A method for preparing a temperature-resistant and salt-resistant nano-micro gel

Abstract

This invention provides a method for preparing temperature- and salt-resistant nano-scale gels, belonging to the field of nano-scale gel technology. The invention involves formulating a gel composition containing polyacrylamide and phenolic resin crosslinking agent, using a thermal stability analysis model to predict optimal gelation conditions for crosslinking reaction to form the gel body, employing a shear parameter game optimization model to guide the selection of mechanical shearing strategies, calculating precise shear parameters through the shear stress transfer equation to implement granulation stability treatment, obtaining nano-scale gel particles, and conducting temperature and salt resistance performance tests in a high-temperature, high-salt environment. Stability is assessed using a particle aggregation stability matrix, and performance is evaluated and optimized using temperature and salt resistance evaluation coefficients. Finally, product quality control is completed to ensure the long-term stability of the particles under extreme conditions, solving the technical problem of difficult stability control of nano-scale gel particles in high-temperature, high-salt environments.

Description

Technical Field

[0001] This invention belongs to the field of nano-micro gel technology, and more specifically, relates to a method for preparing a temperature-resistant and salt-resistant nano-micro gel. Background Technology

[0002] Nanoscale gels, as important functional materials, have broad application prospects in oilfield water shut-off and flood control, soil improvement, and environmental remediation. Traditional nanoscale gel preparation techniques mainly employ simple chemical cross-linking methods. These methods involve the reaction of polymers with cross-linking agents to form a three-dimensional network structure, followed by mechanical shearing to break down large gel blocks into nanoscale particles. This method can achieve gel particles with a certain particle size distribution under normal environmental conditions. However, traditional preparation processes lack precise component distribution control mechanisms. Uneven distribution of components during the cross-linking reaction leads to localized defects in the resulting gel network structure. Stress concentration during mechanical shearing easily results in uneven particle size distribution and irregular shapes. More importantly, nanoscale gel particles prepared by traditional methods are prone to structural damage and performance failure under extreme environmental conditions such as high temperature and high salinity due to insufficient network structure stability and uncontrollable particle aggregation behavior, failing to meet the long-term stability requirements of complex application environments. Summary of the Invention

[0003] In view of this, the present invention provides a method for preparing temperature-resistant and salt-resistant nano-scale gel, which can solve the technical problem of difficulty in controlling the stability of nano-scale gel particles under high temperature and high salt environment in the prior art.

[0004] This invention is achieved as follows: It provides a method for preparing a temperature- and salt-resistant nano-scale gel. The method controls component distribution through a crosslinking uniform distribution matrix, predicts reaction conditions using a thermal stability analysis model, guides shear strategies using a shear parameter game optimization model, and achieves precise control by evaluating stability through a particle aggregation stability matrix. The method includes a gel composition preparation step, where polyacrylamide and a phenolic resin crosslinking agent are mixed, simulated formation water is added, and a crosslinking uniform distribution matrix is ​​constructed to optimize component distribution. The component uniformity coefficient is output to guide the adjustment of mixing process parameters. A gelation reaction step is then performed, where the optimal reaction temperature and reaction time are predicted using a thermal stability analysis model. The process of cross-linking reaction is monitored by a maximum tolerance temperature characteristic matrix to form a three-dimensional network structure of the gel body; a mechanical shearing treatment step is performed, and a shearing strategy selection is determined using a shear parameter game optimization model to output the optimal shearing strategy scheme; a granulation stability treatment step is implemented, and precise shear parameters are calculated through the shear stress transfer equation, and the shear intensity is controlled based on the nano-micro particle size distribution matrix to obtain nano-micro gel particles; a temperature and salt resistance performance test step is performed, and particle stability is evaluated using a particle aggregation stability matrix to output a particle stability index; a performance evaluation and optimization step is performed, and quantitative analysis is conducted through temperature and salt resistance evaluation coefficients; and the final product quality control step is completed.

[0005] Specifically, the step of preparing the gel composition involves mixing polyacrylamide with a relative molecular mass of 6 million to 12 million and a degree of hydrolysis of 3% to 6% with a phenolic resin crosslinking agent at a mass fraction of 0.3% to 0.6% and 0.4% to 0.9%, respectively, and adding simulated formation water with a mineralization of 5000 mg / L to 300000 mg / L.

[0006] Specifically, the gelation reaction step involves placing the gel composition in a temperature environment of 50°C to 120°C, using a thermal stability analysis model to predict the optimal reaction temperature and reaction time based on the component uniformity coefficient output by the polyacrylamide molecular weight, phenolic resin crosslinking agent concentration, simulated formation water salinity, and crosslinking uniform distribution matrix, and controlling the gelation process according to the predicted optimal reaction temperature and reaction time.

[0007] Specifically, the mechanical shearing process involves continuously shearing the gel bulk using a colloid mill, a high-speed shearing machine, or a pipeline shearing crosslinking method. A shearing strategy selection is determined based on the gel bulk viscosity, crosslinking density, and target particle size range using a shearing parameter game optimization model.

[0008] Specifically, the granulation stability treatment step involves calculating the optimal shear strategy based on the uniform shear stress matrix and the shear parameter game optimization model, using the shear stress transfer equation to ensure the integrity of the particle shape and the uniformity of the particle size, thereby obtaining nano-micro gel particles with an average particle size of 0.3 μm to 6.4 μm.

[0009] Specifically, the temperature and salt resistance test involves aging the prepared nano-sized gel particles in a high-temperature environment of 80°C to 180°C and a high-salinity environment of 50,000 mg / L to 300,000 mg / L for 10 to 15 days, and then evaluating the stability of the particles under extreme conditions using a particle aggregation stability matrix.

[0010] Specifically, the performance evaluation and optimization steps involve quantitatively analyzing the performance of gel particles using the particle stability index output from the particle aggregation stability matrix based on the temperature and salt resistance evaluation coefficient. When the temperature and salt resistance evaluation coefficient is less than or equal to 0.4, it indicates excellent performance; when it is greater than 0.4 but less than or equal to 0.75, the formulation needs to be optimized; and when it is greater than 0.75, the gel particles need to be re-prepared.

[0011] Specifically, the final product quality control step involves observing the macroscopic morphology and analyzing the microstructure of the prepared nano-scale gel particles to ensure that the particles can effectively aggregate without network structure destruction under high temperature and high salt conditions, and that the stabilization time reaches more than 90 days.

[0012] Specifically, the crosslinking uniform distribution matrix is ​​used to describe the distribution of polyacrylamide molecular chains and phenolic resin crosslinking agents in three-dimensional space. The matrix elements characterize the crosslinking density at different positions, ensuring the uniform mixing of each component in the gel composition and the consistency of the crosslinking reaction.

[0013] Specifically, the highest tolerance temperature feature matrix is ​​used to record the structural stability parameters of the gel body under different temperature conditions. The matrix rows represent temperature gradients, and the columns represent structural feature parameters. Matrix operations are used to monitor structural changes during the crosslinking reaction process.

[0014] Specifically, the shear parameter game optimization model includes an upper-level game model that aims to maximize shear efficiency and a lower-level game model that aims to optimize particle uniformity. The objective functions of the upper-level game model and the lower-level game model influence each other through shear stress coupling terms, reflecting the mutual constraint relationship between shear efficiency and particle uniformity.

[0015] Specifically, the shear efficiency maximization function is used to determine the optimal equipment configuration and operation strategy for generating nano- and micro-sized gel particles per unit time during mechanical shearing. The inputs include gel bulk viscosity, crosslinking density, target particle size range, equipment power limit, and shearing time constraint. The outputs are the selection of shearing equipment type and the basic shear rate range.

[0016] Specifically, the particle uniformity optimization function is used to formulate a control strategy for the uniformity of particle size distribution and shape regularity of nano- and micro-scale gel particles. The inputs include the target particle size range, gel bulk crosslinking density, shear strength gradient requirements, target value of particle shape factor, and uniformity standard deviation limit. The output is the shear uniformity control strategy.

[0017] The thermal stability analysis model is specifically a sequence prediction network based on the Transformer architecture, which includes a multi-head attention mechanism and a feedforward neural network layer. The number of heads in the multi-head attention mechanism is determined according to the number of components in the gel composition and the number of temperature monitoring points. The model parameters are updated using the cross-entropy loss function and the Adam optimizer.

[0018] Specifically, the temperature adaptability adjustment function is used to adjust the multi-head attention mechanism parameters of the thermal stability analysis model. The temperature sensitivity index is calculated based on the crosslinking density of the gel bulk, the simulated formation water salinity, the molecular weight of polyacrylamide, and the component uniformity coefficient. The prediction accuracy parameters of the model are adjusted by using different numbers of attention heads according to different ranges of the temperature sensitivity index.

[0019] Specifically, the particle aggregation stability matrix is ​​used to evaluate the aggregation behavior and structural stability of nano-scale gel particles in a high-temperature and high-salt environment. The matrix elements include the influence weights of multi-dimensional parameters such as temperature, salinity, and time on particle stability. The temperature and salt resistance evaluation coefficient is calculated by comprehensively considering the structural integrity, aggregation ability, and stability duration of nano-scale gel particles in a high-temperature and high-salt environment.

[0020] This invention addresses the shortcomings of traditional methods, such as uneven component distribution and difficulty in controlling structural stability, by establishing a uniform crosslinking distribution matrix to control the spatial distribution of the crosslinking agent between polyacrylamide and phenolic resin. It utilizes a thermal stability analysis model to predict optimal gelation conditions, employs a shear parameter game optimization model to guide the selection of mechanical shearing strategies, and evaluates structural stability under extreme conditions through a particle aggregation stability matrix. The multi-dimensional matrix analysis system established in this invention can precisely control the formation process of the gel network structure and granulation parameters, ensuring that nano- and micro-sized gel particles have a uniform particle size distribution and a complete network structure. Furthermore, it achieves quantitative performance analysis and closed-loop optimization of process parameters through temperature and salt resistance evaluation coefficients. In summary, this invention solves the technical problem mentioned in the background art of difficulty in controlling the stability of nano- and micro-sized gel particles under high temperature and high salt environments. Attached Figure Description

[0021] Figure 1 These are macroscopic, microscopic, and microscopic morphology images of micro / nano-scale gel particles with an average particle size of 0.6 μm prepared from phenolic resin gel in Example 1 of this invention, agglomerated at 165°C and 280,000 mg / L high temperature and high salt conditions.

