Mixing control method, device, equipment and medium for oilfield dry powder fracturing fluid

Abstract

The present application provides a mixing control method, device, equipment and medium for oilfield dry powder fracturing fluid, which includes: real-time collection of mixing process data; establishment of a virtual model based on the internal structure of the target equipment; use of fluid dynamics to simulate the flow of fluid in the mixing tank and construct a real-time fluid motion model; based on the principle of chemical kinetics, establish a chemical kinetic model of dry powder dissolution, cross-linking reaction and gel breaking reaction, and integrate material characteristic parameters into the chemical kinetic model; input the mixing process data into the virtual model, dynamically calibrate the virtual model through the real-time fluid motion model and chemical kinetic model, dynamically adjust water quality data, material ratio and stirring conditions, and perform multi-scenario real-time simulation analysis of the mixing process; and execute corresponding dry powder fracturing fluid mixing control on the target equipment according to the mixing simulation data. This can improve the quality and efficiency of fracturing fluid mixing and assist in improving the development efficiency of unconventional oil and gas reservoirs.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a mixing control method, device, equipment and medium for oilfield dry powder fracturing fluid. Background Art

[0002] Oilfield dry powder fracturing fluid is a specialized liquid system used in fracturing operations during oil and natural gas extraction. Existing in a dry, solid powder form, it offers advantages such as ease of storage and transportation, as well as convenient on-site preparation. It is widely used in the development of unconventional oil and gas reservoirs, such as shale gas and tight oil reservoirs. For example, fracturing fluid is pumped into the formation at high pressure, fracturing the rock and forming fractures. The cross-linked, high-viscosity fracturing fluid then carries proppant into the fractures, ensuring they remain open even after pumping is stopped.

[0003] Currently, in the mixing process of dry powder fracturing fluids in oilfields, the high molecular weight polymers (such as guar gum and polyacrylamide) in the dry powder are highly hygroscopic and easily agglomerated during storage. Insufficient stirring speed or poor water quality (high mineralization or impurities) during mixing can cause the surface of the dry powder particles to rapidly swell, forming "colloids," preventing the interior from fully dissolving. Incompletely dissolved micelles reduce the viscosity and sand-carrying capacity of the fracturing fluid, and can even clog pumping lines, compromising operation safety. Furthermore, on-site mixing is typically performed in mobile mixing tanks. Due to uneven flow distribution within the tank (e.g., dead zones), different components (thickeners, crosslinkers, and additives) may stratify or experience localized over-concentration. These over-concentrated areas can prematurely crosslink and form jelly, compromising the stability of the fracturing fluid. Uneven concentrations can lead to inconsistent subsequent gel-breaking times, complicating flowback.

[0004] Therefore, in order to solve at least one of the above technical problems, it is urgent to propose a new mixing control solution for oilfield dry powder fracturing fluid. Summary of the Invention

[0005] The present application provides a mixing control method, device, equipment and medium for oilfield dry powder fracturing fluid, which are used to improve the quality and efficiency of fracturing fluid mixing, enhance adaptability to complex production environments, and assist in improving the development efficiency of unconventional oil and gas reservoirs.

[0006] In a first aspect, the present application provides a method for controlling the mixing of an oilfield dry powder fracturing fluid, the method comprising:

[0007] Deploy multi-source sensors in the target equipment to collect real-time mixing process data; the target equipment includes at least: a liquid mixing tank, an agitator, and a feeding device; the mixing process data includes at least: stirring speed, pressure in the liquid mixing tank, fluid temperature, dry powder feeding amount, and liquid mixing water flow rate;

[0008] Based on the internal structure of the target device, a virtual model corresponding to the target device is established; wherein the virtual model is provided with a liquid mixing tank, an agitator, a feeding device, and a pipeline layout between each virtual device; the internal structure includes at least: the tank shape, the agitator structure, and the pipeline layout;

[0009] Based on the Navier-Stokes equations and the continuity equation, fluid dynamics is used to simulate the fluid flow in the liquid mixing tank. The flow field distribution, vortex area, and stirring dead corners are analyzed to construct a real-time fluid motion model corresponding to the target equipment.

[0010] Based on the principles of chemical kinetics, a chemical kinetic model for dry powder dissolution, crosslinking reaction, and gel breaking reaction is established, and material characteristic parameters are integrated into the chemical kinetic model. The chemical kinetic model is used to simulate the reaction process of the target equipment under different conditions.

[0011] The mixing process data is input into a virtual model, and the virtual model is dynamically calibrated through a real-time fluid motion model and a chemical kinetic model. The water quality data, material ratio, and stirring conditions are dynamically adjusted, and a multi-scenario real-time simulation analysis of the mixing process is performed to obtain mixing simulation data of the target equipment; and corresponding dry powder fracturing fluid mixing control is performed on the target equipment based on the mixing simulation data.

[0012] In a second aspect, an embodiment of the present application provides a mixing control device for an oilfield dry powder fracturing fluid, the device comprising:

[0013] The acquisition unit is configured to deploy multiple sensors in a target device to collect mixing process data in real time; wherein the target device includes at least a liquid preparation tank, an agitator, and a feeding device; and the mixing process data includes at least agitation speed, pressure in the liquid preparation tank, fluid temperature, dry powder feeding amount, and liquid preparation water flow rate;

[0014] The first construction unit is configured to establish a virtual model corresponding to the target device based on the internal structure of the target device; wherein the virtual model includes a liquid mixing tank, an agitator, a feeding device, and a pipeline layout between each virtual device; the internal structure includes at least: a tank shape, a stirring paddle structure, and a pipeline layout;

[0015] The second construction unit is configured to use fluid dynamics to simulate the fluid flow in the liquid mixing tank based on the Navier-Stokes equations and the continuity equation, and analyze the flow field distribution, vortex areas, and stirring dead corners to construct a real-time fluid motion model corresponding to the target equipment;

[0016] The third building block is configured to establish a chemical kinetic model for dry powder dissolution, crosslinking reaction, and gel breaking reaction based on chemical kinetic principles, and incorporate material characteristic parameters into the chemical kinetic model; the chemical kinetic model is used to simulate the reaction process of the target device under different conditions;

[0017] The control unit is configured to input the mixing process data into a virtual model, dynamically calibrate the virtual model through a real-time fluid motion model and a chemical kinetic model, dynamically adjust water quality data, material ratios, and stirring conditions, and perform multi-scenario real-time simulation analysis of the mixing process to obtain mixing simulation data of a target device; and execute corresponding dry powder fracturing fluid mixing control on the target device according to the mixing simulation data.

[0018] In a third aspect, an embodiment of the present application provides a computing device, the computing device comprising:

[0019] at least one processor, memory, and input-output unit;

[0020] The memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute the mixing control method of the oilfield dry powder fracturing fluid of the first aspect.

[0021] In a fourth aspect, a computer-readable storage medium is provided, comprising instructions, which, when executed on a computer, cause the computer to execute the mixing control method for the oilfield dry powder fracturing fluid according to the first aspect.

