Gas permeability monitoring method and system based on carbon dioxide geological sequestration

Abstract

The invention discloses a gas permeability monitoring method and system based on carbon dioxide geological sequestration, and relates to the technical field of gas permeability monitoring, and the method comprises the following steps: obtaining geological data and a first constraint condition of a target stratum; disturbing the seismic wave frequency to generate a plurality of seismic monitoring control sets; constructing a three-dimensional geological model and a wave velocity change simulation model of the sequestration area, and performing forward simulation to obtain wave velocity change response data under different frequency combinations; wave velocity disturbance characteristic distribution caused by carbon dioxide injection under different earthquake monitoring control sets is obtained, and the set evaluation model is weighted and trained; constructing a second set according to an evaluation result of the set evaluation model and a corresponding earthquake monitoring control set; and fitting a multi-dimensional curved surface according to the second set, and screening an optimal seismic monitoring control set for seismic data acquisition and inversion, thereby solving the problem that traditional single-frequency-band inversion cannot dynamically match a stratum structure and permeability characteristics.

Description

Technical Field

[0001] The present invention relates to the technical field of gas permeability monitoring. More specifically, the present invention relates to a gas permeability monitoring method and system based on carbon dioxide geological sequestration. Background Art

[0002] Carbon dioxide geological sequestration refers to the process of injecting captured carbon dioxide into deep underground formations (such as depleted oil and gas reservoirs, deep salt caverns, aquifers) for long-term storage to prevent it from entering the atmosphere and mitigate the greenhouse gas effect. This is an important carbon capture and storage (CCS) technology.

[0003] Gas permeability monitoring refers to using sensors, data analysis, modeling algorithms, or other technical means to obtain and analyze the gas permeability of underground formations in real-time or periodically, monitor the sequestration effect, and detect potential leaks.

[0004] During the process of carbon dioxide geological sequestration, seismic wave velocity change inversion is a key physical monitoring method, mainly used to evaluate the physical property changes of the formation after injecting carbon dioxide, especially to identify the carbon dioxide migration path, the diffusion range of the injection layer, and the sequestration safety. Its core principle is that: the propagation speed of seismic waves in the formation is affected by factors such as lithology, porosity, fluid type, and saturation. When carbon dioxide injection causes the replacement of the original fluid in the formation or the change of pore pressure, it will lead to a change in the seismic wave velocity. By analyzing the differences in seismic data at different time points, the small changes in wave velocity can be inverted, and then the distribution and diffusion dynamics of carbon dioxide can be inferred.

[0005] Specifically, this method is usually implemented through time-lapse seismic data, including: collecting baseline seismic data before injection and seismic data at multiple moments after injection, extracting travel time changes or waveform differences after consistency processing; and then reconstructing the formation wave velocity change model through an inversion algorithm. This process highly depends on the signal-to-noise ratio, the consistency of acquisition parameters, and the frequency control strategy, among which frequency control is particularly crucial - low frequency helps to detect deep regions and large-scale changes, while high frequency is beneficial for resolving details and abnormal boundaries. The coordinated use of the two improves the resolution and stability of the inversion.

[0006] Through seismic wave velocity change inversion, not only can the carbon dioxide migration process be dynamically tracked, but also it can assist in evaluating the integrity of the sequestration layer, the effectiveness of the caprock, and the potential leakage risk. Therefore, it has an irreplaceable position in the long-term monitoring of CCS projects.