[0022] Figure 2 These are macroscopic, microscopic, and microscopic morphology images of the phenolic resin gel particles with an average particle size of 4.2 μm prepared by nanoparticle-reinforced gel in Example 2 of the present invention, which aggregated at 155℃ and under high temperature and high salt conditions of 250000 mg / L.

[0023] Figure 3 These are macroscopic, microscopic, and microscopic morphology images of the hydroquinone-type phenolic resin gel particles with an average particle size of 0.7 μm prepared in Example 3 of the present invention, which aggregated at 170°C and 300,000 mg / L high temperature and high salt conditions.

[0024] Figure 4 These are macroscopic, microscopic, and microscopic morphology images of micro / nano-scale gel particles with an average particle size of 0.5 μm prepared from resorcinol-type phenolic resin gel in Example 4 of the present invention, agglomerated at 140℃ and 220000 mg / L high temperature and high salt conditions.

[0025] Figure 5 These are macroscopic, microscopic, and microscopic morphology images of the 0.8 μm micro-nano scale gel particles prepared from catechol-type phenolic resin gel in Example 5 of the present invention, agglomerated at 158°C and 275000 mg / L high temperature and high salt conditions.

[0026] Figure 6 These are macroscopic, microscopic, and microscopic morphology images of micro / nano-scale gel particles with an average particle size of 0.4 μm prepared from organic chromium gel in Example 6 of this invention, agglomerated at 85°C and 120,000 mg / L high temperature and high salt conditions.

[0027] Figure 7 This is the molecular dynamics model of micron-sized gel particles with an average particle size of 4.2 μm prepared from hydroquinone-type phenolic resin gel in Example 2 of the present invention under the conditions of 155℃ and 250,000 mg / L.

[0028] Figure 8 This is a molecular dynamics model of nanoscale gel particles with an average particle size of 0.6 μm prepared from hydroquinone-type phenolic resin gel in Example 1 of the present invention under conditions of 165℃ and 280,000 mg / L.

[0029] Figure 9 This is a graph showing the structural stability potential energy changes of hydroquinone-type phenolic resin gel particles with an average particle size of 4.2 μm (micrometer-sized) and 0.6 μm (nanometer-sized) prepared from hydroquinone-type phenolic resin gels in Examples 1 and 2 of the present invention under high temperature and high salt conditions.

[0030] Figure 10 These are macroscopic and microscopic morphological images of the three-dimensional network structure destruction under high temperature and high salt conditions for the following gels: phenolic resin gel with a size of 15–45 mm prepared in Comparative Example 1; nanoparticle-reinforced phenolic resin gel with a size of 18–42 mm prepared in Comparative Example 2; resorcinol-type phenolic resin gel with a size of 12–38 mm prepared in Comparative Example 3; catechol-type phenolic resin gel with a size of 16–44 mm prepared in Comparative Example 4; organochromium gel with a size of 10–35 mm prepared in Comparative Example 5; and composite crosslinking agent gel with a size of 14–40 mm prepared in Comparative Example 6.

[0031] Figure 11 This is a flowchart of the method of the present invention.

[0032] Figure 12 This is a schematic diagram of the neural network structure of the thermal stability analysis model involved in the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0034] like Figure 11 The diagram shown is a flowchart of a method for preparing a temperature-resistant and salt-resistant nano-microscale gel provided by the present invention. This method includes the following steps:

[0035] S01. Prepare a gel composition by mixing polyacrylamide with a relative molecular mass of 6 million to 12 million and a degree of hydrolysis of 3% to 6% with a phenolic resin crosslinking agent at a mass fraction of 0.3% to 0.6% and 0.4% to 0.9%, respectively, and adding simulated formation water with a mineralization of 5000 mg / L to 300000 mg / L. Construct a crosslinking uniform distribution matrix to optimize the component distribution, and output the component uniformity coefficient to guide the adjustment of mixing process parameters.

[0036] S02. Perform the gelation reaction by placing the gel composition in a temperature environment of 50°C to 120°C. Utilize the thermal stability analysis model to predict the optimal reaction temperature and reaction time based on the molecular weight of polyacrylamide, the concentration of phenolic resin crosslinking agent, the simulated formation water salinity, and the component uniformity coefficient output by the crosslinking uniformity distribution matrix. Monitor the crosslinking reaction process through the highest tolerance temperature characteristic matrix and control the gelation process according to the predicted optimal reaction temperature and reaction time to form a three-dimensional network structure gel body.

[0037] S03. Perform mechanical shearing treatment. Continuously shear the gel bulk using a colloid mill, high-speed shearing machine, or pipeline shearing crosslinking method. Utilize a shearing parameter game optimization model to determine the shearing strategy selection based on the gel bulk viscosity, crosslinking density, and target particle size range. The shearing parameter game optimization model includes an upper-level game model aimed at maximizing shearing efficiency and a lower-level game model aimed at optimizing particle uniformity. The optimal shearing strategy is output to guide the selection of shearing equipment and the setting of basic parameters.

[0038] S04. Implement particle stability treatment. Based on the optimal shear strategy scheme output by the uniform shear stress matrix and the shear parameter game optimization model, calculate the accurate shear parameters through the shear stress transfer equation, and control the shear intensity according to the nano-micro particle size distribution matrix to ensure the integrity of particle shape and uniformity of particle size, and obtain nano-micro gel particles with an average particle size of 0.3μm to 6.4μm.

[0039] S05. Conduct temperature and salt resistance tests. Place the prepared nano-scale gel particles in a high temperature environment of 80℃ to 180℃ and a high salinity environment of 50000mg / L to 300000mg / L for 10 to 15 days. Use the particle aggregation stability matrix to evaluate the stability of the particles under extreme conditions and output the particle stability index for performance evaluation.

[0040] S06. Performance evaluation and optimization: The performance of gel particles is quantitatively analyzed by using the temperature and salt resistance evaluation coefficients based on the particle stability index output by the particle aggregation stability matrix. When the temperature and salt resistance evaluation coefficients are less than or equal to 0.4, the performance is excellent. When the coefficients are greater than 0.4 and less than or equal to 0.75, the formulation needs to be optimized. When the coefficients are greater than 0.75, the gel particles need to be re-prepared.

[0041] S07. Complete the final product quality control, observe the macroscopic morphology and analyze the microstructure of the prepared nano-scale gel particles to ensure that the particles can effectively aggregate without network structure destruction under high temperature and high salt conditions, and the stabilization time reaches more than 90 days.

[0042] The crosslinking uniformity distribution matrix describes the distribution of polyacrylamide molecular chains and phenolic resin crosslinking agents in three-dimensional space. Matrix elements characterize the crosslinking density at different locations, ensuring uniform mixing of components and consistency of the crosslinking reaction in the gel composition. The maximum tolerance temperature characteristic matrix records the structural stability parameters of the gel matrix under different temperature conditions. Matrix rows represent temperature gradients, and columns represent structural characteristic parameters. Matrix operations monitor structural changes during the crosslinking reaction. The nanoparticle size distribution matrix statistically analyzes and controls the particle size distribution of nano-scale gel particles after shearing. Matrix elements record the proportion of particles in different size ranges, guiding precise control of shear strength. The uniform shear stress matrix describes the stress distribution generated at different locations by the shearing equipment, ensuring uniform stress on the gel matrix during shearing and avoiding uneven breakage caused by localized stress concentration. The particle aggregation stability matrix evaluates the aggregation behavior and structural stability of nano-scale gel particles in high-temperature and high-salt environments. Matrix elements include the influence weights of multi-dimensional parameters such as temperature, salinity, and time on particle stability.

[0043] The upper-level game model in the shear parameter game optimization model aims to maximize shear efficiency. Its objective function is a shear efficiency maximization function, whose inputs include gel bulk viscosity, crosslinking density, target particle size range, equipment power limitations, and shear time constraints. Its outputs are the selection of shearing equipment type and the basic shear rate range. The lower-level game model aims to optimize particle uniformity. Its objective function is a particle uniformity optimization function, whose inputs include the target particle size range, gel bulk crosslinking density, shear strength gradient requirements, target particle shape factor value, and uniformity standard deviation limits. Its output is the shear uniformity control strategy. The objective functions of the upper and lower-level game models influence each other through a shear stress coupling term, which reflects the mutual constraint between shear efficiency and particle uniformity. The constraints of the upper-level game model include shear rate range constraints, equipment power limitations, and temperature control requirements. The constraints of the lower-level game model include particle size distribution range limitations, particle integrity requirements, and network structure maintenance conditions.

[0044] The shear efficiency maximization function is used to determine the optimal equipment configuration and operation strategy for generating nano- and micro-sized gel particles per unit time during mechanical shearing. Inputs include gel bulk viscosity, crosslinking density, target particle size range, equipment power limits, and shearing time constraints. Outputs are the selection of shearing equipment type and the basic shear rate range. The particle uniformity optimization function is used to formulate control strategies for the uniformity of particle size distribution and shape regularity of nano- and micro-sized gel particles. Inputs include the target particle size range, gel bulk crosslinking density, shear strength gradient requirements, target particle shape factor value, and uniformity standard deviation limits. Outputs are the shear uniformity control strategy.

[0045] The shear stress transfer equation is used to calculate the precise stress distribution and transfer law inside the gel during mechanical shearing. The inputs include the basic shear rate range, gel viscosity, crosslinking density, shear temperature, and uniform shear stress matrix parameters output by the shear parameter game optimization model. The outputs are the precise shear rate value and shear time value.

[0046] The thermal stability analysis model is structured as a sequence prediction network based on the Transformer architecture, including a multi-head attention mechanism and a feedforward neural network layer. The number of heads in the multi-head attention mechanism is determined based on the number of components in the gel composition and the number of temperature monitoring points. The steps for establishing the training dataset for the thermal stability analysis model include collecting experimental data on the stability of different formulation gel compositions under various temperature conditions, recording the correspondence between the type of phenolic resin crosslinking agent, the molecular weight of polyacrylamide, the simulated formation water salinity, the reaction temperature, the reaction time, and the final temperature resistance performance, and constructing a training set containing 10,000 sets of experimental data and a test set containing 2,000 sets of validation data. The training steps for the thermal stability analysis model include updating the model parameters using the cross-entropy loss function and the Adam optimizer, setting the learning rate to 0.001, the batch size to 32, and the training epochs to 200. The model convergence is monitored through the validation set performance to prevent overfitting.