[0022] In the technical solution provided in the embodiments of the present application, first, multi-source sensors are deployed in the target device to collect mixing process data in real time; wherein the target device includes at least: a liquid mixing tank, an agitator, and a feeding device; the mixing process data includes at least: stirring speed, pressure in the liquid mixing tank, fluid temperature, dry powder feeding amount, and liquid mixing water flow rate. Next, based on the internal structure of the target device, a virtual model corresponding to the target device is established; wherein, the virtual model is provided with a liquid mixing tank, an agitator, a feeding device, and the pipeline layout between each virtual device; the internal structure includes at least: tank shape, agitator structure, and pipeline layout. Then, based on the Navier-Stokes equations and the continuity equation, fluid dynamics is used to perform a dynamic simulation of the fluid flow in the liquid mixing tank, and the flow field distribution, vortex area, and stirring dead angle are analyzed to construct a real-time fluid motion model corresponding to the target device. Then, based on the principles of chemical kinetics, a chemical kinetic model of dry powder dissolution, crosslinking reaction, and gel breaking reaction is established, and the material characteristic parameters are integrated into the chemical kinetic model; the chemical kinetic model is used to simulate the reaction process of the target device under different conditions. Finally, the mixing process data is input into the virtual model, and the virtual model is dynamically calibrated through a real-time fluid motion model and a chemical kinetic model. The water quality data, material ratio, and stirring conditions are dynamically adjusted, and a multi-scenario real-time simulation analysis of the mixing process is performed to obtain the mixing simulation data of the target device. Thus, the corresponding dry powder fracturing fluid mixing control is performed on the target device according to the mixing simulation data. In summary, the mixing control method of oilfield dry powder fracturing fluid helps to improve the accuracy, efficiency, and quality of fracturing fluid mixing through its many advantages, enhances its adaptability to complex production environments, and promotes the intelligent development of production management, which has positive and important significance for oilfield mining operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0024] Figure 1 1 is a flow chart of a method for controlling the mixing of dry powder fracturing fluid in an oil field according to an embodiment of the present application;

[0025] Figure 2 1 is a schematic structural diagram of a mixing control device for an oilfield dry powder fracturing fluid according to an embodiment of the present application;

[0026] Figure 3 It is a structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to solve at least one technical problem in the related art, it is urgent to propose a new mixing control solution for oilfield dry powder fracturing fluid.

[0028] To solve at least one of the above technical problems, an embodiment of the present application provides a mixing control method for oilfield dry powder fracturing fluid.

[0029] Specifically, multi-source sensors are first deployed in the target device to collect real-time mixing process data. The target device includes at least a liquid mixing tank, an agitator, and a dosing device. This mixing process data includes at least stirring speed, pressure within the liquid mixing tank, fluid temperature, dry powder addition amount, and dosing water flow rate. Next, a virtual model of the target device is established based on its internal structure. This virtual model includes the liquid mixing tank, agitator, dosing device, and the pipeline layout between each virtual device. The internal structure includes at least the tank shape, agitator structure, and pipeline layout. Furthermore, based on the Navier-Stokes equations and the continuity equation, fluid dynamics is used to simulate the fluid flow within the liquid mixing tank. The flow field distribution, vortex areas, and mixing dead zones are analyzed to construct a real-time fluid motion model corresponding to the target device. Then, based on the principles of chemical kinetics, chemical kinetic models are established for the dry powder dissolution, crosslinking reaction, and gel breaking reaction. Material characteristic parameters are incorporated into the chemical kinetic model. The chemical kinetic model is used to simulate the reaction processes of the target device under different conditions. Finally, the mixing process data is input into the virtual model, which is dynamically calibrated using real-time fluid motion and chemical kinetic models. Water quality data, material ratios, and stirring conditions are dynamically adjusted, and multi-scenario real-time simulation analysis of the mixing process is performed to obtain mixing simulation data for the target equipment. Based on this mixing simulation data, the corresponding dry powder fracturing fluid mixing control is then executed for the target equipment.

[0030] In the technical solution of this application, by deploying multi-source sensors on key equipment such as liquid mixing tanks, agitators, and feeding devices, full-dimensional data monitoring of the mixing process can be achieved, and core data such as stirring speed, pressure, and temperature can be obtained in real time, providing an accurate and timely data basis for subsequent analysis, effectively avoiding decision-making errors caused by data loss or lag. At the same time, these data can help technicians monitor the operating status of equipment and fluid changes in real time, promptly detect sudden problems such as pipe blockages and stirring anomalies, and provide strong data support for fault diagnosis and early warning, ensuring the safe and stable operation of the mixing process.

[0031] Utilizing 3D modeling technology, a virtual model highly consistent with the actual equipment is constructed, accurately recreating details such as the tank shape, agitator structure, and pipeline layout, providing a reliable digital platform for subsequent fluid dynamics simulation and process optimization. The virtual model can support diverse simulation analyses under different operating conditions and parameter settings, helping technicians to pre-evaluate the impact of different design solutions or process adjustments on the mixing process, thereby optimizing equipment design and mixing processes, avoiding the trial-and-error costs associated with actual operations, and improving the scientific nature and efficiency of design and decision-making. Furthermore, the virtual model facilitates integration with other specialized software and systems, enabling multidisciplinary data interaction and collaborative analysis, and enhancing the comprehensiveness and practicality of the entire mixing control method.

[0032] With the help of computational fluid dynamics algorithms, an in-depth simulation and analysis of the fluid flow in the mixing tank is carried out based on the Navier-Stokes equations and the continuity equation, which can accurately present key information such as flow field distribution, vortex areas, and mixing dead corners. This helps technicians thoroughly understand the movement patterns of the fluid in the mixing tank, thereby optimizing the agitator design and adjusting the mixing parameters in a targeted manner, effectively improving the uniformity of dry powder dissolution and mixing efficiency, and ensuring the mixing quality of the fracturing fluid. Furthermore, the simulation results were experimentally verified using the particle image velocimetry (PIV) algorithm to ensure the accuracy and reliability of the simulation results, enhance the credibility of the model, and provide a solid technical guarantee for practical application. The real-time fluid motion model can dynamically simulate the changes in fluid under different mixing conditions, predict fluid behavior, and provide timely and effective support for real-time adjustment of the mixing strategy, so that the mixing process can better adapt to different working conditions.

[0033] By combining regression analysis and artificial neural network algorithms to establish a chemical kinetics model, the model can fully account for the impact of multiple complex factors, such as water quality, temperature, and material properties, on dry powder dissolution, crosslinking, and gel breaking reactions, achieving highly accurate predictions of chemical reaction progress. This provides a scientific theoretical basis for determining optimal reaction conditions and optimizing formulation design, helping to improve the performance and quality stability of fracturing fluids. Through in-depth research on the laws of chemical reactions, the model can guide technicians in rationally adjusting material ratios and reaction conditions based on different material characteristic parameters, effectively reducing quality issues caused by improper chemical reaction control, lowering production risks, and improving production reliability and consistency. The chemical kinetics model can also predict potential anomalies during the reaction process, such as premature crosslinking and incomplete gel breaking, and issue timely warnings. It also provides appropriate response strategies, helping technicians take preventive measures to ensure a smooth mixing process.

[0034] By inputting the real-time collected mixing process data into the virtual model and dynamically calibrating it using a real-time fluid motion model and chemical kinetics model, the virtual model can reflect the actual operating status of the physical entity in real time, ensuring the timeliness and accuracy of the simulation results, and providing a reliable basis for subsequent decision-making. By comprehensively using various optimization algorithms such as genetic algorithms and response surface methodology, multiple factors such as water quality data, material ratios, and stirring conditions are collaboratively optimized. Through multi-scenario real-time simulation analysis, the interaction and influence between various factors can be fully considered to find the optimal parameter combination, significantly improve the quality and efficiency of fracturing fluid mixing, and enhance overall production efficiency. Simulating and evaluating mixing schemes under different parameter combinations in a virtual environment can predict the implementation effects of various schemes in advance, avoid a large number of tests and adjustments in actual production, save time and costs, and provide strong technical support for the formulation of scientific and reasonable mixing processes.

