the preset time step can be the injection time information of the reinjected water, which can be represented by multiple moments, and the time interval between these moments can be set manually. The geostress curve can be fitted using the Mohr-Coulomb constitutive model in the Visitage simulator. The fitted curve characterizes the spatial distribution of geostress and its variation with injection time. The variation law of geostress distribution can be represented by the fitted geostress curve.

[0102] Step S603: Based on the stress distribution variation law and the preset strain boundary conditions, the stress distribution field and stress concentration points are obtained.

[0103] In this embodiment, the Mohr-Coulomb constitutive model in the Visage simulator can be used to generate a multidimensional image of the fitted curve representing the in-situ stress. This generated multidimensional image represents the in-situ stress distribution field, used to characterize the changes in in-situ stress distribution during water injection. The preset strain boundary conditions can be determined by manually adjusting the strain parameters, which can be the static compressive elastic modulus, specifically the maximum value of the static compressive elastic modulus, Eh. max =0.0009 and the minimum value of static compressive elastic modulus Eh min =0.0018, which can be used to meet the minimum horizontal principal stress equivalent density of 1.5 g / cm³ near the reinjection well. 3 The maximum principal stress equivalent density is 1.9 g / cm³. 3 Vertical stress equivalent density 2.1 g / cm³ 3 Within the range defined by two threshold values ​​for the static compressive modulus of elasticity, the point of maximum stress in the stress distribution field can be identified as a stress concentration point. Understandably, the formation is most prone to fracturing at stress concentration points. Therefore, steel plates need to be added around the formation at these stress concentration points to prevent stress concentration from causing formation fracturing and affecting drilling operations.

[0104] The formation pressure prediction method for high water injection conditions provided in this application calculates the ground stress value based on the formation pressure value using a ground stress calculation model. Then, it generates the distribution of ground stress at different times according to the time scale of the water injection operation. By setting strain boundary conditions, it determines the maximum value of the ground stress value falling within the strain range and identifies the point with the maximum ground stress value as the ground stress concentration point. This allows for timely reinforcement of the ground stress concentration point, preventing it from rupturing due to the continuous water injection operation. This ensures that the production water reinjection process does not lead to formation rupture, effectively guaranteeing the safety and efficiency of drilling operations.

[0105] Figure 7 The flowchart illustrating the implementation of the formation pressure prediction method for high water injection conditions provided in Embodiment 7 of this application is shown. The difference between this method and Embodiment 6 above is that step S603 specifically includes:

[0106] Step S701: Calculate multiple principal stress equivalent densities based on the stress distribution variation law and preset strain boundary conditions.

[0107] In this embodiment, the principal stress equivalent density refers to the stress state at a point in three-dimensional space, which can be represented by a scalar. This scalar is the stress equivalent density. The principal stress equivalent density is an eigenvalue of the stress tensor, representing the magnitude of the maximum shear stress experienced by the rock in the formation at that point. The strain parameter can be the static compressive elastic modulus, and the maximum value of the static compressive elastic modulus is Eh.max =0.0009 and the minimum value of static compressive elastic modulus Eh min =0.0018, the preset strain boundary condition is the range formed by the maximum and minimum values ​​of the static compressive elastic modulus, which can be calculated using the Mohr-Coulomb constitutive model in the Visitage simulator to obtain the minimum horizontal principal stress equivalent density of 1.5 g / cm³ near the reinjection well. 3 The maximum principal stress equivalent density is 1.9 g / cm³. 3 Vertical stress equivalent density 2.1 g / cm³ 3 .

[0108] Step S702: Obtain the geostress distribution field based on the multiple principal stress equivalent densities.

[0109] In this embodiment, the spatial location information of multiple principal stress equivalent densities can be calibrated, and then the location can be reconstructed based on the calibrated spatial location information to obtain the geostress distribution field, which is used to characterize the geostress distribution in three-dimensional space.

[0110] Step S703: Obtain the stress concentration point based on the maximum value of the principal stress equivalent density.

[0111] In this embodiment, the point with the highest ground stress value is identified as the ground stress concentration point, so that the ground stress concentration point can be reinforced in a timely manner to prevent the ground stress concentration point from cracking due to the continuous water injection operation.

[0112] The formation pressure prediction method for high water injection conditions provided in this application determines the maximum value of the geostress within the strain range and identifies the point with the maximum geostress value as the geostress concentration point. This allows for timely reinforcement of the geostress concentration point, preventing it from rupturing due to continuous water injection during the production water reinjection process. This effectively ensures the safety and efficiency of drilling operations.