[0047] The temperature-adaptive adjustment function is used to adjust the multi-head attention mechanism parameters of the thermal stability analysis model. The temperature-adaptive adjustment function is calculated based on the crosslinking density of the gel bulk, the simulated formation water salinity, the molecular weight of polyacrylamide, and the component uniformity coefficient to obtain a temperature sensitivity index. When the temperature sensitivity index belongs to different ranges, different numbers of attention heads are used to adjust the prediction accuracy parameters of the model. When the temperature sensitivity index a∈[0, 0.25), 4 attention heads are used; when a∈[0.25, 0.5), 6 attention heads are used; when a∈[0.5, 0.75), 8 attention heads are used; and when a∈[0.75, 1.0], 12 attention heads are used.

[0048] The temperature and salt resistance evaluation coefficient is calculated by comprehensively considering the structural integrity, aggregation ability, and stability duration of nano-scale gel particles in high-temperature and high-salt environments. This coefficient reflects the overall performance level of the nano-scale gel particles, providing a quantitative basis for optimizing the preparation process. The shear stress coupling term characterizes the degree of interaction between shear efficiency and particle uniformity. Increased shear efficiency affects the selection of particle uniformity control strategies, and vice versa. The component uniformity coefficient quantifies the consistency of component distribution in the gel composition, guiding the optimization of gelation reaction temperature and time. The optimal shearing strategy includes the selection of shearing equipment type, the basic shear rate range, and the shear uniformity control strategy, providing a guiding framework for accurate shear parameter calculation. The particle stability index quantifies the structural stability of nano-scale gel particles under extreme conditions, serving as a key input parameter for performance evaluation. The optimal reaction temperature and reaction time are determined based on the prediction results of the thermal stability analysis model and are used to control the gelation process to obtain the optimal gel bulk structure. The precise shear rate and shear time values ​​are used to directly control the operating parameters of the mechanical shearing equipment, ensuring the acquisition of nano-scale gel particles of predetermined specifications.

[0049] The specific implementation methods of the above steps are described in detail below.

[0050] The specific implementation of step S01 involves constructing an optimized gel composition base system by precisely controlling the proportions and uniformity of each component. First, polyacrylamide with a relative molecular mass of 6 million to 12 million is dissolved at a mass fraction of 0.3% to 0.6% in simulated formation water with a mineralization of 5000 mg / L to 300000 mg / L. A stepwise dissolution method is used, with thorough stirring at 60°C to 80°C for 30 to 60 minutes to ensure complete expansion of the polymer molecular chains. Subsequently, phenolic resin crosslinking agent is added at a mass fraction of 0.4% to 0.9%, using a batch-by-batch addition method to avoid pre-crosslinking caused by excessively high local concentrations. During this process, a crosslinking uniform distribution matrix algorithm is used to monitor and optimize the component distribution in real time. This matrix, based on the principle of three-dimensional spatial grid division, divides the reaction vessel space into several micro-grids. Concentration gradient detection technology is used to monitor the concentration distribution of polyacrylamide and phenolic resin crosslinking agent within each grid in real time, and the component uniformity coefficient is calculated. When the component uniformity coefficient is greater than 0.85, the distribution is considered uniform. When it is less than 0.85, the stirring rate and stirring time need to be adjusted. The stirring rate can be adjusted from 100 rpm to 300 rpm. The purpose of this step is to establish a basic system of gel composition with uniform component distribution and optimal crosslinking reactivity, providing an ideal precursor for subsequent gelation reactions.

[0051] The specific implementation of step S02 involves precise control of the gelation reaction process guided by an intelligent thermal stability analysis model. The prepared gel composition is transferred to a constant-temperature reactor, and a thermal stability analysis model based on the Transformer architecture is used to predict and optimize reaction parameters. This model uses the molecular weight of polyacrylamide, the concentration of the phenolic resin crosslinking agent, the simulated formation water salinity, and the component uniformity coefficient output from step S01 as input parameters. It captures the complex relationships between these parameters through a multi-head attention mechanism, outputting the predicted optimal reaction temperature and reaction time. The reaction temperature control accuracy is ±2℃, and the reaction time control accuracy is ±0.5 hours. Simultaneously, a maximum tolerance temperature characteristic matrix is ​​used to monitor the crosslinking reaction process in real time. This matrix collects temperature data at various points in the reaction system through a temperature sensor array and judges the degree of crosslinking reaction progress by combining the changes in rheological parameters. The crosslinking reaction is considered basically complete when the system viscosity reaches 8 to 12 times the initial viscosity. During the reaction, the temperature is maintained within the range of 50℃ to 120℃, with a preferred temperature of 85℃ to 95℃, and the reaction time is 4 to 8 hours. The purpose of this step is to form a gel matrix with ideal three-dimensional network structure density and thermal stability, providing a suitable raw material basis for subsequent mechanical shearing processing.

[0052] The specific implementation of step S03 involves employing a two-layer game theory optimization strategy to achieve intelligent control of the mechanical shearing process. First, a shearing parameter game theory optimization model is used to optimize the shearing strategy. This model adopts a two-layer game theory algorithm architecture. The upper-layer game model takes maximizing shearing efficiency as its objective function. Input parameters include gel bulk viscosity, crosslinking density, target particle size range, equipment power limit, and shearing time constraint. The Nash equilibrium solution algorithm determines the selection of the shearing equipment type and the basic shearing rate range. The lower-layer game model takes optimizing particle uniformity as its objective function. Input parameters include the target particle size range, gel bulk crosslinking density, shear strength gradient requirement, particle shape factor target value, and uniformity standard deviation limit. The output is a shear uniformity control strategy. The two models influence and constrain each other through a shear stress coupling term. When shearing efficiency increases, the weight allocation of the particle uniformity optimization strategy is affected through the coupling term. Based on the model output, an appropriate shearing method is selected. The colloid mill method is suitable for processing high-viscosity gel bulks, with a shear rate controlled between 8000 and 15000 rpm. The high-speed shearing machine method is suitable for medium-viscosity gel bulks, with a shear rate between 3000 and 8000 rpm. The pipeline shearing crosslinking method is suitable for continuous production, with the shear rate controlled by the pipeline diameter and flow rate. The purpose of this step is to efficiently break down large gel bulks into nano- and micro-sized particles with controllable particle size, while maintaining particle shape regularity and uniform particle size distribution.

[0053] The specific implementation of step S04 involves achieving precise control of particle stability through multi-dimensional stress field analysis and accurate parameter calculation. Based on the optimal shear strategy and uniform shear stress matrix parameters output by the shear parameter game optimization model in step S03, accurate shear parameters are calculated using the shear stress transfer equation. This equation is based on the stress tensor analysis theory in fluid mechanics, comprehensively considering the viscoelastic characteristics of the gel bulk, the influence of shear temperature on material properties, and the influence of the geometric parameters of the shearing equipment on stress distribution. Input parameters include the basic shear rate range, gel bulk viscosity, crosslinking density, shear temperature, and uniform shear stress matrix parameters. The output is accurate shear rate and shear time values. The shear rate calculation accuracy reaches ±50 rpm, and the shear time calculation accuracy is ±0.1 hours. Simultaneously, a nano-micro particle size distribution matrix is ​​used to monitor the shearing process in real time. This matrix uses laser particle size analysis technology to detect the particle size distribution in real time. When the standard deviation of the particle size distribution is less than 0.8 μm and the average particle size is within the target range of 0.3 μm to 6.4 μm, the control requirements are considered met. The shear strength is dynamically adjusted based on real-time particle size distribution data, with an adjustment range of ±10%. The purpose of this step is to ensure the acquisition of nano- and micro-sized gel particles with precise and controllable particle size and good shape integrity, providing standardized test samples for subsequent temperature and salt resistance tests.

[0054] The specific implementation of step S05 involves comprehensive performance testing of nano-scale gel particles using extreme environment simulation and a multi-parameter stability matrix evaluation system. The prepared nano-scale gel particles were placed in high-temperature environments ranging from 80°C to 180°C and high-salinity environments ranging from 50,000 mg / L to 300,000 mg / L for aging tests, with aging times set from 10 to 100 days. The high-temperature aging test employed a programmed temperature increase method, with an initial temperature set at 80°C, increasing by 10°C every 24 hours until the target temperature was reached, and then maintaining the target temperature for the remaining aging time. The high-salinity aging test used saline solutions with different mineralization levels, the main components of which included sodium chloride, calcium chloride, magnesium chloride, and sodium sulfate, prepared according to the actual formation water ion composition ratio. During the aging process, a particle aggregation stability matrix was used to continuously monitor the particle state. This matrix is ​​based on the principle of multi-dimensional parameter monitoring, and the monitored parameters include particle size change, shape factor change, aggregation degree, settling velocity, and network structure integrity. The monitoring frequency was once every 6 hours, and data acquisition was achieved using microscopic observation combined with image analysis technology. When the average particle size change rate is less than 15%, the shape factor change rate is less than 20%, and the aggregation degree is less than 30%, the stability is considered good. The matrix outputs a particle stability index, which comprehensively reflects the overall stability performance of the particles under extreme conditions. The purpose of this step is to comprehensively evaluate the stability performance of nano- and micro-sized gel particles in practical application environments, providing reliable data support for performance optimization.