[0007] For example, the prediction method of formation characteristic parameters based on carbon dioxide geological storage disclosed in the invention patent announcement with the announcement number: CN119067255A includes that the present application discloses a prediction method of formation characteristic parameters based on carbon dioxide geological storage, belonging to the technical field of oil and gas exploration and development. The method includes: obtaining the first geological parameter and the second geological parameter of the target formation; based on the first target prediction model for predicting the first predicted formation characteristic parameter, obtaining the first predicted formation characteristic parameter of the target formation according to the first geological parameter, wherein the first predicted formation characteristic parameter includes the predicted value of formation pressure, the predicted value of carbon dioxide saturation and the predicted value of formation fluid density; based on the second target prediction model for predicting the second predicted formation characteristic parameter, obtaining the second predicted formation characteristic parameter of the target formation according to the first predicted formation characteristic parameter and the second geological parameter, wherein the second predicted formation characteristic parameter includes the predicted value of mineral volume fraction and the predicted value of formation permeability. The present application can improve the prediction efficiency of the formation characteristic parameters during the carbon dioxide geological storage process.

[0008] For example, a carbon dioxide geological storage seal capacity prediction method and device disclosed in the invention patent announcement with the announcement number: CN118798425A includes that the present invention discloses a carbon dioxide geological storage seal capacity prediction method and device. The method includes the following steps: preprocessing the collected carbon dioxide seal capacity data to obtain characteristic data; screening characteristic indicators from a preset index library through feature screening; wherein, the preset index library includes several seal indicators that affect carbon dioxide geological storage; training a preset model according to the characteristic data and the several characteristic indicators to obtain a seal capacity prediction model; predicting the seal capacity of carbon dioxide geological storage in the area to be measured through the seal capacity prediction model. The present invention can improve the efficiency and accuracy of predicting the seal capacity of carbon dioxide geological storage.

[0009] In the above disclosed technical solutions, there are at least the following technical problems: In the inversion process of seismic wave velocity change in carbon dioxide geological storage, the selection of seismic wave frequency plays a key role in monitoring accuracy and penetration depth. If the frequency is set too high, although the inversion resolution can be improved and it is easier to identify small-scale permeability changes, at the same time, the signal attenuation is serious, the penetration ability decreases, and it is easily affected by noise, resulting in unstable inversion results of deep targets; if the frequency is too low, although it has good penetration and the ability to construct the overall velocity field, the spatial resolution is insufficient, it is difficult to capture local wave velocity anomalies, and it is easy to miss small leakage channels or early infiltration signs.

[0010] In the prior art, single frequency bands are mostly used for inversion, lacking a dynamic matching mechanism between the deep and shallow changes of the formation structure, the permeability scale characteristics and the signal characteristics, which easily causes deviation or information loss in the inversion results, affecting the timeliness and accuracy of the seal safety assessment and risk warning.

[0011] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0012] To overcome the above defects of the prior art, an embodiment of the present invention provides a system for a gas permeability monitoring method based on carbon dioxide geological sequestration. By combining multi-frequency perturbations with various seismic source and acquisition methods, a diversified seismic monitoring scheme is constructed. Using three-dimensional geology and wave velocity simulation forward modeling, combined with machine learning to perform weighted evaluation on the scheme, and finally the optimal frequency is selected. It solves the problem that traditional single-frequency inversion cannot dynamically match the formation structure and permeability characteristics, and greatly improves the monitoring accuracy and timeliness of sequestration safety assessment.

[0013] To achieve the above object, the present invention provides the following technical solutions: A gas permeability monitoring method based on carbon dioxide geological sequestration includes the following steps: obtaining geological data of a target formation and determining a first constraint condition required for inversion; based on the first constraint condition, perturbing the seismic wave frequency and randomly combining different seismic source types and acquisition methods to generate a number of seismic monitoring control sets; constructing a three-dimensional geological model of the sequestration area and a wave velocity change simulation model, and performing forward modeling based on each seismic monitoring control set to obtain wave velocity change response data under different frequency combinations; obtaining the distribution of wave velocity perturbation characteristics caused by carbon dioxide injection under different seismic monitoring control sets, and combining the wave velocity change response data to weighted train a set evaluation model; constructing a second set according to the evaluation results of the set evaluation model and the corresponding seismic monitoring control sets; fitting a multi-dimensional surface according to the second set, and screening the optimal seismic monitoring control set for seismic data acquisition and inversion.