[0113] Corresponding to the method in the above embodiments, Figure 8 The diagram shows a structural block diagram of a formation pressure prediction device for high water injection conditions provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiments of this application are shown. Figure 8 The example formation pressure prediction device for high water injection conditions can be the execution subject of the formation pressure prediction method for high water injection conditions provided in the aforementioned embodiment 1.

[0114] Reference Figure 8 The formation pressure prediction device for high water injection conditions includes:

[0115] The information acquisition module 810 is used to acquire water injection speed information, water injection time information, and suspended particle concentration information.

[0116] The water injection volume and formation permeability determination module 820 is used to determine the water injection volume and formation permeability based on the water injection rate information, water injection time information and suspended particle concentration information.

[0117] The initial formation pressure value calculation module 830 is used to perform multi-dimensional calculations of formation pressure based on the water injection volume and formation permeability to obtain initial formation pressure values ​​in multiple dimensions.

[0118] Formation pressure matrix generation module 840 is used to generate multiple formation pressure matrices based on the initial formation pressure values ​​of the multiple dimensions.

[0119] The target formation pressure value calculation module 850 is used to predict and calculate the formation pressure matrix based on a preset weight vector, a nonlinear mapping function, and a gating vector to obtain the target formation pressure value.

[0120] The process by which each module in the formation pressure prediction device for high water injection conditions provided in this application implements its respective function can be found in the foregoing. Figure 1 The description of Embodiment 1 shown will not be repeated here.

[0121] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0122] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0123] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0124] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0125] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first," "second," etc., are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, a first table may be named a second table, and similarly, a second table may be named a first table, without departing from the scope of the various described embodiments. Both the first table and the second table are tables, but they are not the same table.

[0126] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0127] The formation pressure prediction method for high water injection conditions provided in this application can be applied to terminal devices such as mobile phones, tablets, wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPCs), netbooks, and personal digital assistants (PDAs). This application does not impose any restrictions on the specific type of terminal device.

[0128] For example, the terminal device may be a station (STAION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, a set-top box (STB), customer premises equipment (CPE), and / or other devices used for communication over a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks or mobile terminals in future evolved Public Land Mobile Network (PLMN) networks.

[0129] As an example and not a limitation, when the terminal device is a wearable device, the term "wearable device" can also refer to any device that utilizes wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices worn directly on the body or integrated into a user's clothing or accessories. Wearable devices are not merely hardware devices; they achieve powerful functions through software support, data interaction, and cloud interaction. Broadly defined, wearable smart devices include those with comprehensive functions, large sizes, and the ability to perform complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those focused on a specific application function that require interaction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0130] Figure 9 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For example... Figure 9 As shown, the terminal device 9 of this embodiment includes: at least one processor 90 ( Figure 9 (Only one is shown in the image), a memory 91, which stores a computer program 92 that can run on the processor 90. When the processor 90 executes the computer program 92, it implements the steps in the above embodiments of the formation pressure prediction method for high water injection conditions, for example... Figure 1 Steps S101 to S105 are shown. Alternatively, when the processor 90 executes the computer program 92, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 8The functions of modules 810 to 850 are shown.

[0131] The terminal device 9 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The terminal device may include, but is not limited to, a processor 90 and a memory 91. Those skilled in the art will understand that... Figure 9 This is merely an example of terminal device 9 and does not constitute a limitation on terminal device 9. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input transmission devices, network access devices, buses, etc.

[0132] The processor 90 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0133] In some embodiments, the memory 91 may be an internal storage unit of the terminal device 9, such as a hard disk or memory of the terminal device 9. The memory 91 may also be an external storage device of the terminal device 9, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device 9. Furthermore, the memory 91 may include both internal and external storage units of the terminal device 9. The memory 91 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 91 can also be used to temporarily store data that has been sent or will be sent.

[0134] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0135] This application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it causes the terminal device to implement the steps in any of the above method embodiments.