[0055] The specific implementation of step S06 involves establishing a quantitative performance evaluation system to scientifically analyze and optimize the performance of gel particles. Based on the particle stability index output from the particle aggregation stability matrix in step S05, a comprehensive performance evaluation is performed using a temperature and salt resistance evaluation coefficient calculation method. This coefficient calculation method adopts a weighted comprehensive evaluation principle, comprehensively considering three key performance indicators of nano- and micro-scale gel particles in high-temperature and high-salt environments: structural integrity, aggregation ability, and stability duration. The weights of each indicator are 0.4, 0.3, and 0.3, respectively. Structural integrity is evaluated through particle morphology retention rate and network structure damage degree; aggregation ability is evaluated through inter-particle interaction strength and aggregation rate; and stability duration is evaluated through effective functional retention time. A temperature and salt resistance evaluation coefficient of 0.4 or less indicates excellent performance, and the process can proceed directly to the next step. A coefficient greater than 0.4 but less than or equal to 0.75 requires formulation optimization. Optimization may include adjusting the polyacrylamide molecular weight to 8-10 million, increasing the phenolic resin crosslinking agent concentration to 0.6-0.8%, or optimizing shear parameters. A coefficient greater than 0.75 requires re-preparation and verification of raw material quality and process parameter settings. The purpose of this step is to establish scientific performance evaluation standards and optimization guidance mechanisms to ensure that the final product meets the expected technical performance requirements.

[0056] The specific implementation of step S07 involves comprehensive quality control of the final product through multi-scale morphology analysis and long-term stability verification. The prepared nano-scale gel particles are subjected to macroscopic morphology observation and microscopic structure analysis. Macroscopic morphology observation uses a stereomicroscope at 50x to 200x magnification to observe the overall shape, surface smoothness, and dispersion state between particles. Microscopic structure analysis uses a scanning electron microscope at 1000x to 10000x magnification to observe the internal network structure, crosslinking point distribution, and pore structure characteristics of the particles. Simultaneously, a long-term stability verification test is conducted under high temperature and high salt conditions, with the test conditions set at 120℃ and 150,000 mg / L salinity for more than 90 days. During the test, the particle aggregation behavior and network structure changes are detected every 10 days. Dynamic light scattering technology is used to monitor particle size distribution changes, and rheological testing is used to monitor viscoelasticity changes. When the particles can effectively aggregate under the set conditions, and the gel network structure formed after aggregation remains intact, with viscoelastic parameter changes of less than 20%, and a stability time of more than 90 days, the product is considered to be of qualified quality. The purpose of this step is to ensure that the final product has excellent adaptability to high temperature and high salt environment and long-term stability, meeting the stringent requirements of practical applications.

[0057] like Figure 12As shown, the thermal stability analysis model employs a deep sequence prediction network structure based on the Transformer architecture. This model comprises four core components: an input embedding layer, a multi-head attention mechanism layer, a feedforward neural network layer, and an output prediction layer. The input embedding layer converts input parameters such as polyacrylamide molecular weight, phenolic resin crosslinking agent concentration, simulated formation water salinity, and component homogeneity coefficient into high-dimensional feature vectors, with an embedding dimension of 512. The multi-head attention mechanism layer contains eight attention heads, each independently calculating the correlation weights between input features, capturing long-term dependencies and complex nonlinear mappings between parameters through a self-attention mechanism. The feedforward neural network layer uses a two-layer fully connected structure with 2048 hidden layer neurons using the ReLU activation function and 256 output layer neurons. The output prediction layer maps the output of the feedforward neural network layer to the predicted optimal reaction temperature and reaction time through a linear transformation. The model has approximately 12 million parameters and a computational complexity of O(n^2). 2d), where n is the length of the input sequence and d is the embedding dimension. The model training dataset establishment process includes four stages: data collection, data preprocessing, feature engineering, and data augmentation. In the data collection stage, stability experimental data of different gel compositions under various temperature conditions were collected through laboratory pilot-scale, pilot-scale, and industrial-scale experiments. This covered polyacrylamide molecular weights ranging from 4 million to 15 million, phenolic resin crosslinking agent concentrations ranging from 0.2% to 1.2%, simulated formation water salinity ranging from 1000 mg / L to 500000 mg / L, reaction temperatures ranging from 40℃ to 140℃, and reaction times ranging from 1 hour to 12 hours, collecting over 15,000 sets of experimental data. In the data preprocessing stage, the raw experimental data were cleaned, denoised, and standardized, outliers and missing values ​​were removed, and Z-score standardization was used to normalize the numerical features. In the feature engineering phase, derived feature variables were constructed, including the ratio of molecular weight to crosslinking agent concentration, the interaction term between mineralization and temperature, and the product term between time and temperature, etc., to increase the model's ability to represent complex relationships. In the data augmentation phase, interpolation and noise injection methods were used to expand the training samples, ultimately constructing a training set containing 12,000 training data sets, a validation set containing 2,000 validation data sets, and a test set containing 1,000 test data sets. Model training used the cross-entropy loss function to measure the difference between predicted and true values, and the Adam optimizer was used for parameter updates. The learning rate was set to 0.001 and dynamically adjusted using a cosine annealing strategy. The batch size was 32, and the training epochs were 200. An early stopping strategy was used during training to prevent overfitting; training was stopped when the validation set loss did not decrease for 10 consecutive epochs. Model convergence was evaluated by the prediction accuracy on the validation set; the model was considered convergent when the prediction accuracy reached above 92% and the validation set loss stabilized. The temperature-adaptive adjustment function dynamically adjusts the number of attention heads based on different temperature sensitivity indices. When the temperature sensitivity index *a* is in the range [0, 0.25), four attention heads are used to adapt to low-sensitivity scenarios; when *a* is in the range [0.25, 0.5), six attention heads are used to adapt to medium-sensitivity scenarios; when *a* is in the range [0.5, 0.75), eight attention heads are used to adapt to high-sensitivity scenarios; and when *a* is in the range [0.75, 1.0], twelve attention heads are used to adapt to extremely high-sensitivity scenarios. This adjustment mechanism can adaptively optimize the model's prediction accuracy according to actual operating conditions, improving the model's generalization ability in different application scenarios.

[0058] The key technical ideas of this invention are mainly reflected in three aspects: crosslinking uniform distribution matrix construction technology, shear parameter bilayer game optimization technology, and Transformer-based intelligent prediction technology. The crosslinking uniform distribution matrix construction technology, through three-dimensional spatial gridding analysis and real-time concentration gradient monitoring, achieves precise distribution control of polyacrylamide and phenolic resin crosslinking agents in the gel composition. Compared with traditional empirical formulation methods, this technology can quantitatively evaluate the uniformity of component distribution and provide real-time control guidance, significantly improving the consistency and stability of the gel network structure and avoiding network defects and performance fluctuations caused by uneven local component concentrations. The shear parameter bilayer game optimization technology adopts a strategy combining upper-layer shear efficiency maximization game and lower-layer particle uniformity optimization game. Through Nash equilibrium solution, it achieves synergistic optimization of shear efficiency and product quality. Compared with traditional single-objective optimization methods, this technology can ensure precise control of product particle size distribution while guaranteeing production efficiency, resolving the contradiction between high-efficiency shearing and high-quality products, and realizing intelligent decision-making in the preparation process. The Transformer-based intelligent prediction technology utilizes a multi-head attention mechanism to capture the complex relationships between process parameters and achieves accurate prediction of optimal reaction conditions through deep learning. Compared to traditional empirical formulas or simple mathematical models, this technology has stronger nonlinear mapping and generalization capabilities, adapting to variations in different formulations and operating conditions, and significantly improving the intelligence level and prediction accuracy of gelation reaction control. The synergistic effect of these three key technologies constructs a fully intelligent control system from raw material preparation and gelation reaction to mechanical shearing, realizing the precision, standardization, and automation of the nano-micro gel preparation process. Compared to traditional experience-dependent preparation methods, this synergistic technology system can significantly improve product quality stability, reduce production costs, and shorten process development cycles, providing a complete technical solution for the industrial production of temperature- and salt-resistant nano-micro gels.

[0059] It should be noted that in existing gel granulation technologies, the selection of mechanical shearing equipment and the setting of operating parameters mainly rely on empirical judgment, lacking in-depth analysis of the complex relationship between shearing efficiency and particle quality. This often results in contradictory situations such as high shearing efficiency but poor particle uniformity, or good particle quality but low production efficiency. This invention constructs a shearing parameter game optimization model, treating maximizing shearing efficiency and optimizing particle uniformity as dual objectives in a game. The upper-level game model focuses on equipment selection and basic parameter optimization, while the lower-level game model focuses on particle quality control strategies. These two aspects achieve dynamic equilibrium through shear stress coupling terms, thereby obtaining high-quality nano-micro gel particles with uniform particle size distribution while ensuring production efficiency. Traditional gelation reactions mainly use fixed temperature and time parameters, which cannot be dynamically adjusted according to different formulation combinations and environmental conditions. This leads to randomness and uncontrollability in the formed gel network structure, resulting in significant fluctuations in the temperature and salt resistance of the final product. This invention establishes a model based on...

[0060] The thermal stability analysis model based on the Transformer architecture captures the complex correlations between multiple factors such as polyacrylamide molecular weight, crosslinking agent concentration, and mineralization degree through a multi-head attention mechanism. It can accurately predict the optimal reaction temperature and time under different conditions. At the same time, a temperature adaptability adjustment function is introduced to dynamically adjust the model parameters according to the temperature sensitivity index, ensuring the stability of prediction accuracy under different formulation conditions. This enables precise control of the gel network structure formation process and ensures consistent product performance.

[0061] Specifically, the principle of this invention is as follows: This invention solves the technical problem of difficult stability control of nano- and micro-scale gel particles, mainly based on the synergistic mechanism of multi-level structural design and precise parameter control. First, the crosslinking uniform distribution matrix quantitatively describes the distribution state of polyacrylamide molecular chains and phenolic resin crosslinking agents in three-dimensional space, ensuring uniform mixing of each component at the molecular level. This results in the formation of a three-dimensional network structure with uniform density and minimal defects during the crosslinking reaction, laying a solid foundation for the structural stability of the particles under extreme environments. The thermal stability analysis model uses a Transformer architecture sequence prediction network, which captures the complex correlation of network structure changes under different temperature conditions through a multi-head attention mechanism. It can accurately predict the optimal reaction temperature and time, avoiding the blindness of traditional empirical preparation methods and ensuring that the gel network structure achieves the best thermal stability configuration during formation. The shear parameter game optimization model solves the contradiction between efficiency and quality in traditional mechanical shearing by using a two-layer game mechanism of maximizing upper-layer shear efficiency and optimizing lower-layer particle uniformity. The introduction of the shear stress coupling term allows the two optimization objectives to reach a dynamic balance through mutual constraints, thereby obtaining nano- and micro-scale particles with uniform particle size distribution and complete shape. The particle aggregation stability matrix, through multidimensional parameter coupling analysis, can accurately predict and evaluate the behavioral changes of particles in high-temperature and high-salt environments, providing a scientific basis for stability control. The entire technical solution, through the organic combination of matrix modeling, intelligent prediction, and game-theoretic optimization, achieves precise control over the entire process from molecular structure design to macroscopic performance control, thereby ensuring the long-term stability of nano- and micro-scale gel particles in extreme environments.