[0035] By conducting in-depth analysis of mixing simulation data, extracting key performance indicators and comparing them with standard parameters, and generating precise control instructions based on optimization strategies, equipment parameters such as agitator speed and feeding rate can be precisely adjusted to ensure that the quality of the fracturing fluid mix strictly meets requirements, thereby improving the effectiveness and success rate of fracturing operations and ensuring the efficiency and output of oil and gas production. Combined with the artificial neural network's ability to predict abnormal operating conditions, it can proactively detect potential problems and risks in the mixing process and promptly adjust control strategies to achieve rapid response and handling of abnormal situations, effectively preventing the escalation of problems, ensuring the stability and safety of the mixing process, and reducing the probability of production accidents. By implementing scientific and precise mixing control, production interruptions and product failures caused by mixing problems can be significantly reduced, improving production efficiency, reducing production costs, and enhancing oilfield production efficiency.

[0036] In summary, the mixing control method of dry powder fracturing fluid in this oil field can help improve the quality and efficiency of fracturing fluid mixing through many advantages, enhance its adaptability to complex production environments, and assist in improving the development efficiency of unconventional oil and gas reservoirs.

[0037] The oilfield dry powder fracturing fluid mixing control scheme provided in the embodiments of the present application can be executed by an electronic device, which can be a server, server cluster, or cloud server. The electronic device can also be a terminal device such as a mobile phone, computer, tablet computer, wearable device, or dedicated device (such as a dedicated terminal device equipped with a mixing control system for oilfield dry powder fracturing fluid). In an optional embodiment, the electronic device can be installed with a service program for executing the oilfield dry powder fracturing fluid mixing control scheme.

[0038] Figure 1 A schematic diagram of a mixing control method for an oilfield dry powder fracturing fluid provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes the following steps:

[0039] 101, deploy multi-source sensors in the target equipment to collect mixing process data in real time.

[0040] In the embodiment of the present application, the target equipment at least includes: a liquid preparation tank, a stirrer, and a feeding device.

[0041] As a mixing container for dry powder fracturing fluid, the mixing tank serves as a spatial carrier for dry powder dissolution, fluid mixing, and chemical reactions. The mixing tank provides a stable mixing space, and its shape, volume, and internal structure directly affect the fluid flow state and mixing efficiency.

[0042] Agitators use mechanical agitation to propel fluids, dispersing dry powders, accelerating dissolution, and achieving uniform mixing. Suitable for low-viscosity fluids, they feature a simple structure and are often used for initial dispersion of dry powders. Turbine agitators generate strong shear forces and are suitable for high-viscosity fluids, breaking up "gels" and promoting cross-linking reactions. Ribbon agitators scrape the bottom and sides of the tank, reducing dead zones and improving overall uniformity. The agitator speed and paddle type directly influence the flow distribution (e.g., vortex flow vs. laminar flow) and are key to preventing dry powder agglomeration and improving dissolution efficiency.

[0043] The dosing device precisely controls the quantitative delivery of dry powder (thickener, crosslinker, etc.) and dosing water. The dry powder dosing system utilizes screw conveyors and vibrating feeders to prevent moisture absorption and agglomeration of the dry powder and ensure uniform material delivery. The dosing water control system uses valves and flow meters to control the water flow rate to match the dry powder dosage (e.g., liquid-to-solid ratio control). This enables precise control of the material ratio, preventing premature crosslinking or gel breakage caused by uneven dosing.

[0044] Further optionally, the mixing process data at least includes: stirring speed, pressure in the liquid preparation tank, fluid temperature, dry powder addition amount, and liquid preparation water flow rate.

[0045] For example, assuming that the multi-source sensors include at least a pressure sensor, a temperature sensor, a flow sensor, and a liquid level sensor, in step 101, a pressure sensor and a liquid level sensor are deployed on the liquid mixing tank to collect real-time pressure and liquid level data within the tank; a speed sensor is installed on the agitator to obtain stirring speed data; flow sensors are respectively installed on the dry powder delivery pipeline and the liquid mixing water delivery pipeline of the feeding device to monitor the dry powder feeding amount and the liquid mixing water flow rate in real time; a temperature sensor is placed in the fluid area within the liquid mixing tank to collect fluid temperature data, and the collected data is transmitted in real time to the data processing terminal via the Internet of Things communication module.

[0046] Specifically, deploying multi-source sensors such as pressure, temperature, flow, and liquid level in the oilfield dry powder fracturing fluid mixing scenario and realizing real-time data transmission can significantly improve the safety, efficiency, and controllability of the mixing process through full-dimensional data collection and intelligent analysis. Specifically, the pressure sensor and liquid level sensor on the liquid mixing tank work together, not only to monitor the static pressure and volume changes of the fluid in the tank in real time, but also to indirectly evaluate the dry powder dissolution efficiency through density calculation, avoiding insufficient viscosity or pipeline blockage caused by insufficient dissolution; the agitator speed sensor directly links the stirring intensity and fluid shear force, providing key parameters for optimizing the stirring strategy. For example, for high-viscosity fluids, the speed is automatically increased to maintain a turbulent state to ensure uniform dispersion of the dry powder; the dual flow sensor of the feeding device realizes dynamic monitoring of the liquid-to-solid ratio. When the mixing water flow is stable and the dry powder feeding amount fluctuates, the system can automatically adjust the screw conveyor speed to control the liquid-to-solid ratio error within ±2%, which is 6% smaller than the manual adjustment error. Single-well operation can save 3-5% of dry powder materials (for example, 0.6-1 ton is saved when 20 tons of guar gum is consumed per well), significantly reducing material loss.

[0047] At the data transmission level, the IoT communication module supports millisecond-level data transmission. Compared to the 15-30 minute intervals of traditional manual inspections, it can capture instantaneous anomalies such as feed pipe blockage within 5 seconds and trigger an early warning. For example, when the pressure in the liquid mixing tank suddenly rises (>0.5MPa) and the dry powder flow rate suddenly drops (<50% of the set value), the system will automatically stop feeding, start the agitator to reverse at high speed to impact the blockage point, and issue an audible and visual alarm. This shortens the abnormal response time from tens of minutes for manual processing to within a few seconds of automatic control, significantly reducing the risk of equipment damage. At the same time, the "mixing log" formed by the full process data record can be traced back to the details of each feeding, stirring, and cross-linking. When batch quality problems arise, historical data such as temperature fluctuations can be used to quickly locate the fault period. For example, if the temperature exceeds the set range during a certain period, resulting in incomplete cross-linking, it provides a precise basis for process improvement.

