Enhanced monitoring network-based carbon dioxide sequestration amount real-time monitoring method

Abstract

The invention discloses a carbon dioxide sequestration amount real-time monitoring method based on an enhanced monitoring network, and the method comprises the steps: deploying a multi-level sensor network in a target sequestration region, and collecting multi-scale fluid multi-dimensional monitoring data in a carbon dioxide sequestration process in real time through the sensor network; constructing a geological migration dynamic model for the multi-scale fluid multi-dimensional monitoring data based on a phase field method and a dynamic grid method, and obtaining dynamic evolution data of a geological fluid-solid interface in a carbon dioxide sequestration process; carrying out fusion processing on the dynamic evolution data by adopting a data fusion algorithm, and dynamically estimating the current total carbon dioxide sequestration amount and the spatial distribution thereof; an early warning mechanism is set to obtain key monitoring data of the carbon dioxide sequestration state, real-time monitoring of the carbon dioxide sequestration amount is achieved, adaptability and predictive capacity to complex geological environments are improved, carbon dioxide sequestration monitoring precision and response rate are improved, and dynamic perception and risk prevention and control of the carbon dioxide sequestration process can be effectively supported.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon dioxide storage, and in particular to a method for real-time monitoring of carbon dioxide storage volume based on an enhanced monitoring network. Background Art

[0002] As global climate change becomes increasingly severe, carbon capture, utilization and storage (CCUS) technology, as one of the important means to achieve the "dual carbon" goals, is receiving widespread attention. Among them, geological storage of carbon dioxide is a key link in reducing the concentration of greenhouse gases in the atmosphere. However, how to accurately, continuously and dynamically monitor the migration status, storage efficiency and potential leakage risks of carbon dioxide during the storage process remains a core challenge facing current technology.

[0003] In recent years, with the development of emerging technologies such as the Internet of Things (IoT), artificial intelligence (AI), fiber optic sensing, and edge computing, building an enhanced intelligent monitoring network has become a new direction for improving CO2 storage monitoring capabilities. By combining multi-source heterogeneous sensor fusion, distributed data collection, and cloud-based intelligent analysis, it is possible to achieve all-weather, three-dimensional perception of key parameters such as temperature, pressure, gas concentration, and surface deformation in the storage area, providing strong support for the safety assessment, dynamic regulation, and long-term management of CO2 storage.

[0004] The current real-time monitoring methods for CO2 storage still have the following problems:

[0005] (1) Currently, CO2 storage monitoring methods mainly include seismic exploration, downhole logging, pressure and temperature monitoring, etc. Although these methods can reflect formation changes to a certain extent, they still have problems such as response lag, low spatial resolution, and difficulty in achieving real-time continuous monitoring;

[0006] (2) The use of a single sensor or discrete sampling method cannot meet the needs of large-scale storage sites for multi-dimensional, high-precision, and intelligent monitoring. Summary of the Invention

[0007] The purpose of the present invention is to provide a real-time monitoring method for carbon dioxide storage based on an enhanced monitoring network to solve the technical problems in the prior art such as delayed response of CO2 storage monitoring, low spatial resolution, and difficulty in achieving real-time continuous monitoring.

[0008] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0009] The present invention provides a method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network, comprising the following steps:

[0010] Deploy a multi-level sensor network within the target storage area, comprising surface monitoring nodes, underwater monitoring nodes, and remote transmission nodes, to collect multi-scale and multi-dimensional monitoring data of the CO2 storage process in real time through the sensor network;

[0011] Constructing a geological migration dynamic model based on the multi-scale and multi-dimensional fluid monitoring data and the phase field method and the dynamic grid method to obtain the dynamic evolution data of the geological fluid-solid interface during the carbon dioxide storage process;

[0012] fusing the dynamically evolving data using a data fusion algorithm to generate comprehensive monitoring information in a unified format, analyzing the comprehensive monitoring information using an enhanced machine learning model to dynamically estimate the current total amount of carbon dioxide stored and its spatial distribution;

[0013] By evaluating the carbon dioxide storage status in real time, setting up an early warning mechanism and automatically starting the auxiliary verification process, key monitoring data on the carbon dioxide storage status is obtained, and blockchain technology is used to distribute the key monitoring data and early warning records to obtain real-time and traceable carbon dioxide storage monitoring data.

[0014] As a preferred solution of the present invention, real-time collection of multi-scale and multi-dimensional monitoring data of the fluid during the carbon dioxide storage process by the sensor network includes:

[0015] The multi-scale fluid multi-dimensional monitoring data at least includes pressure data, solute migration data, gas concentration distribution data and geological deformation data of multi-scale fluid flow;

[0016] collecting temporal and spatial scale data during the carbon dioxide storage process through the multi-level sensor network;

[0017] A spatial distribution map is constructed for the spatial scale data based on the time scale data to obtain geomechanical dimension data of geological pores during the carbon dioxide injection and storage process.

[0018] As a preferred solution of the present invention, a phase field model is constructed based on the phase field method for the multi-scale fluid multi-dimensional monitoring data, including:

[0019] According to the spatial distribution map, a multi-component and multi-mineral state evolution mathematical model is constructed for the geomechanical dimension data by simulating the phase field model of gas-liquid two-phase flow.