[0062] The following provides a specific embodiment 1 of the present invention, and the specific implementation of each step in this embodiment 1 is described in detail below.

[0063] The specific implementation of step S01 involves constructing an optimized gel composition base system by precisely controlling the proportions and uniformity of each component. First, polyacrylamide with a relative molecular mass of 6 million to 12 million is dissolved at a mass fraction of 0.3% to 0.6% in simulated formation water with a mineralization of 5000 mg / L to 300000 mg / L. A stepwise dissolution method is used, with thorough stirring at 60°C to 80°C for 30 to 60 minutes to ensure complete expansion of the polymer molecular chains. Subsequently, phenolic resin crosslinking agent is added at a mass fraction of 0.4% to 0.9%, using a batch-by-batch addition method to avoid pre-crosslinking caused by excessively high local concentrations. During this process, a crosslinking uniform distribution matrix algorithm is used to monitor and optimize the component distribution in real time. This matrix, based on the principle of three-dimensional spatial grid division, divides the reaction vessel space into several micro-grids. Concentration gradient detection technology is used to monitor the concentration distribution of polyacrylamide and phenolic resin crosslinking agent within each grid in real time, and the component uniformity coefficient is calculated. The component uniformity coefficient U... c The calculation formula is:

[0064]

[0065] In the formula U c C is the component homogeneity coefficient, dimensionless; i The concentration of the component in the i-th grid is expressed in mg / L. The value represents the average concentration of components within all grids, expressed in mg / L; n is the total number of grids, dimensionless. A component uniformity coefficient greater than 0.85 is considered uniformly distributed; a coefficient less than 0.85 requires adjustment of the stirring rate and time. The stirring rate can be adjusted from 100 rpm to 300 rpm. The purpose of this step is to establish a basic gel composition system with uniform component distribution and optimal crosslinking reactivity, providing an ideal precursor for subsequent gelation reactions.

[0066] The specific implementation of step S02 involves precise control of the gelation reaction process guided by an intelligent thermal stability analysis model. The prepared gel composition is transferred to a constant-temperature reactor, and a thermal stability analysis model based on the Transformer architecture is used to predict and optimize reaction parameters. This model uses the molecular weight of polyacrylamide, the concentration of the phenolic resin crosslinking agent, the simulated formation water salinity, and the component homogeneity coefficient output from step S01 as input parameters. It captures the complex relationships between these parameters through a multi-head attention mechanism, outputting predicted values ​​for the optimal reaction temperature and reaction time. The core prediction function of the thermal stability analysis model is:

[0067] T opt , t opt =f thermal (M w Ccrosslink S water U c );

[0068] In the formula T opt The optimal reaction temperature, in °C; t opt The optimal reaction time is expressed in hours (h); f thermal M is the thermal stability analysis function; w The value is the molecular weight of polyacrylamide, in ten thousand Da; C crosslink S represents the concentration of the phenolic resin crosslinking agent, expressed in %; water To simulate formation water salinity, the unit is mg / L; U c The component uniformity coefficient is dimensionless. The reaction temperature is controlled with an accuracy of ±2℃, and the reaction time with an accuracy of ±0.5 hours. Simultaneously, a maximum tolerance temperature characteristic matrix is ​​used to monitor the crosslinking reaction process in real time. This matrix collects temperature data at various points in the reaction system through a temperature sensor array and judges the degree of crosslinking progress by combining the changes in rheological parameters. The crosslinking reaction is considered basically complete when the system viscosity reaches 8 to 12 times the initial viscosity. During the reaction, the temperature is maintained within the range of 50℃ to 120℃, with an optimal temperature of 85℃ to 95℃, and the reaction time is 4 to 8 hours. The purpose of this step is to form a gel matrix with ideal three-dimensional network structure density and thermal stability, providing a suitable raw material basis for subsequent mechanical shearing treatment.

[0069] The specific implementation of step S03 involves employing a two-layer game theory optimization strategy to achieve intelligent control of the mechanical shearing process. First, a shearing strategy is optimized using a shearing parameter game theory optimization model. This model adopts a two-layer game theory algorithm architecture. The upper-layer game model uses maximizing shearing efficiency as the objective function. Input parameters include gel bulk viscosity, crosslinking density, target particle size range, equipment power limit, and shearing time constraint. The selection of the shearing equipment type and the basic shearing rate range are determined through a Nash equilibrium solution algorithm. The shearing efficiency maximization function η... max The expression is:

[0070]

[0071] In the formula η max The function value that maximizes shear efficiency is expressed in m. 3 / (kW·h); V output The volume of nano-sized gel particles produced per unit time, in meters (m). 3 / h;P input The power input to the device is expressed in kW; t shearThe shearing time is expressed in hours (h). The lower-level game theory model uses particle uniformity optimization as the objective function. Input parameters include the target particle size range, gel bulk crosslinking density, shear strength gradient requirement, target particle shape factor, and uniformity standard deviation constraint. The output is a shear uniformity control strategy. The particle uniformity optimization function is U. particle The expression is:

[0072]

[0073] In the formula U particle The function value for particle uniformity optimization is dimensionless; σ d This represents the standard deviation of particle size distribution, in μm. The average particle size is expressed in μm. The two-layer model interacts and constrains each other through a shear stress coupling term, Ψ. coupling The expression is:

[0074] Ψ coupling =α·η max ·U particle ;

[0075] In the formula Ψ coupling This is a shear stress coupling term, in meters. 3 / (kW·h); α is the coupling coefficient, dimensionless, ranging from 0.1 to 0.3. Increased shear efficiency affects the weight allocation of the particle uniformity optimization strategy through the coupling term. Based on the model output, an appropriate shearing method is selected. The colloid mill method is suitable for processing high-viscosity gel bulks, with a shear rate controlled between 8000 rpm and 15000 rpm. The high-speed shearing machine method is suitable for medium-viscosity gel bulks, with a shear rate between 3000 rpm and 8000 rpm. The pipeline shearing crosslinking method is suitable for continuous production, with the shear rate controlled by the pipeline diameter and flow rate. The purpose of this step is to efficiently break large gel bulks into nano- and micro-sized particles with controllable particle size, while maintaining particle shape regularity and uniform particle size distribution.

[0076] The specific implementation of step S04 involves achieving precise control of particle stability through multi-dimensional stress field analysis and accurate parameter calculation. Based on the optimal shear strategy and uniform shear stress matrix parameters output by the shear parameter game optimization model in step S03, accurate shear parameters are calculated using the shear stress transfer equation. This equation is based on the stress tensor analysis theory in fluid mechanics, comprehensively considering the viscoelastic characteristics of the gel bulk, the influence of shear temperature on material properties, and the influence of shearing equipment geometry parameters on stress distribution. Input parameters include the basic shear rate range, gel bulk viscosity, crosslinking density, shear temperature, and uniform shear stress matrix parameters. The output is accurate shear rate and shear time values. The expression for the shear stress transfer equation is:

[0077] τ=μ eff ·γ+β·σ n ;

[0078] γ precise =γ base ·f(T,ρ cross ,τ);

[0079] In the formula, τ is the shear stress, with units of Pa; μ eff The effective shear viscosity is expressed in Pa·s; γ is the shear rate, expressed in s. -1 β is the stress transfer coefficient, dimensionless, ranging from 0.05 to 0.15; σ n Normal stress, unit is Pa; γ precise For precise shear rate, the unit is s. -1 ;γ base Base shear rate, in seconds -1 ;f(T,ρ cross τ) is the temperature-crosslinking density stress correction function, dimensionless; T is the shear temperature, in °C; ρ cross Crosslinking density, in mol / m 3 The shear rate calculation accuracy reaches ±50 rpm, and the shear time calculation accuracy is ±0.1 hours. Simultaneously, a nano-micro particle size distribution matrix is ​​used to monitor the shearing process in real time. This matrix uses laser particle size analysis technology to detect the particle size distribution in real time. When the standard deviation of the particle size distribution is less than 0.8 μm and the average particle size is within the target range of 0.3 μm to 6.4 μm, the control requirements are considered met. The shear strength is dynamically adjusted based on the real-time particle size distribution data, with an adjustment range of ±10%. The purpose of this step is to ensure the acquisition of nano-micro gel particles with precise and controllable particle size and good shape integrity, providing standardized test samples for subsequent temperature and salt resistance performance testing.

[0080] The specific implementation of step S05 involves comprehensive performance testing of nano-scale gel particles using extreme environment simulation and a multi-parameter stability matrix evaluation system. The prepared nano-scale gel particles were placed in high-temperature environments ranging from 80°C to 180°C and high-salinity environments ranging from 50,000 mg / L to 300,000 mg / L for aging tests, with aging times set from 10 to 100 days. The high-temperature aging test employed a programmed temperature increase method, with an initial temperature set at 80°C, increasing by 10°C every 24 hours until the target temperature was reached, and then maintaining the target temperature for the remaining aging time. The high-salinity aging test used saline solutions with different mineralization levels, the main components of which included sodium chloride, calcium chloride, magnesium chloride, and sodium sulfate, prepared according to the actual formation water ion composition ratio. During the aging process, a particle aggregation stability matrix was used to continuously monitor the particle state. This matrix is ​​based on the principle of multi-dimensional parameter monitoring, and the monitored parameters include particle size change, shape factor change, aggregation degree, settling velocity, and network structure integrity. The monitoring frequency was once every 6 hours, and data acquisition was achieved using microscopic observation combined with image analysis technology. Particle stability is considered good when the average particle size change rate is less than 15%, the shape factor change rate is less than 20%, and the aggregation degree is less than 30%. Particle stability index S particle The calculation formula is:

[0081] S particle =w1·(1-Δd) avg / d 0avg )+w2·(1-ΔSF / SF0)+w3·(1-A agg );

[0082] In the formula S particle Δd is the particle stability index, dimensionless, ranging from 0 to 1; w1, w2, and w3 are the weighting coefficients for particle size variation, shape factor variation, and aggregation degree, respectively, dimensionless, with values ​​of 0.4, 0.3, and 0.3; Δd avg d represents the average particle size variation, in μm. 0avg ΔSF is the initial average particle size in μm; ΔSF is the change in shape factor, dimensionless; SF0 is the initial shape factor, dimensionless; A agg The aggregation degree is dimensionless. The matrix output is a particle stability index, which comprehensively reflects the overall stability of the particles under extreme conditions. The purpose of this step is to comprehensively evaluate the stability performance of nano- and micro-sized gel particles in practical application environments, providing reliable data support for performance optimization.