[0048] In terms of intelligent control, sensor data is combined with PID regulation and fuzzy control algorithms to achieve dynamic optimization of mixing parameters. Taking temperature control as an example, when the measured temperature falls below the target value (e.g., 25°C), the system automatically adjusts the heating power proportional to the temperature difference (full power for a 5°C temperature difference, 30% power for a 2°C temperature difference), avoiding temperature overshoot and energy waste caused by traditional constant heating. When approaching the upper temperature limit (e.g., 45°C), air cooling is prioritized over direct power outage, improving temperature control accuracy to ±1°C and ensuring that the cross-linking reaction proceeds within the designed range. Furthermore, high-frequency fluctuation analysis of the pressure sensor (e.g., FFT spectrum) can proactively identify mechanical faults such as agitator bearing wear or loose tank baffles. Combined with accumulated data from the flow sensor, the screw conveyor screw replacement cycle is predicted (e.g., wear is checked every 500 tons of dry powder conveyed), reducing unplanned equipment downtime by over 40% and lowering maintenance costs.

[0049] At the process collaboration level, sensor data is deeply linked to the digital twin system, and real-time parameters such as the liquid level in the mixing tank and the speed of the agitator are synchronized to the 3D virtual model. Operators can conduct remote inspections using VR equipment, visually viewing the fluid flow status and equipment operation details, thereby improving monitoring efficiency. Integration with the fracturing construction scheduling system enables dynamic matching of mixing progress and pumping speed. When the volume of the mixing tank falls below the safe value, the pump group is automatically notified to slow down, avoiding the risk of "fluid outage". At the same time, by predicting the completion time of mixing and optimizing the process connection, the single-well operation cycle can be shortened by 10-15%. The data-driven mixing model also reduces the waste slurry rate from 10% in traditional processes to below 5%. Combined with precise proportioning and energy consumption optimization (such as two-stage mixing, which reduces energy consumption by 15%), the overall production cost is reduced by approximately 8-12%, providing technical support for the efficient development of unconventional oil and gas reservoirs.

[0050] 102. Establish a virtual model corresponding to the target device based on the internal structure of the target device.

[0051] In the embodiment of the present application, the virtual model is provided with a liquid dispensing tank, a stirrer, a feeding device, and a pipeline layout between each virtual device. Further optionally, the internal structure includes at least: a tank shape, a stirring paddle structure, and a pipeline layout.

[0052] For example, in 102, a three-dimensional geometric model of the liquid mixing tank, agitator, and feeding device is constructed according to the actual dimensional parameters of the target equipment; during the modeling process, internal structural details corresponding to the tank shape, agitator structure, and pipeline layout are generated in the three-dimensional geometric model; through model mapping technology, the three-dimensional geometric model with added internal structural details is imported into the digital twin platform to form a virtual model that corresponds one-to-one to the target equipment.

[0053] Specifically, the principle of building a virtual model based on the internal structure of the target device is to replicate the geometric features and spatial relationships of the physical entity through digital means. Specifically, through 3D modeling technology, the actual dimensional parameters such as the tank shape of the liquid distribution tank (such as the cylindrical cone bottom structure), the agitator structure (such as the size and angle of the propulsion or anchor blades), and the pipeline layout (such as the diameter, bending angle and interface position of the dry powder delivery pipeline with the tank body) are converted into a precise 3D geometric model. The model also fully presents the flow path direction and structural details related to the fluid mechanics characteristics of the mixing area (such as the position of the baffle and the size of the guide tube). This process imports the 3D model containing internal structural details into the digital twin platform through model mapping technology, so that the virtual model and the target device are completely aligned in terms of geometry, spatial layout and key structural features, forming a digital mirror image of the physical entity.

[0054] As can be understood, firstly, by accurately replicating the equipment's internal structure, the virtual model can realistically simulate the fluid flow path within the mixing tank, the vortex pattern generated by the impeller, and the mixing trajectory of the dry powder and mixing water, providing a visual analysis platform for process parameter optimization (for example, by predicting the shear force distribution of the fluid at different impeller speeds through simulation). Secondly, the digital preservation of internal structural details enables the virtual model to possess dynamic mapping capabilities. For example, when a pipeline in the actual equipment is clogged, the virtual model can use pressure sensor data to drive a color warning (such as red highlight) for the corresponding pipeline segment and, combined with fluid dynamics simulation, locate the blockage point. Furthermore, the complete geometric model lays the foundation for subsequent advanced applications such as stress analysis and heat transfer simulation. For example, these applications can assess pressure bearing capacity based on tank shape parameters or predict power consumption curves based on impeller structure, assisting in equipment selection and energy efficiency optimization. Overall, this internal structure-based virtual modeling transitions the physical equipment from physical existence to digital computability. It enables pre-validation, real-time monitoring, and retrospective analysis of the entire mixing process lifecycle (from design and commissioning to operation and maintenance) in a virtual space, significantly improving the accuracy and reliability of intelligent operations.

[0055] 103. Based on the Navier-Stokes equation and the continuity equation, fluid dynamics is used to simulate the fluid flow in the liquid mixing tank, and the flow field distribution, vortex area and stirring dead corner are analyzed to construct a real-time fluid motion model corresponding to the target equipment.

[0056] For example, in 103, the virtual model of the liquid mixing tank is imported into the computational fluid dynamics software; the physical property parameters of the fluid, including density, viscosity, etc., are set; based on the Navier-Stokes equations and the continuity equation, the finite volume method FVM or the finite element method FEM is used to numerically solve the fluid flow in the liquid mixing tank; by post-processing the calculation results, the flow field distribution is analyzed, and the vortex area and the stirring dead corner are identified; based on the analysis results, a real-time fluid motion model describing the fluid motion law in the liquid mixing tank is established; the particle image velocimetry PIV algorithm is used to perform simulation experiments to verify the computational fluid dynamics simulation results, and the real-time fluid motion model is optimized based on the verification results, and the real-time fluid motion model is associated with the virtual model; wherein, tracer particles are added to the liquid mixing tank experimental device, a particle image sequence in the simulated flow field is obtained, the particle motion velocity of each point in the flow field is calculated, and the accuracy of the fluid dynamics simulation results is compared and verified.

[0057] Specifically, the principle of simulating fluid flow in a distribution tank based on the Navier-Stokes equations and the continuity equation is to describe the mass and momentum conservation characteristics of the fluid through mathematical equations, and then use computational fluid dynamics (CFD) algorithms to discretize the continuous fluid space in the distribution tank into a finite number of control volumes (finite volume method FVM) or units (finite element method FEM). The flow velocity, pressure, and other physical quantities of each discrete unit are numerically solved to simulate the flow state of the fluid in the tank. Taking the finite volume method as an example, its core is to integrate the Navier-Stokes equations within each control volume, solve the physical quantities of each node through iterative calculation, and ultimately obtain data such as the velocity distribution, pressure distribution, and turbulent kinetic energy distribution of the entire flow field.

[0058] For example, consider a cylindrical tank with a diameter of 2 meters and a height of 3 meters, equipped with a turbine-type impeller (0.8 meter diameter, 200 rpm). The simulation objective is to analyze the flow distribution near the impeller and whether there are any dead zones at the bottom of the tank. First, a virtual model of the tank is imported into ANSYS Fluent software, with water (density 1000 kg / m³, viscosity 0.001 Pa·s) as the fluid. The RANS (Reynolds-averaged Navier-Stokes) equations are combined with the k-ε turbulence model for a closed-loop system of equations. The tank is meshed into 1 million tetrahedral cells, with local refinement performed in the impeller region. After convergence, post-processing results show that a high-speed turbulent region (flow velocity approximately 2.5 m / s) forms near the impeller. The flow in the upper portion of the tank exhibits a spiraling upward flow pattern. A low-speed vortex region (flow velocity <0.1 m / s) exists near the tank wall at the bottom, which is identified as a dead zone.