[0020]

[0021] in, represents the total concentration of each component, represents the concentration gradient, D represents the molecular diffusion coefficient of the solute in the solution, e mrepresents the reaction rate on the surface of mineral m, θ m represents the fluid-solid interface on the mineral m surface, μ represents the fluid velocity, and t represents time;

[0022] In the phase field model, the phase field variables are controlled by a step function, the phase field variables are used as independent variables and controlled by the Cahn-Hilliard equation, and the phase field data are coupled with the flow field and the solute field. The expression is:

[0023]

[0024] Where w represents the flow velocity, t represents time, and α represents the fluid density. represents the pressure gradient, μ represents the dynamic viscosity of the fluid on the mineral surface, represents the square of the difference in fluid velocity at the fluid-solid interface, and H(Φ) represents the step function;

[0025] Finite element analysis software is used to solve the multi-component and multi-mineral state evolution mathematical model to obtain multiphase flow-solid field data.

[0026] As a preferred embodiment of the present invention, numerical analysis is performed on the multiphase flow-solid field data to obtain the reaction rate of the fluid-solid interface, including:

[0027] Expand the corresponding spatial distribution map of the multiphase flow-solid field data through a Python simulation program to establish a reaction relationship between the mineral surface and the fluid;

[0028] By calculating the energy of particles that can nucleate on a specific surface area within Δt time, the free energy change function Q at the fluid-solid interface is constructed. s , whose expression is:

[0029]

[0030] Where A represents the fluid-solid interface factor, δ represents the fluid-solid interface free energy, lnΩ represents the supersaturation function, K0 represents the initial surface area ratio of the medium, and T represents the fluid-solid interface temperature;

[0031] The free energy change function Q s The volume mesh of the interface between the fluid and the solid is defined, and the reaction rate at the fluid-solid interface is labeled using an exponential probability distribution function.

[0032] As a preferred solution of the present invention, a geological transport dynamic model is constructed in combination with the dynamic grid method to obtain dynamic evolution data of the geological fluid-solid interface during the carbon dioxide storage process, including:

[0033] According to the free energy change function Q at the fluid-solid interface sThe spatial distribution map is meshed using a moving mesh method to segment the fluid-solid interface on the spatial distribution map into 300×300 original data blocks;

[0034] Decompose the volume grid into a 31×23 domain decomposition map, associate the original data block of each subdomain with a cell of the decomposition map, and calculate the porosity size in each cell;

[0035] The geological transport dynamic model state of the geological fluid-solid interface is derived according to the porosity size to obtain real-time dynamic evolution data.

[0036] As a preferred solution of the present invention, a geological evolution scale analysis model based on the carbon dioxide injection amount is established for the dynamic evolution data, including:

[0037] Using the time scale data as the time axis, the pore pressure within each porosity size is analyzed, and the scale of geological evolution is determined by the range of pore pressure disturbance, wherein the pore pressure disturbance includes the accumulation and diffusion of the corresponding pore pressure;

[0038] During the CO2 injection process, the influence of the pore pressure and the displacement of the geological fluid-solid interface fault is evaluated;

[0039] A nonlinear regression equation is established between the CO2 injection parameters and the factors affecting the geological evolution scale, and its expression is:

[0040] M max =alog(X)+b

[0041] Where a and b represent the factors affecting the scale of geological evolution, and X represents the goodness of fit of the nonlinear regression equation. The greater the goodness of fit, the greater the influence of the corresponding factor on the scale of geological evolution.

[0042] The impact threshold of the geological evolution scale is set through the nonlinear regression equation, and a geological evolution scale analysis model is constructed to obtain geological evolution status data.

[0043] As a preferred solution of the present invention, a data fusion algorithm is used to fuse the geological evolution state data to generate comprehensive monitoring information in a unified format, including:

[0044] Cleaning and denoising the geological evolution state data, converting the data into a unified format using a Z-score normalization method, and obtaining geological evolution state data in a unified format;

[0045] The principal component analysis method is used to extract key characteristic variables from the geological evolution state data in the unified format, and a spatiotemporal matching matrix is ​​constructed by combining the time scale data and the spatial scale data. The key characteristic variables are aligned in a unified spatiotemporal coordinate system to form a structured data set with temporal correlation.

[0046] Performing fusion modeling on the structured data set to output comprehensive monitoring indicators reflecting the overall status of the storage area;

[0047] According to the historical data performance and real-time feedback mechanism, the weight coefficient of each sensor node or data dimension in the fusion process is dynamically adjusted to obtain comprehensive monitoring information with dynamic weight distribution.

[0048] As a preferred solution of the present invention, an enhanced machine learning model is used to analyze the comprehensive monitoring information to dynamically estimate the current total amount of carbon dioxide stored and its spatial distribution, including:

[0049] Long short-term memory (LSTM) network was used as an enhanced machine learning model to train the historical dataset, and the characteristic parameters of the LSTM network were adjusted through cross-validation.

[0050] Using the trained model to conduct real-time analysis of the comprehensive monitoring information, dynamically estimate the current carbon dioxide storage capacity and predict future trends;

[0051] A geological model was established using 3D visualization tools combined with GIS to simulate the spatial distribution of the total amount of carbon dioxide storage.