[0136] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0137] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0138] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0139] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0140] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0141] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0142] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for predicting formation pressure under high water injection conditions, characterized in that, include: Acquire information on water injection rate, water injection time, and suspended particulate concentration; Based on the water injection rate information, water injection time information, and suspended particle concentration information, the water injection volume and formation permeability are determined. Based on the injected water volume and formation permeability, the formation pressure is calculated in multiple dimensions to obtain initial formation pressure values ​​in multiple dimensions. The initial formation pressure values ​​include formation pressure data values, formation pressure dimension values, and formation pressure dimension axis identifiers. Specifically, by analyzing the time-varying law of formation permeability, the pressure distribution characteristics of each block of rock are obtained. The pressure distribution characteristics are extracted to obtain initial formation pressure values ​​in multiple dimensions, including the horizontal dimension, the vertical dimension, and the depth dimension. Based on the initial formation pressure values ​​of the multiple dimensions, multiple formation pressure matrices are generated; Based on a preset weight vector, nonlinear mapping function, and gate vector, the formation pressure matrix is ​​predicted and calculated to obtain the target formation pressure value. The step of predicting and calculating the formation pressure matrix based on a preset weight vector, a nonlinear mapping function, and a gating vector to obtain the target formation pressure value specifically includes: The formation pressure matrix, the preset weight vector, and the preset nonlinear mapping function are convolved to obtain the formation pressure feature matrix. Specifically, the formation pressure matrix and the weight vector are convolved first, and then all the values ​​of the result of the convolution operation are used as the independent variables of the nonlinear mapping function to calculate the function value of the nonlinear mapping function. Finally, a matrix is ​​generated from all the calculated function values ​​of the nonlinear mapping function. A global representation feature matrix of formation pressure is generated based on the maximum value of all values ​​in the formation pressure feature matrix. Based on a preset nonlinear mapping function and multiple preset gate vectors, the global representation feature matrix of formation pressure is weighted and summed to obtain multiple formation pressure state weight matrices. Based on a preset nonlinear mapping function and a preset gating vector, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the formation pressure offset weight matrix. The target formation pressure matrix is ​​calculated by multiplying the formation pressure state weight matrix and the formation pressure offset weight matrix by a dot product. The target formation pressure value is obtained based on the target formation pressure matrix. The gating vector includes a first gating vector, a second gating vector, and a third gating vector; The formation pressure state weight matrix includes a first formation pressure state weight matrix, a second formation pressure state weight matrix, and a third formation pressure state weight matrix. The step of performing a weighted summation calculation on the global representation feature matrix of formation pressure based on a preset nonlinear mapping function and a preset gating vector to obtain multiple formation pressure state weight matrices specifically includes: Based on the first gating vector, the preset nonlinear mapping function, and the preset time state matrix, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the first formation pressure state weight matrix. Specifically, the global representation feature matrix of formation pressure is denoted as X, the first gating vector is denoted as E1, the preset nonlinear mapping function is denoted as δ(), the time state matrix is ​​denoted as T, and the process of calculating the first formation pressure state weight matrix Y1 is as follows: ； Based on the second gating vector, the preset nonlinear mapping function, and the preset time state matrix, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the second formation pressure state weight matrix; specifically, the second gating vector is denoted as E2, and the process of calculating the second formation pressure state weight matrix Y2 is as follows: ； Based on the third gating vector, the preset nonlinear mapping function, and the preset time state matrix, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the third formation pressure state weight matrix; specifically, the third gating vector is represented as E3, and the process of calculating the third formation pressure state weight matrix Y3 is as follows: 。 2. The formation pressure prediction method for high water injection conditions as described in claim 1, characterized in that, The step of generating multiple formation pressure matrices based on the initial formation pressure values ​​of the multiple dimensions specifically includes: Based on the formation pressure data values ​​and formation pressure dimension values, generate an array of formation pressure data values ​​and an array of formation pressure dimension values. Based on the formation pressure dimension axis identifier, the formation pressure data value array and the formation pressure dimension value array are concatenated to obtain multiple one-dimensional arrays of formation pressure values; Multiple formation pressure matrices are generated based on the one-dimensional array of formation pressure values, a preset number of matrix rows, and a preset number of matrix columns.

3. The formation pressure prediction method for high water injection conditions as described in claim 1, characterized in that, The step of calculating the target formation pressure matrix based on the formation pressure state weight matrix and the formation pressure offset weight matrix specifically includes: The first formation pressure state weight matrix and the formation pressure offset weight matrix are multiplied by a dot product to obtain the first formation pressure state variable matrix. The second formation pressure state weight matrix and the preset time state matrix are multiplied by a dot product to obtain the second formation pressure state variable matrix. The first formation pressure state variable matrix and the second formation pressure state variable matrix are summed to obtain the global formation pressure state variable matrix. A nonlinear transformation is performed on the global variable matrix of formation pressure state to obtain the global mapping matrix of formation pressure state; The target formation pressure matrix is ​​obtained by performing a dot product operation on the third formation pressure state weight matrix and the formation pressure state global mapping matrix.