[0059] To validate the simulation results, fluorescent tracer particles (50 μm in diameter) were added to the physical experimental setup. A laser sheet illuminated the midsection of the tank, and a high-speed camera captured images of the particle motion at 200 fps. Using the particle image velocimetry (PIV) algorithm, cross-correlation analysis was performed on 50 consecutive image frames to calculate the velocity vectors at each point in the flow field. Comparison revealed that the predicted velocity in the agitator paddle region was within 5% of the experimental value, and the location of the low-velocity region at the tank bottom was consistent with experimental observations. However, the simulation did not fully capture the secondary vortices caused by the baffle design. Based on this, the baffle angle parameters of the virtual model were adjusted and the simulation was repeated. The velocity distribution error between the modified model and the experimental results was reduced to within 3%, ultimately establishing a highly accurate real-time fluid motion model.

[0060] After identifying dead zones at the tank bottom through simulation, researchers experimented with adding a diverter cone (60° angle) or adjusting the agitator installation height (from 0.5 to 0.8 meters) in the virtual model. The simulation results showed that the diverter cone increased the bottom flow velocity to 0.3 m / s, reducing the dead zone by 70%, providing a clear solution for physical equipment modification. Simulations of power consumption and fluid shear stress at different rotational speeds (150-250 rpm) revealed that reducing the rotational speed from 200 to 180 rpm reduced power consumption by 12%, but only decreased shear stress by 8%, and extended dissolution time by 5 minutes. Considering the requirements for fracturing fluid dissolution efficiency, 180 rpm was ultimately selected as the economical rotational speed, saving approximately 150 kWh of electricity per well.

[0061] For winter low-temperature scenarios (fluid viscosity increases to 0.002 Pa·s), simulations show that the fluid turbulence intensity decreases by 15% at the same rotational speed. Therefore, the impeller speed needs to be increased to 220 rpm to maintain dissolution efficiency and avoid project delays caused by on-site commissioning.

[0062] Through the closed loop of numerical simulation-experimental verification-model correction, the complex flow field in the mixing tank is accurately characterized, and the traditional experience-based adjustment of stirring parameters is transformed into data-driven scientific optimization. The fluid dynamic characteristics of the mixing process are transformed from invisible to calculable, verifiable, and optimizable, which significantly improves the uniformity and efficiency of fracturing fluid mixing, while reducing the cost of physical experiments and the risk of equipment debugging.

[0063] 104. Based on the principle of chemical kinetics, a chemical kinetic model of dry powder dissolution, cross-linking reaction and gel breaking reaction is established, and the material characteristic parameters are integrated into the chemical kinetic model.

[0064] In the embodiment of the present application, the chemical kinetics model is used to simulate the reaction process of the target device under different conditions.

[0065] Specifically, as an optional embodiment, in 104, the hygroscopicity curve and dissolution kinetic parameters of the polymer, the reaction rate constant and temperature sensitivity data of the cross-linking agent, and the parameters affecting dissolution and cross-linking of different water quality data are obtained; based on the principle of chemical kinetics, chemical kinetic models of dry powder dissolution, cross-linking reaction and gel breaking reaction are established respectively; a regression analysis algorithm is used to establish a regression model between water quality parameters and indicators related to dry powder dissolution and cross-linking reaction based on the collected experimental data, and the regression equation coefficients are determined by the least squares method; an artificial neural network ANN is used, with water quality parameters as input layer nodes and indicators related to dry powder dissolution and cross-linking reaction as output layer nodes, a hidden layer is set, and the neural network is trained through a large amount of experimental data to learn complex nonlinear relationships; the results obtained from the regression analysis and the ANN network are used as the model input of the chemical kinetic model, and the parameters of the chemical kinetic model are calibrated and verified through experimental data to ensure that the chemical kinetic model can simulate the reaction process of the target equipment under different conditions.

[0066] Taking the combination of guar gum dissolution and borate cross-linking reaction as an example, in the data collection and model establishment process, guar gum as a high molecular polymer has a hygroscopicity curve showing that the moisture content increases from 5% to 12% within 24 hours under a humidity greater than 60%. The dissolution kinetic parameter activation energy E a =45kJ / mol indicates that the dissolution rate increases by 1.8 times for every 10℃ increase in temperature; Borax is used as a crosslinking agent, and the reaction rate constant k=0.02min⁻¹ (25℃, pH=9). The temperature sensitivity data shows an Arrhenius exponential growth (E a =60kJ / mol), and cross-linking time was prolonged by over 50% when calcium and magnesium ion concentrations exceeded 500ppm. Based on this, dissolution and cross-linking models, as well as a bimolecular reaction model, were constructed, incorporating an ionic strength correction term. In regression analysis and neural network applications, a linear regression equation Y = 12.5 + 0.005X1 -1.2X2 was fitted, with salinity and pH as independent variables and cross-linking time as the dependent variable. The experimentally verified error was <5%. The artificial neural network was designed with a 3-node input layer (for salinity, pH, and temperature), 10-node hidden layer (with ReLU activation function), and 2-node output layer (for dissolution rate and cross-linking degree). After training with 200 sets of experimental data, the nonlinear fitting accuracy outperformed the regression model (MSE reduced by 30%). During model calibration and validation, the regression and ANN output data were input into a chemical kinetic model to adjust parameters. For example, at salinity = 1500ppm, the calibrated model predicted cross-linking time that was consistent with the experimental value. In highly salinized water, the model's predicted dissolution and cross-linking times were close to the measured values, demonstrating its adaptability to complex working conditions.

[0067] In the precise prediction and control of chemical reactions, model analysis and adjustment of the dry nitrogen purge process reduced the guar gum dissolution time from 30 minutes to 18 minutes, improving efficiency by 40%. The use of a titanate crosslinker in low-temperature well operations reduced the crosslinking time from >20 minutes to 8 minutes. Faced with complex operating conditions, when the salinity of the oilfield reinjection water increased, the model triggered a water quality treatment procedure to avoid construction interruptions. The gel-breaking reaction model set an upper limit for ammonium persulfate dosage and a cooling system, reducing the risk of gel-breaker failure from 5% to less than 1%. In terms of process costs, model-based formulation optimization reduced guar gum usage, saving approximately 12,000 yuan per well. Virtual model rehearsals reduced physical experiments by 50%, saving approximately 60% in reagent costs and time. In terms of intelligent decision support, the model's abnormality warning response speed was 80% faster than manual intervention. Retrospective analysis addressed the low fracturing fluid flowback rate and improved it to over 75%.

[0068] In summary, the chemical kinetics model that integrates regression analysis and artificial neural networks achieves multi-dimensional and precise modeling of the dissolution, cross-linking, and gel-breaking reactions during the mixing process of dry powder fracturing fluid, breaking through the limitations of traditional empirical formulas, making the reaction process calculable, interventional, and traceable, improving the adaptability of the mixing process to multiple materials and complex water quality, reducing material costs and experimental risks, and providing core technical support for the intelligent and efficient oilfield fracturing operations.

[0069] 105. Input the mixing process data into the virtual model, dynamically calibrate the virtual model through a real-time fluid motion model and a chemical kinetic model, dynamically adjust the water quality data, material ratio, and stirring conditions, and perform multi-scenario real-time simulation analysis on the mixing process to obtain mixing simulation data of the target device; and perform corresponding dry powder fracturing fluid mixing control on the target device according to the mixing simulation data.