[0052] As a preferred solution of the present invention, by real-time assessment of the carbon dioxide storage status, setting up an early warning mechanism and automatically starting an auxiliary verification process, key monitoring data of the carbon dioxide storage status is obtained, including:

[0053] Using the geological evolution data, the carbon dioxide storage status is evaluated in real time, risk indicators are calculated, and warning thresholds and rules for each risk indicator are set;

[0054] When risk indicators exceed the threshold, the corresponding level of early warning mechanism is automatically triggered and notifications are sent to relevant personnel, and auxiliary verification processes are automatically initiated, including increasing monitoring frequency, calling additional sensor data, or dispatching on-site technicians;

[0055] Collect and analyze key monitoring data to confirm the actual status of the storage system, feed back auxiliary verification results to the monitoring system to update risk assessment model parameters, and record early warning events to form a knowledge base.

[0056] As a preferred solution of the present invention, blockchain technology is used to perform distributed storage of the key monitoring data and early warning records to obtain real-time and traceable carbon dioxide storage monitoring data, including:

[0057] Standardize the key monitoring data and early warning records, build a blockchain network consisting of multiple nodes, package the pre-processed key monitoring data into blocks, add timestamps and digital signatures, and then upload them to the blockchain;

[0058] Use smart contracts to manage data access rights and achieve real-time data synchronization in the blockchain network;

[0059] Through API-based query interfaces and visualization tools, a real-time retrieval and analysis model for historical data is established, providing an intuitive visualization interface.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] The present invention constructs a multi-level sensor network, collects time-scale data and space-scale data, establishes the correlation between the time evolution law of the carbon dioxide migration process and the spatial distribution characteristics, and uses time series to drive spatial modeling, which helps to reveal the migration path and storage efficiency of carbon dioxide in different time periods and improve the model prediction ability. The phase field method is introduced to model the gas-liquid two-phase flow behavior during the carbon dioxide injection process, which can more realistically reflect the diffusion, replacement and retention of carbon dioxide in the pores of the formation, construct a dynamic geological migration model, and analyze the chemical reactions and adsorption / desorption processes between carbon dioxide and various minerals through data, thereby improving the adaptability and prediction ability to complex geological environments, and solving the problems of low carbon dioxide storage monitoring accuracy, delayed response, serious data islands, and insufficient security in the existing technology.

[0062] Data fusion algorithms are used to integrate multi-dimensional monitoring data to form comprehensive monitoring information in a unified format. Enhanced machine learning models are used to continuously learn and infer the storage process, dynamically estimate the total amount of carbon dioxide stored and its spatial distribution, improve assessment accuracy and response speed, and set up a graded early warning mechanism based on the real-time storage status. Potential leakage risks are proactively identified before anomalies occur, and auxiliary verification processes are automatically triggered. Combined with multi-source data cross-verification, the reliability and accuracy of the early warning results are ensured. The entire monitoring method has the characteristics of high precision, strong intelligence, wide coverage, and good safety, and can effectively support dynamic perception, scientific decision-making, and risk prevention and control of the carbon dioxide storage process. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0064] Figure 1 This is a flow chart of a method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0066] like Figure 1 As shown, the present invention provides a method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network, comprising the following steps:

[0067] Deploy a multi-level sensor network within the target storage area, comprising surface monitoring nodes, underwater monitoring nodes, and remote transmission nodes, to collect multi-scale and multi-dimensional monitoring data of the CO2 storage process in real time through the sensor network;

[0068] In this embodiment, by deploying a multi-level sensor network consisting of surface, underwater and remote transmission nodes in the target storage area, multi-scale fluid parameters such as temperature, pressure, gas concentration, flow rate, etc. in the storage area can be collected in an all-weather and three-dimensional manner, and multi-source heterogeneous data can be fused and processed, effectively improving the monitoring coverage breadth and spatial resolution.

[0069] Constructing a geological migration dynamic model based on the multi-scale and multi-dimensional fluid monitoring data and the phase field method and the dynamic grid method to obtain the dynamic evolution data of the geological fluid-solid interface during the carbon dioxide storage process;

[0070] In this embodiment, a geological transport dynamic model based on the phase field method and the dynamic grid method is used to construct a dynamic model, which can truly restore the diffusion, migration and retention behavior of carbon dioxide in the formation under complex geological conditions, realize the upgrade of the carbon dioxide storage process from static monitoring to dynamic evolution, and improve the prediction ability of the storage status change trend.

[0071] fusing the dynamically evolving data using a data fusion algorithm to generate comprehensive monitoring information in a unified format, analyzing the comprehensive monitoring information using an enhanced machine learning model to dynamically estimate the current total amount of carbon dioxide stored and its spatial distribution;

[0072] In this embodiment, a data fusion algorithm is used to integrate multi-dimensional monitoring data to form comprehensive monitoring information in a unified format. With the help of an enhanced machine learning model, continuous learning and reasoning are carried out on the storage process to dynamically estimate the total amount of carbon dioxide stored and its spatial distribution, thereby improving assessment accuracy and response speed.

[0073] By evaluating the carbon dioxide storage status in real time, setting up an early warning mechanism and automatically starting the auxiliary verification process, key monitoring data on the carbon dioxide storage status is obtained, and blockchain technology is used to distribute the key monitoring data and early warning records to obtain real-time and traceable carbon dioxide storage monitoring data.