[0014] In a preferred embodiment, the obtaining geological data of the target formation and determining the first constraint condition required for inversion are specifically: based on core test results and empirical models, obtaining lithology types and their corresponding pore permeability characteristics; converting the lithology types and their corresponding pore permeability characteristics into sensitivity factors of wave velocity perturbation based on core test fitting method to form a lithology-perturbation response matrix, quantitatively expressing the degree of wave velocity change caused by the permeability of different lithologies to carbon dioxide; setting the first constraint condition for inversion according to the lithology-perturbation response matrix in combination with the pressure gradient, injection rate and formation noise level in the obtained injection history and predicted working conditions.

[0015] In a preferred embodiment, based on the first constraint condition, the seismic wave frequency is perturbed, and combined with different source types and acquisition methods randomly combined to generate a number of seismic monitoring control sets, specifically: calculate the first initial frequency based on geological data combined with empirical formulas; obtain the optimal frequency bandwidth range by combining formation scattering characteristics and signal-to-noise ratio analysis, and the formation scattering characteristics and signal-to-noise ratio analysis include satisfying the maximum signal-to-noise ratio and covering the preset geological interface response frequency band; adjust the first initial frequency to capture the dynamic wave velocity change caused by infiltration, and use the frequency point matching the time constant of the infiltration process as the second initial frequency; randomly perturb the second initial frequency under the condition of satisfying the first constraint condition to generate a number of frequencies to be screened; randomly combine the number of frequencies to be screened with different source types and acquisition methods to generate a number of seismic monitoring control sets.

[0016] In a preferred embodiment, the construction of the three-dimensional geological model of the storage area and the wave velocity change simulation model is specifically as follows: based on the drilling, logging and seismic data of the target formation, construct a three-dimensional geological model of the storage area; grid the geological model; map the pore pressure change and saturation change caused by carbon dioxide injection into the physical response of wave velocity change; based on the finite difference method, perform forward simulation on the wave velocity change to obtain the wave field response under different combinations of seismic wave frequencies, source types and acquisition parameters, and obtain the spatial distribution characteristics of wave velocity perturbation; through the analysis of the simulation results, extract the corresponding relationship between the wave velocity change and the carbon dioxide infiltration characteristics as the model output result.

[0017] In a preferred embodiment, based on each seismic monitoring control set, forward simulation is performed to obtain the wave velocity change response data under different frequency combinations, specifically: perform forward simulation on each seismic monitoring control set based on the geological model and the wave velocity change simulation model, extract the characteristics of the simulation results and associate them with the formation position and time to obtain the wave velocity change response data.

[0018] In a preferred embodiment, to obtain the distribution of wave velocity perturbation characteristics caused by carbon dioxide injection under different seismic monitoring control sets, specifically: in the forward simulation results corresponding to each control set, identify the time series wave velocity perturbation images in the carbon dioxide injection area and its adjacent areas, and extract the perturbation characteristic distribution; the perturbation characteristic distribution includes the perturbation morphology boundary of the injection influence area, forming a spatial mask of the perturbation area; the volume change rate, perturbation boundary propagation speed and direction of the perturbation area at different times; the wave velocity gradient distribution and uniformity index inside the perturbation area; the matching degree between the main propagation path representing the perturbation front and the preset infiltration direction; construct the perturbation characteristic distribution into a perturbation characteristic vector.

[0019] In a preferred embodiment, a collective evaluation model is trained by weighting wave velocity change response data, specifically as follows: A gradient boosting tree algorithm is used to construct the collective evaluation model; the model takes a combined feature composed of a perturbation feature vector and the corresponding wave velocity change response data as input to construct training samples; the training samples include the perturbation feature vector and the wave velocity change response data; according to the inversion requirements of the target formation, a collective evaluation index is set as the label value for supervised learning; the label value includes an effective perturbation identification ability feature evaluated based on the spatial matching degree between the perturbation distribution and the target storage layer; a time response accuracy feature evaluated based on the synchronization between the first appearance time of the perturbation feature and injection; a wave velocity perturbation structure clarity feature evaluated based on the interpretability of the perturbation profile in a key area; based on the input and the label of supervised learning, regression training is performed using the gradient boosting tree model, and the collective evaluation model is output.