[0083] The specific implementation of step S06 involves establishing a quantitative performance evaluation system to scientifically analyze and optimize the performance of gel particles. Based on the particle stability index output from the particle aggregation stability matrix in step S05, a comprehensive performance evaluation is performed using a temperature and salt resistance evaluation coefficient calculation method. This coefficient calculation method adopts a weighted comprehensive evaluation principle, comprehensively considering three key performance indicators: structural integrity, aggregation ability, and stability duration of nano- and micro-scale gel particles in high-temperature and high-salt environments. The weights of each indicator are 0.4, 0.3, and 0.3, respectively. The temperature and salt resistance evaluation coefficient R... eval The calculation formula is:

[0084] R eval =1-(0.4·I struct +0.3·C coalesc +0.3·T stable );

[0085] In the formula R eval I is a dimensionless coefficient for evaluating temperature and salt resistance, ranging from 0 to 1. struct C is a structural integrity index, dimensionless, with a value range of 0 to 1; coalesc T is a cohesion capacity index, dimensionless, with a value range of 0 to 1; stable The stability duration index is dimensionless and ranges from 0 to 1. Structural integrity is assessed through particle morphology retention and network structure damage, while coalescence ability is assessed through interparticle interaction strength and coalescence rate. Stability duration is assessed through the duration of effective functional retention. A temperature and salt resistance evaluation coefficient of ≤0.4 indicates excellent performance, allowing direct progression to the next step. Coefficients greater than 0.4 but less than or equal to 0.75 require formulation optimization, including adjusting the polyacrylamide molecular weight to 8-10 million, increasing the phenolic resin crosslinking agent concentration to 0.6-0.8%, or optimizing shear parameters. Coefficients greater than 0.75 require re-preparation and verification of raw material quality and process parameter settings. The purpose of this step is to establish scientific performance evaluation standards and optimization guidance mechanisms to ensure the final product meets the expected technical performance requirements.

[0086] The specific implementation of step S07 involves comprehensive quality control of the final product through multi-scale morphology analysis and long-term stability verification. The prepared nano-scale gel particles are subjected to macroscopic morphology observation and microscopic structure analysis. Macroscopic morphology observation uses a stereomicroscope at 50x to 200x magnification to observe the overall shape, surface smoothness, and dispersion state between particles. Microscopic structure analysis uses a scanning electron microscope at 1000x to 10000x magnification to observe the internal network structure, crosslinking point distribution, and pore structure characteristics of the particles. Simultaneously, a long-term stability verification test is conducted under high temperature and high salt conditions, with the test conditions set at 120℃ and 150,000 mg / L salinity for more than 90 days. During the test, the particle aggregation behavior and network structure changes are detected every 10 days. Dynamic light scattering technology is used to monitor particle size distribution changes, and rheological testing is used to monitor viscoelasticity changes. When the particles can effectively aggregate under the set conditions, and the gel network structure formed after aggregation remains intact, with viscoelastic parameter changes of less than 20%, and a stability time of more than 90 days, the product is considered to be of qualified quality. The purpose of this step is to ensure that the final product has excellent adaptability to high temperature and high salt environment and long-term stability, meeting the stringent requirements of practical applications.

[0087] The thermal stability analysis model employs a deep sequence prediction network structure based on the Transformer architecture. This model comprises four core components: an input embedding layer, a multi-head attention mechanism layer, a feedforward neural network layer, and an output prediction layer. The input embedding layer converts input parameters such as polyacrylamide molecular weight, phenolic resin crosslinking agent concentration, simulated formation water salinity, and component homogeneity coefficient into high-dimensional feature vectors, with an embedding dimension of 512. The multi-head attention mechanism layer contains eight attention heads, each independently calculating the correlation weights between input features, capturing long-term dependencies and complex nonlinear mappings between parameters through a self-attention mechanism. The feedforward neural network layer uses a two-layer fully connected structure with 2048 hidden layer neurons using the ReLU activation function and 256 output layer neurons. The output prediction layer maps the output of the feedforward neural network layer to predicted optimal reaction temperature and reaction time values ​​through a linear transformation. The model has approximately 12 million parameters and a computational complexity of O(n^2). 2d), where n is the length of the input sequence and d is the embedding dimension. The model training dataset establishment process includes four stages: data collection, data preprocessing, feature engineering, and data augmentation. In the data collection stage, stability experimental data of different gel compositions under various temperature conditions were collected through laboratory pilot-scale, pilot-scale, and industrial-scale experiments. This covered polyacrylamide molecular weights ranging from 4 million to 15 million, phenolic resin crosslinking agent concentrations ranging from 0.2% to 1.2%, simulated formation water salinity ranging from 1000 mg / L to 500000 mg / L, reaction temperatures ranging from 40℃ to 140℃, and reaction times ranging from 1 hour to 12 hours, collecting over 15,000 sets of experimental data. In the data preprocessing stage, the raw experimental data were cleaned, denoised, and standardized, outliers and missing values ​​were removed, and Z-score standardization was used to normalize the numerical features. In the feature engineering stage, derived feature variables are constructed, including the ratio of molecular weight to crosslinking agent concentration, the interaction term between mineralization and temperature, and the product term between time and temperature, etc., to increase the model's ability to represent complex relationships. In the data augmentation stage, interpolation and noise injection methods are used to expand the training samples, ultimately constructing a training set containing 12,000 training data sets, a validation set containing 2,000 validation data sets, and a test set containing 1,000 test data sets. Model training uses the cross-entropy loss function to measure the difference between predicted and true values, employs the Adam optimizer for parameter updates, sets the learning rate to 0.001 and uses a cosine annealing strategy for dynamic adjustment, uses a batch size of 32, and conducts 200 training epochs. An early stopping strategy is used during training to prevent overfitting; training stops when the validation set loss does not decrease for 10 consecutive epochs. Model convergence is evaluated by the prediction accuracy on the validation set; the model is considered convergent when the prediction accuracy reaches above 92% and the validation set loss is stable. The temperature adaptability adjustment function dynamically adjusts the number of attention heads according to different temperature sensitivity indices. The formula for calculating the temperature sensitivity index 'a' is:

[0088]

[0089] In the formula, a is the temperature sensitivity index, which is dimensionless; ρ cross This represents the bulk crosslinking density of the gel, in mol / m³. 3 S water To simulate formation water salinity, the unit is mg / L; U c M is the component homogeneity coefficient, dimensionless; w This refers to the molecular weight of polyacrylamide, in ten thousand Da; 10 6This is a dimensionless constant. When the temperature sensitivity index *a* is in the range [0, 0.25), four attention heads are used to adapt to low-sensitivity scenarios; when *a* is in the range [0.25, 0.5), six attention heads are used to adapt to medium-sensitivity scenarios; when *a* is in the range [0.5, 0.75), eight attention heads are used to adapt to high-sensitivity scenarios; and when *a* is in the range [0.75, 1.0], twelve attention heads are used to adapt to extremely high-sensitivity scenarios. This adjustment mechanism can adaptively optimize the model's prediction accuracy according to actual operating conditions, improving the model's generalization ability in different application scenarios.

[0090] The key technical ideas of this invention are mainly reflected in three aspects: crosslinking uniform distribution matrix construction technology, shear parameter bilayer game optimization technology, and Transformer-based intelligent prediction technology. The crosslinking uniform distribution matrix construction technology, through three-dimensional spatial gridding analysis and real-time concentration gradient monitoring, achieves precise distribution control of polyacrylamide and phenolic resin crosslinking agents in the gel composition. Compared with traditional empirical formulation methods, this technology can quantitatively evaluate the uniformity of component distribution and provide real-time control guidance, significantly improving the consistency and stability of the gel network structure and avoiding network defects and performance fluctuations caused by uneven local component concentrations. The shear parameter bilayer game optimization technology adopts a strategy combining upper-layer shear efficiency maximization game and lower-layer particle uniformity optimization game. Through Nash equilibrium solution, it achieves synergistic optimization of shear efficiency and product quality. Compared with traditional single-objective optimization methods, this technology can ensure precise control of product particle size distribution while guaranteeing production efficiency, resolving the contradiction between high-efficiency shearing and high-quality products, and realizing intelligent decision-making in the preparation process. The Transformer-based intelligent prediction technology utilizes a multi-head attention mechanism to capture the complex relationships between process parameters and achieves accurate prediction of optimal reaction conditions through deep learning. Compared to traditional empirical formulas or simple mathematical models, this technology has stronger nonlinear mapping and generalization capabilities, adapting to variations in different formulations and operating conditions, and significantly improving the intelligence level and prediction accuracy of gelation reaction control. The synergistic effect of these three key technologies constructs a fully intelligent control system from raw material preparation and gelation reaction to mechanical shearing, realizing the precision, standardization, and automation of the nano-micro gel preparation process. Compared to traditional experience-dependent preparation methods, this synergistic technology system can significantly improve product quality stability, reduce production costs, and shorten process development cycles, providing a complete technical solution for the industrial production of temperature- and salt-resistant nano-micro gels.

[0091] To better understand and implement this invention, Example 2, a specific application scenario, is provided below: The technical team needed to develop profile control and water shut-off materials for a deep, high-temperature, and high-salinity oil reservoir. The reservoir's formation temperature reached 165°C, and the formation water salinity was as high as 280,000 mg / L. Traditional millimeter-sized gels quickly failed under such harsh conditions and could not meet long-term sealing requirements. The technical team decided to use the temperature-resistant and salt-resistant nano-microscale gel preparation method of this invention to solve this technical problem.