[0070] First, in 105, the real-time collected mixing process data is input into the interface corresponding to the virtual model. The real-time fluid motion model is used to update and simulate the fluid flow state in the mixing tank based on the computational fluid dynamics algorithm. The chemical reaction process is updated and simulated using the chemical kinetics model combined with the regression analysis algorithm and the artificial neural network algorithm. According to the preset multi-scenario simulation rules, a genetic algorithm is used to treat the ratio of thickener, crosslinker, and additive as chromosomal genes, and a fitness function is defined based on the fracturing fluid performance as the evaluation standard. The material ratio is optimized through selection, crossover, and mutation operations. Fracturing fluid properties include: crosslinking time, gel breaking time, and fracturing fluid viscosity. The response surface methodology is used to arrange experimental points through central composite design and Box-Behnken design. Regression analysis is performed on the experimental data to establish a response surface model and determine the optimal material ratio range. Under different parameter combinations, based on the optimized material ratio and the optimal material ratio range, the real-time fluid motion model and chemical kinetics model are respectively run to simulate the mixing process in real time. Key data during the simulation is recorded to form mixing simulation data for the target equipment.

[0071] For example, in an intelligent oilfield fracturing fluid mixing system, a solution based on real-time data-driven virtual models and multi-algorithm collaborative optimization can achieve efficient and precise design of material ratios. First, real-time data collected during the mixing process, such as pressure, temperature, flow rate, and liquid level, is input into the virtual model interface. A real-time fluid motion model uses a computational fluid dynamics algorithm to update the flow field distribution within the mixing tank (e.g., vortex area and velocity vector). A chemical kinetic model, combined with regression analysis and an artificial neural network algorithm, dynamically simulates the dissolution, crosslinking, and gel-breaking reaction processes (e.g., viscosity curve and reaction conversion rate). Subsequently, to optimize the ratio of thickeners, crosslinkers, and additives, a genetic algorithm encodes the addition ratio of each component into chromosomal genes. A fitness function is constructed based on performance indicators such as crosslinking time, gel-breaking time, and fracturing fluid viscosity. The population is iteratively evolved through selection (e.g., roulette wheel selection), crossover (two-point crossover), and mutation (Gaussian mutation). For example, 100 random mixing solutions are initially generated, and after 20 generations of evolution, the optimal solution is converged.

[0072] The response surface methodology (RSM) employs a central composite design (e.g., three-factor, five-level) or Box-Behnken design to arrange experimental points. A quadratic polynomial regression fit is performed on the experimental data to construct a response surface model (e.g., Y = β0 + ∑βᵢXᵢ + ∑βᵢᵢXᵢ² + ∑βᵢⱼXᵢXⱼ). Contour plots or three-dimensional surface plots visualize the relationship between mix parameters and fracturing fluid performance, identifying the optimal mix region (e.g., the parameter combination with the shortest crosslinking time). Under different operating parameter combinations (e.g., temperature between 20°C and 60°C, salinity between 500 and 3000 ppm), the optimized mix ratios derived from the genetic algorithm and the optimal region parameters determined by the RSM are input into real-time fluid motion and chemical kinetic models. Multiple simulations are then conducted to verify the model. Key data such as stirring power, dissolution time, and crosslinking strength are recorded, resulting in a simulation database covering over 500 operating conditions.

[0073] A genetic algorithm revealed that crosslinker concentration influences fracturing fluid viscosity by 45% (temperature influences 25%, pH influences 30%), leading to adjustments in crosslinker addition strategies. A response surface model predicted that the optimal viscosity retention (92%) was achieved at 40°C, pH 9, and a crosslinker concentration of 0.8%. The measured value was 90.5%, with an error of less than 2%. In a shale gas fracturing operation, this solution reduced guar gum dosage from 35 kg / m³ to 32 kg / m³, controlled crosslinking time to 8-12 minutes (design target 10 ± 2 minutes), and achieved a breaker fluid viscosity of less than 5 mPa·s (target value <10 mPa·s). This saved 12,000 yuan in material costs per well and increased operational efficiency by 15%. By integrating real-time data with multiple algorithms, the system has achieved a leap from "trial and error" to "data-driven precision design," providing an intelligent solution for fracturing treatment in complex oil and gas reservoirs.

[0074] Next, at step 105, the mixing simulation data is analyzed to extract at least one key performance indicator (KPI) of the fracturing fluid: dissolution uniformity, crosslinking performance, or gel-breaking time. The KPI is then compared with a preset standard parameter range. If the KPI exceeds the standard parameter range, a corresponding control instruction is generated according to a preset control strategy. The control instruction is used to adjust the agitator speed, change the feeding rate of the feeding device, and adjust the additive injection rate. The control instruction is then sent to the actuator of the target device via the data transmission module, thereby controlling the dry powder fracturing fluid mixing of the target device.

[0075] For example, in the control of dry powder fracturing fluid mixing in oilfields, the analysis of mixing simulation data and the subsequent execution of control instructions are key to ensuring fracturing fluid quality and operational safety. This data-driven process enables closed-loop management from performance evaluation and anomaly diagnosis to precise control.

[0076] In actual operations, the system first extracts key performance indicators from the mixing simulation data. For example, computational fluid dynamics simulation results are used to evaluate the uniformity of fracturing fluid dissolution. If the standard deviation of the fluid viscosity in the mixing tank exceeds 0.3 mPa·s, the dissolution is determined to be uneven. The crosslinking time and degree of crosslinking output by the chemical kinetic model are used to determine whether the crosslinking performance meets the standard. For example, if the crosslinking time exceeds ±10% of the design value (assuming a designed crosslinking time of 10 minutes, a range of 9-11 minutes is considered abnormal), or if the viscosity of the fracturing fluid after crosslinking does not reach the target value (for example, below 80 mPa·s). The progress of the gel-breaking reaction is monitored to determine whether the gel-breaking time meets the requirements. If the gel-breaking time is too long (over 12 hours) or the residual viscosity of the liquid after gel-breaking is too high (>5 mPa·s), both indicate abnormal gel-breaking performance.

[0077] The system then compares these key performance indicators against pre-set standard parameter ranges. These standard parameters are developed based on extensive experimental data and field operation experience, covering the performance requirements of fracturing fluids under different geological conditions and construction techniques. For example, for a low-permeability reservoir, the pre-set fracturing fluid dissolution uniformity must meet viscosity standard deviations of less than 0.2 mPa·s, crosslinking time controlled between 9 and 11 minutes, gel breaking time between 8 and 10 hours, and residual viscosity after gel breaking of less than 3 mPa·s.

[0078] If a key performance indicator falls outside the standard parameter range, the system immediately generates corresponding control instructions based on the pre-set control strategy. For example, if it detects uneven dissolution of the fracturing fluid, the system will generate an instruction to increase the agitator speed (e.g., from 200 rpm to 250 rpm) to enhance fluid shear and accelerate dry powder dissolution. If the crosslinking time is too long, indicating insufficient crosslinker addition, the system will issue an instruction to increase the crosslinker injection rate (e.g., from 5 L / min to 7 L / min). If the gel breaking time is too long, the system will automatically adjust the gel breaker injection rate (e.g., increasing the ammonium persulfate addition ratio from 0.8% to 1.0%).