4. The formation pressure prediction method for high water injection conditions as described in claim 1, characterized in that, After predicting and calculating the formation pressure matrix based on a preset weight vector, a nonlinear mapping function, and a gating vector to obtain the target formation pressure value, the method further includes: Based on a preset geostress calculation model, the geostress value is calculated according to the target formation pressure value; Based on the aforementioned geostress value and the preset time step information, a geostress distribution variation law is generated; Based on the stress distribution variation law and the preset strain boundary conditions, the stress distribution field and stress concentration points are obtained.

5. The formation pressure prediction method for high water injection conditions as described in claim 4, characterized in that, The step of obtaining the geostress distribution field and geostress concentration points based on the geostress distribution variation law and the preset strain boundary conditions specifically includes: Based on the aforementioned stress distribution variation law and the preset strain boundary conditions, multiple principal stress equivalent densities are calculated. The geostress distribution field is obtained based on the multiple principal stress equivalent densities. The stress concentration point is obtained based on the maximum value of the principal stress equivalent density.

6. A formation pressure prediction device for high water injection conditions, characterized in that, include: The information acquisition module is used to acquire information on water injection rate, water injection time, and suspended particle concentration. The water injection volume and formation permeability determination module is used to determine the water injection volume and formation permeability based on the water injection rate information, water injection time information and suspended particle concentration information. The initial formation pressure calculation module is used to perform multi-dimensional calculations of formation pressure based on the injected water volume and formation permeability to obtain initial formation pressure values ​​in multiple dimensions. The initial formation pressure values ​​include formation pressure data values, formation pressure dimension values, and formation pressure dimension axis identifiers. Specifically, by analyzing the time-varying law of formation permeability, the pressure distribution characteristics of each block of rock are obtained, and the pressure distribution characteristics are extracted to obtain initial formation pressure values ​​in multiple dimensions, including the horizontal dimension, the vertical dimension, and the depth dimension. The formation pressure matrix generation module is used to generate multiple formation pressure matrices based on the initial formation pressure values ​​of the multiple dimensions. The target formation pressure value calculation module is used to predict and calculate the formation pressure matrix based on a preset weight vector, a nonlinear mapping function, and a gate vector to obtain the target formation pressure value. The step of predicting and calculating the formation pressure matrix based on a preset weight vector, a nonlinear mapping function, and a gating vector to obtain the target formation pressure value specifically includes: The formation pressure matrix, the preset weight vector, and the preset nonlinear mapping function are convolved to obtain the formation pressure feature matrix. Specifically, the formation pressure matrix and the weight vector are convolved first, and then all the values ​​of the result of the convolution operation are used as the independent variables of the nonlinear mapping function to calculate the function value of the nonlinear mapping function. Finally, a matrix is ​​generated from all the calculated function values ​​of the nonlinear mapping function. A global representation feature matrix of formation pressure is generated based on the maximum value of all values ​​in the formation pressure feature matrix. Based on a preset nonlinear mapping function and multiple preset gate vectors, the global representation feature matrix of formation pressure is weighted and summed to obtain multiple formation pressure state weight matrices. Based on a preset nonlinear mapping function and a preset gating vector, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the formation pressure offset weight matrix. The target formation pressure matrix is ​​calculated by multiplying the formation pressure state weight matrix and the formation pressure offset weight matrix by a dot product. The target formation pressure value is obtained based on the target formation pressure matrix. The gating vector includes a first gating vector, a second gating vector, and a third gating vector; The formation pressure state weight matrix includes a first formation pressure state weight matrix, a second formation pressure state weight matrix, and a third formation pressure state weight matrix. The step of performing a weighted summation calculation on the global representation feature matrix of formation pressure based on a preset nonlinear mapping function and a preset gating vector to obtain multiple formation pressure state weight matrices specifically includes: Based on the first gating vector, the preset nonlinear mapping function, and the preset time state matrix, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the first formation pressure state weight matrix. Specifically, the global representation feature matrix of formation pressure is denoted as X, the first gating vector is denoted as E1, the preset nonlinear mapping function is denoted as δ(), the time state matrix is ​​denoted as T, and the process of calculating the first formation pressure state weight matrix Y1 is as follows: ； Based on the second gating vector, the preset nonlinear mapping function, and the preset time state matrix, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the second formation pressure state weight matrix; specifically, the second gating vector is denoted as E2, and the process of calculating the second formation pressure state weight matrix Y2 is as follows: ； Based on the third gating vector, the preset nonlinear mapping function, and the preset time state matrix, a nonlinear calculation is performed on the global representation feature matrix of formation pressure to obtain the third formation pressure state weight matrix; specifically, the third gating vector is represented as E3, and the process of calculating the third formation pressure state weight matrix Y3 is as follows: 。 7. A terminal device, characterized in that, The terminal device includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 5.

Images