[0092] In the gel composition formulation and optimization stage, the technical team first precisely formulated the gel composition. Polyacrylamide with a relative molecular mass of 8 million and a degree of hydrolysis of 4% was selected as the main polymer matrix, with a mass fraction controlled at 0.5%. Hydroquinone-type phenolic resin was chosen as the crosslinking agent, with a mass fraction set at 0.7%. Simulated formation water with a mineralization of 280,000 mg / L was used as the dispersion medium, with a mass fraction of 98.8%. During the formulation process, the technical team constructed a crosslinking uniformity distribution matrix to describe the distribution of polyacrylamide molecular chains and phenolic resin crosslinking agent in three-dimensional space. This matrix, with 12×12 matrix elements, characterizes the crosslinking density at different positions, ensuring uniform mixing of each component. After matrix operation analysis, the component uniformity coefficient was obtained as 0.87, a value that guided the precise adjustment of subsequent mixing process parameters.

[0093] In the intelligent gelation reaction control stage, the technical team employed a thermal stability analysis model based on the Transformer architecture to predict optimal reaction conditions. This model includes eight multi-head attention mechanisms and a feedforward neural network layer. Input parameters include a polyacrylamide molecular weight of 8 million, a phenolic resin crosslinking agent concentration of 0.7%, a simulated formation water salinity of 280,000 mg / L, and a component homogeneity coefficient of 0.87. A temperature sensitivity index of 0.68 was calculated using a temperature adaptability adjustment function, and predictions were made using eight attention heads according to the set adjustment strategy. The model predicted an optimal reaction temperature of 115℃ and an optimal reaction time of 18 hours. During gelation, a maximum tolerance temperature feature matrix continuously monitored the crosslinking reaction process. The matrix rows represent the temperature gradient from 50℃ to 120℃, and the columns represent structural characteristic parameters including crosslinking density, viscosity change rate, and network integrity index. Placing the prepared gel composition in a constant temperature environment of 115℃ for 18 hours successfully yielded a gel matrix with a stable three-dimensional network structure.

[0094] In the precision mechanical shearing stage, the technical team used a game-theoretic optimization model of shearing parameters to determine the optimal shearing strategy. The upper-level game model aims to maximize shearing efficiency, with inputs including a gel bulk viscosity of 3500 mPa·s and a crosslinking density of 2.3 × 10⁻⁶. -4The target particle size range is 0.4 μm to 5.8 μm, the equipment power limit is 8 kW, and the shearing time constraint is 90 min. The output is a Waring Blender high-speed shearing machine with a basic shear rate range of 8000–12000 rpm. The lower-level game theory model aims to optimize particle uniformity, with inputs including a target particle size range of 0.4 μm to 5.8 μm and a gel bulk crosslinking density of 2.3 × 10⁻⁶ mol / g. -4 With a shear strength gradient requirement, a particle shape factor target value of 0.92, and a uniformity standard deviation limit of 0.15, the output shear uniformity control strategy is a piecewise variable-rate shearing mode. Through shear stress coupling term analysis, a mutually restrictive relationship exists between shear efficiency and particle uniformity. Based on the uniform shear stress matrix and the optimal shearing strategy, the technical team calculated the precise shear parameters using the shear stress transfer equation: a shear rate of 10500 rpm and a shearing time of 75 min.

[0095] During the granulation stabilization stage, the technical team strictly controlled the shear strength based on the nanoparticle size distribution matrix. This matrix statistically analyzes and controls the particle size distribution of the nano-scale gel particles after shearing, with matrix elements recording the percentage of particles in different size ranges. By precisely controlling the shear strength, the integrity of the particle shape and the uniformity of the particle size were ensured, ultimately yielding nano-scale gel particles with average particle sizes of 0.6 μm and 4.2 μm.

[0096] In the comprehensive experimental verification phase, the technical team designed 7 experimental groups and 6 control groups to fully verify the performance of the nano-scale gel. Experiment 1 used 0.6μm nano-scale gel particles, aging them for 60 days at 165℃ and 280,000 mg / L high temperature and high salt concentration with a concentration of 3% and an average particle size of 0.6μm. Figure 1 As shown, macroscopically, the gel particles exhibited agglomeration; microscopic observation revealed significant particle aggregation; and microscopic morphology analysis indicated that the particle shape remained intact, the network structure was not disrupted, and the stabilization time reached 95 days. Experiment 2 used 4.2 μm nano-sized gel particles, with a concentration of 3% and an average particle size of 4.2 μm, and aged them for 60 days at 155℃ and 250,000 mg / L. Figure 2 As shown, the particles also exhibit good agglomeration properties, with a stabilization time of 92 days.

[0097] Experimental Example 3 used a nano-silica-reinforced gel particle formulation. 0.08% nano-SiO2 was added to the original formulation to prepare reinforced gel particles with an average particle size of 0.7 μm. The 4% concentration particles were aged for 60 days under extreme conditions of 170℃ and 300,000 mg / L. Figure 3As shown, the reinforced gel particles exhibit superior temperature and salt resistance, with a stability time extended to 105 days. Example 4 uses resorcinol-type phenolic resin crosslinking agents to prepare nano-scale gel particles. The selected materials include 0.6% polyacrylamide (relative molecular mass 6.5 million, degree of hydrolysis 5%), 0.4% resorcinol-type phenolic resin crosslinking agent, 98.95% simulated formation water with a mineralization of 8000 mg / L, and 0.05% oxalic acid pH adjuster. The prepared gel particles, with an average particle size of 0.5 μm and a concentration of 5%, were aged for 60 days at 140℃ and 220,000 mg / L. Figure 4 As shown, the particle agglomeration effect is good, and the stabilization time reaches 88 days.

[0098] Experimental Example 5 used a catechol-type phenolic resin crosslinking agent, selecting polyacrylamide with a relative molecular mass of 9.6 million and a degree of hydrolysis of 3% at a mass fraction of 0.5%, and the catechol-type phenolic resin crosslinking agent at a mass fraction of 0.8%. The simulated formation water with a mineralization of 200,000 mg / L was 98.7%. Gel particles with an average particle size of 0.8 μm and a concentration of 2.5% were prepared and aged for 60 days at 158℃ and 275,000 mg / L. Figure 5 As shown, it exhibits excellent structural stability, with a stability time of 91 days. Experimental Example 6, instead of using a catechol-type phenolic resin crosslinking agent, used an organic chromium crosslinking agent to prepare nano-scale gel particles. The selected materials included 0.5% polyacrylamide with a relative molecular mass of 8 million and a degree of hydrolysis of 5%, 0.6% organic chromium crosslinking agent, and 98.9% simulated formation water with a mineralization of 1000 mg / L. The prepared gel particles had an average particle size of 0.4 μm and a concentration of 8%. Although the temperature resistance was relatively low, after aging at 85℃ and 120,000 mg / L for 60 days, as shown... Figure 6 As shown, the particles still maintain good integrity and remain stable for 96 days.

[0099] Experiment 7 used a composite crosslinking agent of phenolic resin and organochromium. Polyacrylamide with a relative molecular mass of 7 million and a degree of hydrolysis of 4.5% was selected, with a mass fraction of 0.45%. The mass fraction of the phenolic resin crosslinking agent was 0.3%, and the mass fraction of the organochromium crosslinking agent was 0.25%. The simulated formation water with a mineralization of 150,000 mg / L was 99.0%. Composite gel particles with an average particle size of 1.2 μm and a concentration of 6% were prepared and aged for 60 days at 148℃ and 200,000 mg / L, exhibiting balanced comprehensive performance and a stability time of 89 days.

[0100] Molecular dynamics simulations indicate that the technical team conducted in-depth theoretical research. For example... Figure 7As shown, the molecular dynamics model of 4.2 μm micron-sized gel particles at 155℃ and 250,000 mg / L indicates that the movement of polymer chain segments within the particles is relatively slow, and the crosslinked network remains stable. Figure 8 As shown, under more stringent conditions of 165℃ and 280,000 mg / L, the molecular chain segment movement of 0.6 μm nanoscale gel particles intensifies, but the crosslinking points still maintain network integrity. Figure 9 As shown, the potential energy variation curves of micron-sized and nano-sized gel particles under high temperature and high salt conditions indicate that nano-sized particles exhibit lower potential energy fluctuations, demonstrating better structural stability. The potential energy variation range is controlled within ±15 kJ / mol, far below the critical value of 50 kJ / mol for structural failure.

[0101] In the comparative experimental analysis phase, the technical team set up six comparative examples to verify the advantages of nano- and micro-sized particles. Comparative Example 1 used the same formulation as Experimental Example 1, namely, 0.5% polyacrylamide (relative molecular mass 8 million, degree of hydrolysis 4%), 0.7% hydroquinone-type phenolic resin crosslinking agent, and 98.8% simulated formation water (mineralization 280,000 mg / L), but maintained the traditional millimeter-sized dimensions of 15–45 mm. After aging at 165℃ and 280,000 mg / L for 3 days, significant network structure destruction occurred. Comparative Example 2 used the same nano-SiO2-enhanced formulation as Experimental Example 3, namely, 0.5% polyacrylamide, 0.7% hydroquinone-type phenolic resin crosslinking agent, 0.08% nano-SiO2, and 98.72% simulated formation water (mineralization 280,000 mg / L), but prepared into millimeter-sized particles of 18–42 mm. The network structure was destroyed after aging for 2 days at 170℃ and 300,000 mg / L.