[0079] Finally, control commands are sent to the target device's actuators via a data transmission module (such as 5G or Industrial Ethernet). These actuators respond quickly to the commands, such as adjusting the speed of the agitator's motor inverter, changing the flow rate of the dosing device's metering pump, and adjusting the valve opening of the additive injection system, achieving precise control of the target device. During command execution, the system continuously monitors the equipment's operating status and changes in the fracturing fluid's properties, forming a closed-loop control system consisting of data collection, analysis, decision-making, command execution, and feedback on results.

[0080] This control method significantly improves the automation and precision of fracturing fluid mixing. Compared to traditional manual control, its response speed is over 80% faster, enabling responses to abnormal indicators within 30 seconds. The control accuracy of key performance indicators has increased by 30%, for example, narrowing the fluctuation range of cross-linking time from ±2 minutes to ±1 minute. This effectively reduces the construction risks caused by unstable fracturing fluid performance, while also reducing material waste and production costs, providing a strong guarantee for efficient oilfield development.

[0081] In the embodiments of the present application, it helps to improve the quality and efficiency of fracturing fluid mixing, enhances the adaptability to complex production environments, and assists in improving the development efficiency of unconventional oil and gas reservoirs.

[0082] In another embodiment of the present application, a mixing control device for oilfield dry powder fracturing fluid is provided. Figure 2 As shown, the device includes the following units:

[0083] The acquisition unit is configured to deploy multiple sensors in a target device to collect mixing process data in real time; wherein the target device includes at least a liquid preparation tank, an agitator, and a feeding device; and the mixing process data includes at least agitation speed, pressure in the liquid preparation tank, fluid temperature, dry powder feeding amount, and liquid preparation water flow rate;

[0084] The first construction unit is configured to establish a virtual model corresponding to the target device based on the internal structure of the target device; wherein the virtual model includes a liquid mixing tank, an agitator, a feeding device, and a pipeline layout between each virtual device; the internal structure includes at least: a tank shape, a stirring paddle structure, and a pipeline layout;

[0085] The second construction unit is configured to use fluid dynamics to simulate the fluid flow in the liquid mixing tank based on the Navier-Stokes equations and the continuity equation, and analyze the flow field distribution, vortex areas, and stirring dead corners to construct a real-time fluid motion model corresponding to the target equipment;

[0086] The third building block is configured to establish a chemical kinetic model for dry powder dissolution, crosslinking reaction, and gel breaking reaction based on chemical kinetic principles, and incorporate material characteristic parameters into the chemical kinetic model; the chemical kinetic model is used to simulate the reaction process of the target device under different conditions;

[0087] The control unit is configured to input the mixing process data into a virtual model, dynamically calibrate the virtual model through a real-time fluid motion model and a chemical kinetic model, dynamically adjust water quality data, material ratios, and stirring conditions, and perform multi-scenario real-time simulation analysis of the mixing process to obtain mixing simulation data of a target device; and execute corresponding dry powder fracturing fluid mixing control on the target device according to the mixing simulation data.

[0088] This device can implement each step in the above-mentioned method for controlling the mixing of oilfield dry powder fracturing fluid, which will not be described in detail here.

[0089] In another embodiment of the present application, an electronic device is provided, including: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory is used to store computer programs; the processor is used to implement the mixing control method of the oilfield dry powder fracturing fluid described in the method embodiment when executing the program stored in the memory. The communication bus 1140 mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 1140 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0090] The communication interface 1120 is used for communication between the electronic device and other devices.

[0091] The memory 1130 may include a random access memory (RAM) or a non-volatile memory (non-volatile memory), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0092] The processor 1110 may be a general purpose processor, including a central processing unit (CPU).

[0093] It can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0094] Accordingly, an embodiment of the present application further provides a computer-readable storage medium storing a computer program, which, when executed, can implement the steps that can be performed by the electronic device in the above method embodiment.

Claims

1. A mixing control method for oilfield dry powder fracturing fluid, characterized in that: include: Deploy multi-source sensors in the target equipment to collect real-time mixing process data; the target equipment includes at least: a liquid mixing tank, an agitator, and a feeding device; the mixing process data includes at least: stirring speed, pressure in the liquid mixing tank, fluid temperature, dry powder feeding amount, and liquid mixing water flow rate; Based on the internal structure of the target device, a virtual model corresponding to the target device is established; wherein the virtual model is provided with a liquid mixing tank, an agitator, a feeding device, and a pipeline layout between each virtual device; the internal structure includes at least: the tank shape, the agitator structure, and the pipeline layout; Based on the Navier-Stokes equations and the continuity equation, fluid dynamics is used to simulate the fluid flow in the liquid mixing tank. The flow field distribution, vortex area, and stirring dead corners are analyzed to construct a real-time fluid motion model corresponding to the target equipment. Based on the principles of chemical kinetics, a chemical kinetic model for dry powder dissolution, crosslinking reaction, and gel breaking reaction is established, and material characteristic parameters are integrated into the chemical kinetic model. The chemical kinetic model is used to simulate the reaction process of the target equipment under different conditions. Inputting the mixing process data into a virtual model, dynamically calibrating the virtual model through a real-time fluid motion model and a chemical kinetic model, dynamically adjusting water quality data, material ratios, and stirring conditions, and performing multi-scenario real-time simulation analysis of the mixing process to obtain mixing simulation data for a target device; and executing corresponding dry powder fracturing fluid mixing control on the target device based on the mixing simulation data; Among them, based on the principle of chemical kinetics, a chemical kinetic model of dry powder dissolution, cross-linking reaction and gel breaking reaction is established, and material characteristic parameters are integrated into the chemical kinetic model, including: obtaining the hygroscopicity curve and dissolution kinetic parameters of the polymer, the reaction rate constant and temperature sensitivity data of the cross-linking agent, and the parameters affecting dissolution and cross-linking of different water quality data; based on the principle of chemical kinetics, chemical kinetic models of dry powder dissolution, cross-linking reaction and gel breaking reaction are established respectively; a regression analysis algorithm is used to establish a regression model between water quality parameters and indicators related to dry powder dissolution and cross-linking reaction based on the collected experimental data, and the regression equation coefficients are determined by the least squares method; an artificial neural network ANN is used, with water quality parameters as input layer nodes and indicators related to dry powder dissolution and cross-linking reaction as output layer nodes, a hidden layer is set and the neural network is trained through a large amount of experimental data to learn complex nonlinear relationships; the results obtained from the regression analysis and the ANN network are used as the model input of the chemical kinetic model, and the parameters of the chemical kinetic model are calibrated and verified through experimental data to ensure that the chemical kinetic model can simulate the reaction process of the target equipment under different conditions.

2. The mixing control method for oilfield dry powder fracturing fluid according to claim 1, characterized in that: The multi-source sensors include at least: a pressure sensor, a temperature sensor, a flow sensor, and a liquid level sensor; the multi-source sensors are deployed in the target device to collect mixing process data in real time, including: Deploy pressure sensors and liquid level sensors on the liquid distribution tank to collect real-time pressure and liquid level data in the liquid distribution tank; Install a speed sensor on the stirrer to obtain stirring speed data; Flow sensors are installed on the dry powder delivery pipeline and the liquid water delivery pipeline of the feeding device to monitor the dry powder feeding amount and the liquid water flow in real time; Temperature sensors are arranged in the fluid area of ​​the liquid distribution tank to collect fluid temperature data, and the collected data are transmitted to the data processing terminal in real time through the Internet of Things communication module.