[0020] In a preferred embodiment, a second set is constructed according to the evaluation result of the collective evaluation model and the corresponding seismic monitoring control set, specifically as follows: Each second set includes a seismic monitoring control set and the output value of a corresponding collective evaluation model.

[0021] In a preferred embodiment, a multi-dimensional surface is fitted according to the second set, and the optimal seismic monitoring control set is selected for seismic data acquisition and inversion, specifically as follows: The second set is mapped to a multi-dimensional coordinate system, and the points mapped by several second sets are fitted into a surface by the curve fitting method; the surface is differentiated, and the seismic monitoring control set corresponding to the most stable peak is analyzed and obtained for seismic data acquisition and inversion.

[0022] A system for a gas permeability monitoring method based on carbon dioxide geological storage includes a constraint module, a control set module, a geology module, a collective evaluation module, and a screening module; the constraint module is used to obtain geological data of the target formation and determine the first constraint conditions required for inversion; the control set module is used to perturb the seismic wave frequency based on the first constraint conditions and randomly combine different seismic source types and acquisition methods to generate several seismic monitoring control sets; the geology module is used to construct a three-dimensional geological model of the storage area and a wave velocity change simulation model, and perform forward simulation based on each seismic monitoring control set to obtain wave velocity change response data under different frequency combinations; the collective evaluation module is used to obtain the wave velocity perturbation feature distribution caused by carbon dioxide injection under different seismic monitoring control sets, and train the collective evaluation model by weighting the wave velocity change response data; the screening module is used to construct a second set according to the evaluation result of the collective evaluation model and the corresponding seismic monitoring control set; a multi-dimensional surface is fitted according to the second set, and the optimal seismic monitoring control set is selected for seismic data acquisition and inversion.

[0023] Technical effects and advantages of the gas permeability monitoring method and system based on carbon dioxide geological storage according to the present invention: 1. According to the present invention, by disturbing the seismic wave frequency based on geological constraint conditions and combining random combinations of different seismic source types and acquisition methods, a diversified seismic monitoring control set is generated. This innovation breaks through the limitation of traditional single-frequency band monitoring, effectively covers the depth changes of the formation structure and different permeability scale characteristics, realizes dynamic matching, improves the capture ability of wave velocity disturbance and the accuracy of inversion, thereby enhancing the fineness and reliability of carbon dioxide storage gas permeability monitoring.

[0024] 2. According to the present invention, by constructing a detailed three-dimensional geological model and mapping the pore pressure and saturation changes caused by carbon dioxide injection into the physical response of wave velocity changes, forward simulation of multi-parameter combination is carried out by combining the finite difference method. This method accurately reproduces the spatial distribution characteristics of wave velocity disturbance in the storage area, helps to reveal the internal relationship between wave velocity disturbance and the carbon dioxide penetration process, and enhances the physical authenticity of the simulation results and the basic reliability of the inversion model.

[0025] 3. According to the present invention, by proposing to extract detailed wave velocity disturbance characteristic distributions from the forward simulation results, including multi-dimensional indexes such as the morphology of the disturbance area, the expansion speed, and the wave velocity gradient, and combining the wave velocity change response data, a set evaluation model is constructed by using the gradient boosting tree algorithm. This innovation realizes the intelligent weighted evaluation of the seismic monitoring control set, significantly improves the timeliness and accuracy of the inversion results, reduces the risk of subjective human selection, and realizes the optimization of the monitoring scheme combining automation and high efficiency.