[0074] In this embodiment, key monitoring data and early warning records are stored on the chain to ensure that the data cannot be tampered with, is auditable, and traceable, thereby improving the transparency and credibility of the monitoring system.

[0075] In this embodiment, by constructing a multi-level sensor network, a dynamic geological migration model, intelligent data analysis and a blockchain evidence storage mechanism, problems such as low accuracy, delayed response, serious data silos and insufficient security in existing carbon dioxide storage monitoring technologies are solved.

[0076] The sensor network collects multi-scale and multi-dimensional monitoring data of the fluid during the carbon dioxide storage process in real time, including:

[0077] The multi-scale fluid multi-dimensional monitoring data at least includes pressure data, solute migration data, gas concentration distribution data and geological deformation data of multi-scale fluid flow;

[0078] In this embodiment, the sensor network can synchronously collect various physical and chemical parameters such as pressure, solute migration, gas concentration distribution and geological deformation during the carbon dioxide storage process, covering multi-scale information from microscopic pores to macroscopic formations, realizing all-round and three-dimensional perception of the storage process, and significantly improving the integrity and accuracy of data collection.

[0079] collecting temporal and spatial scale data during the carbon dioxide storage process through the multi-level sensor network;

[0080] In this embodiment, by collecting time scale data and spatial scale data, the relationship between the temporal evolution law and spatial distribution characteristics of the carbon dioxide migration process is established. The use of time series to drive spatial modeling helps to reveal the migration path and storage efficiency of carbon dioxide in different time periods and improve the model's prediction capabilities.

[0081] A spatial distribution map is constructed for the spatial scale data based on the time scale data to obtain geomechanical dimension data of geological pores during the carbon dioxide injection and storage process.

[0082] In this embodiment, by constructing a spatial distribution map, the diffusion range, accumulation area and potential leakage path of carbon dioxide in the formation can be intuitively reflected, providing a visual basis for risk assessment and regulation strategy formulation.

[0083] Constructing a phase field model based on the phase field method for the multi-scale fluid multi-dimensional monitoring data, including:

[0084] According to the spatial distribution map, a multi-component and multi-mineral state evolution mathematical model is constructed for the geomechanical dimension data by simulating the phase field model of gas-liquid two-phase flow.

[0085]

[0086] in, represents the total concentration of each component, represents the concentration gradient, D represents the molecular diffusion coefficient of the solute in the solution, e m represents the reaction rate on the surface of mineral m, θ m represents the fluid-solid interface on the mineral m surface, μ represents the fluid velocity, and t represents time;

[0087] In this embodiment, a multi-component and multi-mineral state evolution mathematical model is used to describe the evolution state of the complex phase interface under non-equilibrium state during the carbon dioxide injection process, and the phase field method is introduced to model the gas-liquid two-phase flow behavior during the carbon dioxide injection process. This can more realistically reflect the diffusion, replacement and retention of CO2 in the formation pores. Compared with the traditional fixed interface model, the phase field method can automatically capture the fluid interface evolution process and avoid the errors caused by manually setting boundary conditions.

[0088] In the phase field model, the phase field variables are controlled by a step function, the phase field variables are used as independent variables and controlled by the Cahn-Hilliard equation, and the phase field data are coupled with the flow field and the solute field. The expression is:

[0089]

[0090] Where w represents the flow velocity, t represents time, and α represents the fluid density. represents the pressure gradient, μ represents the dynamic viscosity of the fluid on the mineral surface, represents the square of the difference in fluid velocity at the fluid-solid interface, and H(Φ) represents the step function;

[0091] In this embodiment, the spatial distribution map obtained in the early stage is used as the initial input, and geomechanical dimension data, such as porosity and stress field changes, are combined to construct a multi-component and multi-mineral state evolution model. The chemical reactions and adsorption / desorption processes between carbon dioxide and various minerals can be simultaneously considered, thereby improving the adaptability and predictive ability to complex geological environments.

[0092] Finite element analysis software is used to solve the multi-component and multi-mineral state evolution mathematical model to obtain multiphase flow-solid field data.

[0093] In this embodiment, the Cahn-Hilliard equation is used to control the phase field variables, which can ensure that the model has good stability and convergence. The step function is used to accurately characterize the changing process of the fluid-solid interface and effectively simulate key physical behaviors such as CO2 penetration and wettability changes on the rock surface.

[0094] Numerical analysis is performed on the multiphase flow-solid field data to obtain the reaction rate of the fluid-solid interface, including:

[0095] Expand the corresponding spatial distribution map of the multiphase flow-solid field data through a Python simulation program to establish a reaction relationship between the mineral surface and the fluid;

[0096] In this embodiment, based on the multiphase flow-solid field data and combined with the Python simulation program to expand the spatial distribution map, it is possible to more realistically restore the chemical reactions occurring in the formation with the mineral surface after CO2 injection, realize the transition modeling from macroscopic flow behavior to microscopic chemical reaction process, and enhance the understanding of the CO2 geological storage mechanism.

[0097] By calculating the energy of particles that can nucleate on a specific surface area within Δt time, the free energy change function Q at the fluid-solid interface is constructed. s , whose expression is:

[0098]

[0099] Where A represents the fluid-solid interface factor, δ represents the fluid-solid interface free energy, lnΩ represents the supersaturation function, K0 represents the initial surface area ratio of the medium, and T represents the fluid-solid interface temperature;

[0100] The free energy change function Q s The volume mesh of the interface between the fluid and the solid is defined, and the reaction rate at the fluid-solid interface is labeled using an exponential probability distribution function.