3. The mixing control method for oilfield dry powder fracturing fluid according to claim 1, characterized in that: The step of establishing a virtual model corresponding to the target device based on the internal structure of the target device includes: Construct 3D geometric models of the liquid mixing tank, agitator, and feeding device according to the actual dimensional parameters of the target equipment; During the modeling process, the internal structural details of the tank shape, agitator structure, and pipeline layout are generated in the three-dimensional geometric model; through model mapping technology, the three-dimensional geometric model with added internal structural details is imported into the digital twin platform to form a virtual model that corresponds one-to-one to the target equipment.

4. The mixing control method for oilfield dry powder fracturing fluid according to claim 1, characterized in that: Based on the Navier-Stokes equation and the continuity equation, fluid dynamics is used to simulate the fluid flow in the liquid mixing tank, and the flow field distribution, eddy current area and stirring dead angle are analyzed to construct a real-time fluid motion model corresponding to the target equipment, including: Import the virtual model of the liquid distribution tank into the computational fluid dynamics software; Set the physical property parameters of the fluid, including density and viscosity; Based on the Navier-Stokes equation and the continuity equation, the finite volume method FVM or the finite element method FEM is used to numerically solve the fluid flow in the liquid distribution tank; By post-processing the calculation results, the flow field distribution is analyzed, and the vortex area and stirring dead corner are identified; Based on the analysis results, a real-time fluid motion model is established to describe the fluid motion law in the liquid distribution tank; The particle image velocimetry (PIV) algorithm was used to conduct simulation experiments to verify the computational fluid dynamics simulation results. Based on the verification results, the real-time fluid motion model was optimized and associated with the virtual model. Tracer particles were added to the liquid mixing tank experimental device to obtain a sequence of particle images in the simulated flow field, and the particle motion velocity at each point in the flow field was calculated to compare and verify the accuracy of the fluid dynamics simulation results.

5. The mixing control method for oilfield dry powder fracturing fluid according to claim 1, characterized in that: The mixing process data is input into the virtual model, the virtual model is dynamically calibrated through a real-time fluid motion model and a chemical kinetic model, water quality data, material ratio and stirring conditions are dynamically adjusted, and a multi-scenario real-time simulation analysis of the mixing process is performed to obtain mixing simulation data of the target equipment, including: Input the real-time collected mixing process data into the corresponding interface of the virtual model; Through the real-time fluid motion model, the fluid flow state in the liquid distribution tank is updated and simulated based on the computational fluid dynamics algorithm. Through the chemical kinetic model, the chemical reaction process is updated and simulated in combination with the regression analysis algorithm and the artificial neural network algorithm. Based on pre-set multi-scenario simulation rules, a genetic algorithm is used to treat the ratio of thickeners, crosslinkers, and additives as chromosome genes. A fitness function is defined based on the performance of the fracturing fluid. The material ratio is optimized through selection, crossover, and mutation operations. Fracturing fluid performance includes crosslinking time, gel breaking time, and fracturing fluid viscosity. Using the response surface methodology, the experimental points were arranged through central composite design and Box-Behnken design, and the response surface model was established by regression analysis of the experimental data to determine the optimal material ratio area; Under different parameter combinations, based on the optimized material ratio and the optimal material ratio area, the real-time fluid motion model and chemical kinetics model are run respectively to simulate the mixing process in real time, record the key data in the simulation process, and form the mixing simulation data of the target equipment.

6. The mixing control method for oilfield dry powder fracturing fluid according to claim 5, characterized in that: The performing corresponding dry powder fracturing fluid mixing control on the target equipment according to the mixing simulation data includes: Analyze the mixing simulation data to extract at least one key performance indicator of the fracturing fluid, including dissolution uniformity, crosslinking performance, and gel breaking time; and compare the key performance indicator with a preset standard parameter range; If the key performance indicators exceed the standard parameter range, corresponding control instructions are generated according to the preset control strategy; among which, the control instructions are used to adjust the agitator speed, change the feeding speed of the feeding device, and adjust the injection amount of additives; The control instructions are sent to the actuator of the target device through the data transmission module to realize the mixing control of the dry powder fracturing fluid of the target device.

7. A mixing control device for oilfield dry powder fracturing fluid, characterized in that: The device comprises: The acquisition unit is configured to deploy multiple sensors in a target device to collect mixing process data in real time; wherein the target device includes at least a liquid preparation tank, an agitator, and a feeding device; and the mixing process data includes at least agitation speed, pressure in the liquid preparation tank, fluid temperature, dry powder feeding amount, and liquid preparation water flow rate; The first construction unit is configured to establish a virtual model corresponding to the target device based on the internal structure of the target device; wherein the virtual model includes a liquid mixing tank, an agitator, a feeding device, and a pipeline layout between each virtual device; the internal structure includes at least: a tank shape, a stirring paddle structure, and a pipeline layout; The second construction unit is configured to use fluid dynamics to simulate the fluid flow in the liquid mixing tank based on the Navier-Stokes equations and the continuity equation, and analyze the flow field distribution, vortex areas, and stirring dead corners to construct a real-time fluid motion model corresponding to the target equipment; The third building block is configured to establish a chemical kinetic model for dry powder dissolution, crosslinking reaction, and gel breaking reaction based on chemical kinetic principles, and incorporate material characteristic parameters into the chemical kinetic model; the chemical kinetic model is used to simulate the reaction process of the target device under different conditions; A control unit is configured to input the mixing process data into a virtual model, dynamically calibrate the virtual model through a real-time fluid motion model and a chemical kinetic model, dynamically adjust water quality data, material ratios, and stirring conditions, and perform multi-scenario real-time simulation analysis of the mixing process to obtain mixing simulation data of a target device; and execute corresponding dry powder fracturing fluid mixing control on the target device according to the mixing simulation data; The third construction unit establishes a chemical kinetic model of dry powder dissolution, cross-linking reaction and gel breaking reaction based on the principle of chemical kinetics. When incorporating material characteristic parameters into the chemical kinetic model, it is specifically configured as follows: obtaining the hygroscopicity curve and dissolution kinetic parameters of the polymer, the reaction rate constant and temperature sensitivity data of the cross-linking agent, and the parameters affecting dissolution and cross-linking of different water quality data; establishing chemical kinetic models of dry powder dissolution, cross-linking reaction and gel breaking reaction respectively based on the principle of chemical kinetics; using a regression analysis algorithm, based on the collected experimental data, establishing a regression model between water quality parameters and indicators related to dry powder dissolution and cross-linking reaction, and determining the regression equation coefficients by the least squares method; using an artificial neural network ANN, taking water quality parameters as input layer nodes and indicators related to dry powder dissolution and cross-linking reaction as output layer nodes, setting a hidden layer and training the neural network through a large amount of experimental data to learn complex nonlinear relationships; using the results obtained from the regression analysis and the ANN network as the model input of the chemical kinetic model, and calibrating and verifying the parameters of the chemical kinetic model through experimental data to ensure that the chemical kinetic model can simulate the reaction process of the target equipment under different conditions.

8. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the mixing control method of the oilfield dry powder fracturing fluid according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium, characterized in that The storage medium stores a computer software program, and when the computer software program is executed by the processor, the mixing control method for oilfield dry powder fracturing fluid according to any one of claims 1 to 6 is implemented.

Images