[0102] Comparative Example 3 used a resorcinol-based formulation, consisting of 0.6% polyacrylamide (6.5 million molecular weight, 5% degree of hydrolysis), 0.4% resorcinol-based phenolic resin crosslinking agent, 0.05% oxalic acid pH adjuster, and 98.95% simulated formation water with a mineralization of 8000 mg / L. The resulting product was prepared in millimeter-sized forms ranging from 12 to 38 mm. After aging at 140℃ and 220,000 mg / L for 4 days, the structure was destroyed. Comparative Example 4 used an catechol-based formulation, consisting of 0.5% polyacrylamide (9.6 million molecular weight, 3% degree of hydrolysis), 0.8% catechol-based phenolic resin crosslinking agent, and 98.7% simulated formation water with a mineralization of 200,000 mg / L. The resulting product was prepared in millimeter-sized forms ranging from 16 to 44 mm. After aging at 158℃ and 275,000 mg / L for 3 days, the network was destroyed. Comparative Example 5 used an organochromium formulation, consisting of 0.5% polyacrylamide (8 million molecular weight, 5% degree of hydrolysis), 0.6% organochromium crosslinking agent, and 98.9% simulated formation water (1000 mg / L mineralization), prepared into millimeter-sized units of 10–35 mm. After aging at 85℃ and 120,000 mg / L for 5 days, the structure was destroyed. Comparative Example 6 used a composite crosslinking agent formulation, consisting of 0.45% polyacrylamide (7 million molecular weight, 4.5% degree of hydrolysis), 0.3% phenolic resin crosslinking agent, 0.25% organochromium crosslinking agent, and 99.0% simulated formation water (150,000 mg / L mineralization), prepared into millimeter-sized units of 14–40 mm. After aging at 148℃ and 200,000 mg / L for 4 days, the network was destroyed. Figure 10 As shown, all the conventional millimeter-scale gels in the comparative examples exhibited significant collapse of the three-dimensional network structure in a high-temperature and high-salt environment, with the macroscopic morphology showing a broken state and the microstructure completely failing.

[0103] During performance evaluation and data analysis, the technical team established a particle aggregation stability matrix to assess the performance of all experimental samples. This matrix includes the influence weights of multi-dimensional parameters such as temperature, salinity, and time on particle stability. The particle stability index for each experimental example was obtained through matrix operations, as shown in Table 1.

[0104] Table 1. Particle stability index and temperature and salt resistance evaluation coefficients for each experimental case.

[0105]

[0106] According to the criteria for evaluating temperature and salt resistance, the evaluation coefficients of all experimental examples were less than or equal to 0.4, indicating excellent performance and meeting the requirements of practical applications. The technical team also conducted systematic tests on particles with different size ranges, and the results are shown in Table 2.

[0107] Table 2 Comparison of properties of nano-microscale gel particles with different particle sizes

[0108]

[0109] Data shows that smaller nano-sized gel particles have better temperature and salt resistance.

[0110] During the field application verification phase, after laboratory validation, the technical team applied the prepared nano-sized gel particles to field tests in the target reservoir. Hydroquinone-type nano-sized gel particles with an average particle size of 0.6 μm were selected, with an injection concentration of 3% and an injection volume of 120 mg / L. 3 Field monitoring data showed that on the 30th day after injection, the water cut of the target layer decreased from 89% to 62%, demonstrating a significant sealing effect. On the 60th day, the water cut stabilized at around 65%, and on the 90th day, it remained at 67%, indicating that the nano-sized gel particles exhibit good long-term stability under actual formation conditions.

[0111] Analysis of technological advancements shows that this invention represents a significant technological leap forward compared to traditional millimeter-scale gel plugging technologies. From a molecular structure perspective, nano- and micro-scale granulation significantly improves the surface area to volume ratio of the gel, increasing the contact area with formation fluids and enhancing coalescence efficiency. Simultaneously, smaller particles possess lower surface energy, enabling them to maintain a more stable thermodynamic equilibrium in high-temperature, high-salt environments. From a mass and heat transfer perspective, the relatively smaller temperature and concentration gradients within nano- and micro-scale particles reduce structural stress caused by differences in internal and external environments, avoiding internal cracking and network disruption common in large-size gels. The Brownian motion of nano- and micro-scale particles enhances their transport capabilities in porous media, allowing them to penetrate deeper into formation micropores and achieve more precise selective plugging. From a cross-linking network structure perspective, the mechanical shearing during nano- and micro-scale granulation not only alters particle size but, more importantly, optimizes the conformational distribution of polymer chain segments. Shearing causes the previously disordered long-chain molecules to reorient, forming a denser and more ordered cross-linking network, improving the network's mechanical strength and chemical stability. Furthermore, the nano-micro particle technology, through the introduction of multi-level game optimization models and intelligent prediction algorithms, achieves precise control of the preparation process, ensuring the consistency and reproducibility of product quality. This intelligent preparation method provides a reliable technical path for the industrial production of gel-based plugging materials, and promotes the development of oilfield chemical profile control and water shut-off technology to a higher level.

[0112] It should be noted that the variables involved in this invention are explained in detail in Table 3 below.

[0113] Table 3. Variable Explanation Table

[0114]

[0115] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for preparing a temperature-resistant and salt-resistant nano-microscale gel, characterized in that, The process involves controlling component distribution through a crosslinking uniform distribution matrix, predicting reaction conditions using a thermal stability analysis model, guiding shear strategies using a shear parameter game optimization model, and achieving precise control by evaluating stability through a particle aggregation stability matrix. This includes: preparing the gel composition by mixing polyacrylamide and phenolic resin crosslinking agents, adding simulated formation water, constructing a crosslinking uniform distribution matrix for component distribution optimization, and outputting a component uniformity coefficient to guide the adjustment of mixing process parameters; performing the gelation reaction by predicting the optimal reaction temperature and time using a thermal stability analysis model, monitoring the crosslinking reaction process through a maximum tolerance temperature characteristic matrix, and forming a three-dimensional network structure gel body; executing a mechanical shearing treatment step by determining the shear strategy selection using a shear parameter game optimization model and outputting the optimal shear strategy scheme; implementing a particle stability treatment step by calculating precise shear parameters through a shear stress transfer equation, controlling shear intensity based on a nano-micro particle size distribution matrix, and obtaining nano-micro-sized gel particles; conducting temperature and salt resistance performance testing by evaluating particle stability using a particle aggregation stability matrix and outputting a particle stability index; performing performance evaluation and optimization by quantitative analysis using temperature and salt resistance evaluation coefficients; and finally, completing the final product quality control step. Particle stability index The calculation formula is: ； In the formula This is the particle stability index. , , These are the weighting coefficients for particle size variation, shape factor variation, and aggregation degree, respectively. This represents the change in average particle size. The initial average particle size, This represents the change in shape factor. The initial shape factor, Cohesion degree; The thermal stability analysis model employs a deep sequence prediction network structure based on the Transformer architecture, comprising four core components: an input embedding layer, a multi-head attention mechanism layer, a feedforward neural network layer, and an output prediction layer. The input embedding layer converts input parameters such as polyacrylamide molecular weight, phenolic resin crosslinking agent concentration, simulated formation water salinity, and component homogeneity coefficient into high-dimensional feature vectors, with an embedding dimension set to 512. The multi-head attention mechanism layer contains eight attention heads, each independently calculating the correlation weights between input features, capturing long-term dependencies and complex nonlinear mappings between parameters through a self-attention mechanism. The feedforward neural network layer uses a two-layer fully connected structure. The output prediction layer maps the output of the feedforward neural network layer to predicted optimal reaction temperature and reaction time values ​​through a linear transformation.

2. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 1, characterized in that, The specific steps for preparing the gel composition involve mixing polyacrylamide with a relative molecular mass of 6 million to 12 million and a degree of hydrolysis of 3% to 6% with a phenolic resin crosslinking agent at a mass fraction of 0.3% to 0.6% and 0.4% to 0.9%, respectively, and adding simulated formation water with a mineralization of 5000 mg / L to 300000 mg / L.

3. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 2, characterized in that, The gelation reaction step specifically involves placing the gel composition in a temperature environment of 50°C to 120°C, using a thermal stability analysis model to predict the optimal reaction temperature and reaction time based on the component uniformity coefficient output by the polyacrylamide molecular weight, phenolic resin crosslinking agent concentration, simulated formation water salinity, and crosslinking uniform distribution matrix, and controlling the gelation process according to the predicted optimal reaction temperature and reaction time.

4. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 3, characterized in that, The mechanical shearing process specifically involves continuously shearing the gel bulk using a colloid mill, a high-speed shearing machine, or a pipeline shearing crosslinking method. A shearing strategy selection is determined based on the gel bulk viscosity, crosslinking density, and target particle size range using a shearing parameter game optimization model.

5. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 4, characterized in that, The granulation stability treatment step specifically involves using the optimal shear strategy scheme output by the uniform shear stress matrix and the shear parameter game optimization model, calculating the precise shear parameters through the shear stress transfer equation, ensuring the integrity of particle shape and uniformity of particle size, and obtaining nano-micro gel particles with an average particle size of 0.3μm to 6.4μm.

6. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 5, characterized in that, The specific steps of the temperature and salt resistance test are as follows: the prepared nano-sized gel particles are placed in a high temperature environment of 80℃ to 180℃ and a high salinity environment of 50000mg / L to 300000mg / L for 10 to 15 days, and the stability of the particles under extreme conditions is evaluated using the particle aggregation stability matrix.

7. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 6, characterized in that, The performance evaluation and optimization steps specifically involve quantitatively analyzing the performance of gel particles based on the particle stability index output by the particle aggregation stability matrix using the temperature and salt resistance evaluation coefficient. When the temperature and salt resistance evaluation coefficient is less than or equal to 0.4, it indicates excellent performance; when it is greater than 0.4 but less than or equal to 0.75, the formulation needs to be optimized; and when it is greater than 0.75, the gel particles need to be re-prepared.

8. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 7, characterized in that, The final product quality control steps specifically involve observing the macroscopic morphology and analyzing the microstructure of the prepared nano-scale gel particles to ensure that the particles effectively aggregate under high temperature and high salt conditions without network structure destruction, and that the stabilization time reaches more than 90 days.

9. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 8, characterized in that, The crosslinking uniform distribution matrix is ​​specifically used to describe the distribution of polyacrylamide molecular chains and phenolic resin crosslinking agents in three-dimensional space. The matrix elements characterize the crosslinking density at different positions, ensuring the uniform mixing of each component in the gel composition and the consistency of the crosslinking reaction.

10. The method for preparing the temperature-resistant and salt-resistant nano-microscale gel according to claim 9, characterized in that, The highest tolerance temperature feature matrix is ​​specifically used to record the structural stability parameters of the gel body under different temperature conditions. The matrix rows represent temperature gradients and the columns represent structural feature parameters. Matrix operations are used to monitor structural changes during the crosslinking reaction process.

Images