[0101] In this embodiment, based on the free energy change function, the volume grid of the fluid-solid interface is defined, which helps to accurately solve the finite element numerical method at the microscale and supports the subsequent dynamic simulation and prediction of processes such as mineral dissolution / precipitation and pore structure evolution.

[0102] In this embodiment, the exponential probability distribution function is used to model the fluid-solid interface reaction rate, taking into account the impact of local environmental heterogeneity on the reaction rate. Compared with the traditional uniform rate assumption, this method is more flexible and applicable, and is particularly suitable for complex and changeable formation environments.

[0103] In this embodiment, the energy change of nucleated particles on a specific surface area within time t is calculated to establish a free energy change function, and an exponential probability distribution function is introduced to mark the reaction rate at the fluid-solid interface, thereby achieving quantitative modeling of the key chemical reaction processes in the CO2 storage process.

[0104] Combined with the dynamic grid method, a geological transport dynamic model is constructed to obtain dynamic evolution data of the geological fluid-solid interface during the carbon dioxide storage process, including:

[0105] According to the free energy change function Q at the fluid-solid interface s The spatial distribution map is meshed using a moving mesh method to segment the fluid-solid interface on the spatial distribution map into 300×300 original data blocks;

[0106] Decompose the volume grid into a 31×23 domain decomposition map, associate the original data block of each subdomain with a cell of the decomposition map, and calculate the porosity size in each cell;

[0107] In this embodiment, a dynamic mesh method is used to finely mesh the spatial distribution map based on the free energy change function at the fluid-solid interface. This high-resolution segmentation helps to more accurately capture the interaction and reaction process between geological fluids and solids. By further decomposing the volume grid into domain decomposition diagrams, the analysis of each subdomain can be refined, making the calculated porosity size more accurate, thereby improving the accuracy of the entire geological transport dynamic model.

[0108] In this embodiment, the dynamic mesh technology can automatically adjust the mesh structure according to changes in the actual physical field. This is particularly important for simulating complex and changeable underground environments. When carbon dioxide injection causes changes in the pressure, temperature, and chemical composition inside the rock, the dynamic mesh can respond to these changes in real time, ensuring that the model always maintains a high level of accuracy. The domain decomposition method also allows different mesh densities to be used in different areas, providing more detailed analysis for critical areas that require higher resolution, such as near the fluid-solid interface, while reducing the consumption of computing resources.

[0109] The geological transport dynamic model state of the geological fluid-solid interface is derived according to the porosity size to obtain real-time dynamic evolution data.

[0110] In this embodiment, the geological migration dynamic model of the geological fluid-solid interface is derived based on the porosity size, which can more accurately obtain the migration pattern of carbon dioxide in the geological layer and its impact on the surrounding rock structure. By obtaining dynamic evolution data in real time, it is possible to monitor problems that may arise during the carbon dioxide storage process, such as the development of leakage paths or changes in reservoir stability, so that timely measures can be taken to prevent potential risks.

[0111] A geological evolution scale analysis model based on the carbon dioxide injection volume is established for the dynamic evolution data, including:

[0112] Using the time scale data as the time axis, the pore pressure within each porosity size is analyzed, and the scale of geological evolution is determined by the range of pore pressure disturbance, wherein the pore pressure disturbance includes the accumulation and diffusion of the corresponding pore pressure;

[0113] In this example, by analyzing the pore pressure within each porosity size and its accumulation and diffusion, the scale of geological evolution caused by CO2 injection can be accurately determined. This method not only considers local pressure changes, but also focuses on its propagation effects in time and space.

[0114] During the CO2 injection process, the influence of the pore pressure and the displacement of the geological fluid-solid interface fault is evaluated;

[0115] In this embodiment, the relationship between pore pressure and the displacement of geological fluid-solid interface faults can be combined to more comprehensively evaluate the impact of carbon dioxide injection on geological structural stability and identify potential geological disaster risk points.

[0116] A nonlinear regression equation is established between the CO2 injection parameters and the factors affecting the geological evolution scale, and its expression is:

[0117] M max =a log(X)+b

[0118] Where a and b represent the factors affecting the scale of geological evolution, and X represents the goodness of fit of the nonlinear regression equation. The greater the goodness of fit, the greater the influence of the corresponding factor on the scale of geological evolution.

[0119] The impact threshold of the geological evolution scale is set through the nonlinear regression equation, and a geological evolution scale analysis model is constructed to obtain geological evolution status data.

[0120] In this embodiment, by establishing a nonlinear regression equation between carbon dioxide injection parameters, such as injection rate and total amount, and factors affecting the scale of geological evolution, the complex nonlinear relationship between the two can be captured, providing more accurate prediction capabilities than traditional linear models; using a goodness of fit index to measure the importance of different influencing factors helps to identify which factors have a decisive effect on the scale of geological evolution, thereby providing a basis for optimizing the injection strategy.

[0121] In this embodiment, setting the impact threshold of the geological evolution scale according to the result of the nonlinear regression equation can help determine under what conditions the geological evolution will exceed the safety range, triggering the early warning mechanism, and taking timely measures to prevent the occurrence of geological disasters.

[0122] The geological evolution state data are fused and processed using a data fusion algorithm to generate comprehensive monitoring information in a unified format, including:

[0123] Cleaning and denoising the geological evolution state data, converting the data into a unified format using a Z-score normalization method, and obtaining geological evolution state data in a unified format;

[0124] The principal component analysis method is used to extract key characteristic variables from the geological evolution state data in the unified format, and a spatiotemporal matching matrix is ​​constructed by combining the time scale data and the spatial scale data. The key characteristic variables are aligned in a unified spatiotemporal coordinate system to form a structured data set with temporal correlation.

[0125] In this embodiment, a spatiotemporal matching matrix is ​​constructed by combining time scale and spatial scale data to ensure that the data of different sensor nodes are accurately aligned in a unified spatiotemporal coordinate system, thereby solving the information misalignment problem caused by inconsistent data acquisition frequency and uneven spatial distribution in traditional methods, and improving the accuracy and interpretability of data fusion results.

[0126] Performing fusion modeling on the structured data set to output comprehensive monitoring indicators reflecting the overall status of the storage area;

[0127] According to the historical data performance and real-time feedback mechanism, the weight coefficient of each sensor node or data dimension in the fusion process is dynamically adjusted to obtain comprehensive monitoring information with dynamic weight distribution.

[0128] In this embodiment, principal component analysis is used to extract key characteristic variables, which can reduce data dimensions while retaining the main information, reduce the interference of redundant information on model calculations, improve modeling efficiency and stability, and help discover key evolution laws hidden in multidimensional data.

[0129] An enhanced machine learning model is used to analyze the comprehensive monitoring information to dynamically estimate the current total amount of CO2 stored and its spatial distribution, including:

[0130] Long short-term memory (LSTM) network was used as an enhanced machine learning model to train the historical dataset, and the characteristic parameters of the LSTM network were adjusted through cross-validation.

[0131] In this embodiment, a long short-term memory network (LSTM) is used to effectively capture the temporal evolution patterns of the CO2 storage process. Compared with traditional statistical methods or static models, LSTM has stronger time series modeling capabilities and can achieve high-precision dynamic estimation of the total storage volume.

[0132] Using the trained model to conduct real-time analysis of the comprehensive monitoring information, dynamically estimate the current carbon dioxide storage capacity and predict future trends;

[0133] In this embodiment, real-time analysis of the comprehensive monitoring information can not only evaluate the current storage status, but also predict the trend of changes in the total storage volume in the future, which helps to identify potential risks in advance, optimize injection strategies, and improve the controllability and safety of the storage project.

[0134] A geological model was established using 3D visualization tools combined with GIS to simulate the spatial distribution of the total amount of carbon dioxide storage.

[0135] In this embodiment, a geological model is established using three-dimensional visualization tools and a geographic information system to visually display the spatial distribution of the total amount of carbon dioxide storage in a graphical manner. It supports viewing the storage status by multiple dimensions such as region, depth, and time, allowing managers to quickly grasp the overall storage pattern and local abnormal areas.

[0136] By assessing the CO2 storage status in real time, setting up early warning mechanisms and automatically initiating auxiliary verification processes, key monitoring data on the CO2 storage status can be obtained, including:

[0137] Using the geological evolution data, the carbon dioxide storage status is evaluated in real time, risk indicators are calculated, and warning thresholds and rules for each risk indicator are set;

[0138] In this embodiment, the sealing process is continuously evaluated based on geological evolution status data, which can dynamically capture potential anomalies and improve the system's response capability to emergencies. Compared with traditional periodic inspections or static evaluation methods, it has higher timeliness and accuracy.

[0139] When risk indicators exceed the threshold, the corresponding level of early warning mechanism is automatically triggered and notifications are sent to relevant personnel, and auxiliary verification processes are automatically initiated, including increasing monitoring frequency, calling additional sensor data, or dispatching on-site technicians;

[0140] In this embodiment, when the risk indicator exceeds the limit, the system automatically starts the auxiliary verification process, including increasing the monitoring frequency, retrieving additional sensor data, or scheduling manual verification, realizing the transition from "passive discovery" to "active response", significantly shortening the risk handling time, and ensuring the safety of the sealing system.

[0141] Collect and analyze key monitoring data to confirm the actual status of the storage system, feed back auxiliary verification results to the monitoring system to update risk assessment model parameters, and record early warning events to form a knowledge base.

[0142] In this embodiment, the collected key monitoring data is fed back to the monitoring system to correct and update the risk assessment model parameters, forming a "monitoring-assessment-response-feedback" closed-loop mechanism, continuously improving the model prediction accuracy and the system's intelligent decision-making capabilities.

[0143] Blockchain technology is used to distribute and store the key monitoring data and early warning records, obtaining real-time and traceable CO2 storage monitoring data, including:

[0144] Standardize the key monitoring data and early warning records, build a blockchain network consisting of multiple nodes, package the pre-processed key monitoring data into blocks, add timestamps and digital signatures, and then upload them to the blockchain;

[0145] Use smart contracts to manage data access rights and achieve real-time data synchronization in the blockchain network;

[0146] Through API-based query interfaces and visualization tools, a real-time retrieval and analysis model for historical data is established, providing an intuitive visualization interface.

[0147] In this embodiment, a blockchain network consisting of multiple nodes is constructed to avoid data loss problems caused by single point failures in traditional centralized databases. Data is distributed and backed up in multiple nodes. Even if some nodes fail, the continuous operation of the system and data availability can still be guaranteed.

[0148] In this embodiment, based on the API interface and visualization tools, users can quickly retrieve historical data, analyze trend changes, and locate abnormal events, providing a graphical display interface to facilitate managers to grasp the sealing status in real time and assist in scientific decision-making and emergency response.

[0149] The present invention constructs a multi-level sensor network, collects time-scale data and space-scale data, establishes the correlation between the time evolution law of the carbon dioxide migration process and the spatial distribution characteristics, and uses time series to drive spatial modeling, which helps to reveal the migration path and storage efficiency of carbon dioxide in different time periods and improve the model prediction ability. The phase field method is introduced to model the gas-liquid two-phase flow behavior during the carbon dioxide injection process, which can more realistically reflect the diffusion, replacement and retention of carbon dioxide in the pores of the formation, construct a dynamic geological migration model, and analyze the chemical reactions and adsorption / desorption processes between carbon dioxide and various minerals through intelligent data, thereby improving the adaptability and prediction ability to complex geological environments, and solving the problems of low carbon dioxide storage monitoring accuracy, delayed response, serious data islands, and insufficient security in the existing technology.

[0150] Data fusion algorithms are used to integrate multi-dimensional monitoring data to form comprehensive monitoring information in a unified format. Enhanced machine learning models are used to continuously learn and infer the storage process, dynamically estimate the total amount of carbon dioxide stored and its spatial distribution, improve assessment accuracy and response speed, and set up a graded early warning mechanism based on the real-time storage status. Potential leakage risks are proactively identified before anomalies occur, and auxiliary verification processes are automatically triggered. Combined with multi-source data cross-verification, the reliability and accuracy of the early warning results are ensured. The entire monitoring method has the characteristics of high precision, strong intelligence, wide coverage, and good safety, and can effectively support dynamic perception, scientific decision-making, and risk prevention and control of the carbon dioxide storage process.

[0151] The above embodiments are merely exemplary embodiments of the present application and are not intended to limit the scope of the present application. The scope of protection of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and scope of protection of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present application.

Claims

1. A method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network, characterized in that: The following steps are involved: Deploy a multi-level sensor network within the target storage area, comprising surface monitoring nodes, underwater monitoring nodes, and remote transmission nodes, to collect multi-scale and multi-dimensional monitoring data of the CO2 storage process in real time through the sensor network; Constructing a geological migration dynamic model based on the multi-scale and multi-dimensional fluid monitoring data and the phase field method and the dynamic grid method to obtain the dynamic evolution data of the geological fluid-solid interface during the carbon dioxide storage process; fusing the dynamically evolving data using a data fusion algorithm to generate comprehensive monitoring information in a unified format, analyzing the comprehensive monitoring information using an enhanced machine learning model to dynamically estimate the current total amount of carbon dioxide stored and its spatial distribution; By evaluating the carbon dioxide storage status in real time, setting up an early warning mechanism and automatically starting the auxiliary verification process, key monitoring data on the carbon dioxide storage status is obtained, and blockchain technology is used to distribute the key monitoring data and early warning records to obtain real-time and traceable carbon dioxide storage monitoring data.

2. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 1, characterized in that: The sensor network collects multi-scale and multi-dimensional monitoring data of the fluid during the carbon dioxide storage process in real time, including: The multi-scale fluid multi-dimensional monitoring data at least includes pressure data, solute migration data, gas concentration distribution data and geological deformation data of multi-scale fluid flow; collecting temporal and spatial scale data during the carbon dioxide storage process through the multi-level sensor network; A spatial distribution map is constructed for the spatial scale data based on the time scale data to obtain geomechanical dimension data of geological pores during the carbon dioxide injection and storage process.

3. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 2, characterized in that: Constructing a phase field model based on the phase field method for the multi-scale fluid multi-dimensional monitoring data, including: According to the spatial distribution map, a multi-component and multi-mineral state evolution mathematical model is constructed for the geomechanical dimension data by simulating the phase field model of gas-liquid two-phase flow. in, represents the total concentration of each component, represents the concentration gradient, D represents the molecular diffusion coefficient of the solute in the solution, e m represents the reaction rate on the surface of mineral m, θ m represents the fluid-solid interface on the mineral m surface, μ represents the fluid velocity, and t represents time; In the phase field model, the phase field variables are controlled by a step function, the phase field variables are used as independent variables and controlled by the Cahn-Hilliard equation, and the phase field data are coupled with the flow field and the solute field. The expression is: Among them, w represents the flow field velocity, t represents time, α represents the fluid density, represents the pressure gradient, μ represents the dynamic viscosity of the fluid on the mineral surface, represents the square of the difference in fluid velocity at the fluid-solid interface, and H(Φ) represents the step function; Finite element analysis software is used to solve the multi-component and multi-mineral state evolution mathematical model to obtain multiphase flow-solid field data.

4. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 3, characterized in that: Numerical analysis is performed on the multiphase flow-solid field data to obtain the reaction rate of the fluid-solid interface, including: Expand the corresponding spatial distribution map of the multiphase flow-solid field data through a Python simulation program to establish a reaction relationship between the mineral surface and the fluid; By calculating the energy of particles that can nucleate on a specific surface area within Δt time, the free energy change function Q at the fluid-solid interface is constructed. s , whose expression is: Where A represents the fluid-solid interface factor, δ represents the fluid-solid interface free energy, lnΩ represents the supersaturation function, K0 represents the initial surface area ratio of the medium, and T represents the fluid-solid interface temperature; The free energy change function Q s The volume mesh of the interface between the fluid and the solid is defined, and the reaction rate at the fluid-solid interface is labeled using an exponential probability distribution function.

5. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 4, characterized in that: Combined with the dynamic grid method, a geological transport dynamic model is constructed to obtain dynamic evolution data of the geological fluid-solid interface during the carbon dioxide storage process, including: According to the free energy change function Q at the fluid-solid interface s The spatial distribution map is meshed using a moving mesh method to segment the fluid-solid interface on the spatial distribution map into 300×300 original data blocks; Decompose the volume grid into a 31×23 domain decomposition map, associate the original data block of each subdomain with a cell of the decomposition map, and calculate the porosity size in each cell; The geological transport dynamic model state of the geological fluid-solid interface is derived according to the porosity size to obtain real-time dynamic evolution data.

6. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 5, characterized in that: A geological evolution scale analysis model based on the carbon dioxide injection volume is established for the dynamic evolution data, including: Using the time scale data as the time axis, the pore pressure within each porosity size is analyzed, and the scale of geological evolution is determined by the range of pore pressure disturbance, wherein the pore pressure disturbance includes the accumulation and diffusion of the corresponding pore pressure; During the CO2 injection process, the influence of the pore pressure and the displacement of the geological fluid-solid interface fault is evaluated; A nonlinear regression equation is established between the CO2 injection parameters and the factors affecting the geological evolution scale, and its expression is: M max =alog(X)+b Where a and b represent the factors affecting the scale of geological evolution, and X represents the goodness of fit of the nonlinear regression equation. The greater the goodness of fit, the greater the influence of the corresponding factor on the scale of geological evolution. The impact threshold of the geological evolution scale is set through nonlinear regression equations, a geological evolution scale analysis model is constructed, and geological evolution status data is obtained.

7. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 6, characterized in that: The geological evolution state data are fused and processed using a data fusion algorithm to generate comprehensive monitoring information in a unified format, including: Cleaning and denoising the geological evolution state data, converting the data into a unified format using a Z-score normalization method, and obtaining geological evolution state data in a unified format; The principal component analysis method is used to extract key characteristic variables from the geological evolution state data in the unified format, and a spatiotemporal matching matrix is ​​constructed by combining the time scale data and the spatial scale data. The key characteristic variables are aligned in a unified spatiotemporal coordinate system to form a structured data set with temporal correlation. Performing fusion modeling on the structured data set to output comprehensive monitoring indicators reflecting the overall status of the storage area; According to the historical data performance and real-time feedback mechanism, the weight coefficient of each sensor node or data dimension in the fusion process is dynamically adjusted to obtain comprehensive monitoring information with dynamic weight distribution.

8. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 7, characterized in that: An enhanced machine learning model is used to analyze the comprehensive monitoring information to dynamically estimate the current total amount of CO2 stored and its spatial distribution, including: Long short-term memory (LSTM) network was used as an enhanced machine learning model to train the historical dataset, and the characteristic parameters of the LSTM network were adjusted through cross-validation. Using the trained model to conduct real-time analysis of the comprehensive monitoring information, dynamically estimate the current carbon dioxide storage capacity and predict future trends; A geological model was established using 3D visualization tools combined with GIS to simulate the spatial distribution of the total amount of carbon dioxide storage.

9. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 8, characterized in that: By assessing the CO2 storage status in real time, setting up early warning mechanisms and automatically initiating auxiliary verification processes, key monitoring data on the CO2 storage status can be obtained, including: Using the geological evolution data, the carbon dioxide storage status is evaluated in real time, risk indicators are calculated, and warning thresholds and rules for each risk indicator are set; When risk indicators exceed the threshold, the corresponding level of early warning mechanism is automatically triggered and notifications are sent to relevant personnel, and auxiliary verification processes are automatically initiated, including increasing monitoring frequency, calling additional sensor data, or dispatching on-site technicians; Collect and analyze key monitoring data to confirm the actual status of the storage system, feed back auxiliary verification results to the monitoring system to update risk assessment model parameters, and record early warning events to form a knowledge base.

10. The method for real-time monitoring of carbon dioxide storage based on an enhanced monitoring network according to claim 9, characterized in that: Blockchain technology is used to distribute and store the key monitoring data and early warning records, obtaining real-time and traceable CO2 storage monitoring data, including: Standardize the key monitoring data and early warning records, build a blockchain network consisting of multiple nodes, package the pre-processed key monitoring data into blocks, add timestamps and digital signatures, and then upload them to the blockchain; Use smart contracts to manage data access rights and achieve real-time data synchronization in the blockchain network; Through API-based query interfaces and visualization tools, a real-time retrieval and analysis model for historical data is established, providing an intuitive visualization interface.

Images