[0026] 4. According to the present invention, by using the output of the set evaluation model to construct a second set, through the multi-dimensional surface fitting method, the discrete evaluation results are mapped into a continuous surface, and then the most stable peak point is found by analyzing the surface derivative, and the optimal seismic monitoring control set is selected. This technical means effectively avoids the local optimal trap, ensures the global optimality of the monitoring scheme in the multi-parameter space, improves the stability and practicability of the acquisition and inversion schemes, and ensures the scientificity and accuracy of the storage safety evaluation. Description of the Drawings

[0027] Figure 1 It is a schematic flow chart of the gas permeability monitoring method based on carbon dioxide geological storage according to the present invention; Figure 2 It is a schematic structural diagram of the gas permeability monitoring system based on carbon dioxide geological storage according to the present invention. Detailed Embodiments

[0028] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0029] Embodiment 1, Figure 1 A gas permeability monitoring method based on carbon dioxide geological sequestration of the present invention is given, including the following steps: S1. Obtain the geological data of the target formation and determine the first constraint condition required for inversion.

[0030] In this embodiment, obtaining the geological data of the target formation and determining the first constraint condition required for inversion are specifically as follows: Based on the core test results and empirical models, obtain the lithology types and their corresponding pore permeability characteristics; Based on the core test fitting method, convert the lithology types and their corresponding pore permeability characteristics into sensitivity factors of wave velocity perturbation, form a lithology-perturbation response matrix, and quantitatively express the degree of wave velocity change caused by the penetration of carbon dioxide by different lithologies; According to the lithology-perturbation response matrix, combined with the injection history and the pressure gradient, injection rate, and formation noise level in the predicted working conditions, set the first constraint condition for inversion.

[0031] It should be noted that core testing is one of the main means to obtain the true physical properties of underground formations. By testing parameters such as permeability, porosity, density, and wave velocity on the collected core samples, the pore permeability characteristic data of different lithology types under experimental conditions can be obtained. Empirical models such as the Kozeny-Carman formula and the Wyllie time-average formula are used to estimate and complement the pore permeability characteristics according to existing empirical laws when the test data is insufficient or there are errors, ensuring the formation of a complete and reliable formation physical property database.

[0032] It should be noted that lithology types usually include sandstone, shale, limestone, siltstone, etc. For each lithology, due to differences in mineral composition, particle structure, and cementation degree, there are significant differences in its porosity and permeability. Porosity reflects the size of the rock's storage space, while permeability affects the flow ability in the pores. The two jointly determine the response intensity of the rock to the pore pressure and saturation changes caused by injection, which are the basic physical properties affecting wave velocity perturbation.

[0033] The core test fitting method refers to jointly fitting actual core test data with seismic interpretation data, logging curves, etc., to construct a mapping relationship among lithology-porosity / permeability-wave velocity perturbation. This method usually uses multiple regression or machine learning means to identify the correlation between lithology characteristics (such as shale content, particle size distribution) and the corresponding wave velocity changes, so as to establish a parameter sensitivity model for seismic wave simulation.

[0034] The wave velocity perturbation sensitivity factor is used to measure the wave velocity response intensity of a certain lithology after injection. Its calculation is based on core experiment data, laboratory injection test results and numerical simulation results. Its value is expressed as the wave velocity change rate caused by unit pressure change or saturation change, and is used to accurately model the elastic wave response under injection influence.

[0035] The lithology-perturbation response matrix is a two-dimensional array. The rows represent different lithologies, and the columns represent perturbation conditions (such as pressure increment, saturation change). The matrix elements are the wave velocity perturbation sensitivity factors under the corresponding conditions. This matrix, as the core input for forward simulation and inversion constraints, can be used to distinguish the response differences of different lithologies to the infiltration process and improve the simulation accuracy.

[0036] The injection history includes the injection start time, cumulative injection volume, injection rate change curve, well location layout, etc., which is an important basis for establishing the formation evolution background. The prediction conditions are based on the geological storage plan to simulate and predict the injection conditions (such as rate, pressure) in the future for a period of time, so as to evaluate the long-term response characteristics of the storage area.

[0037] The first constraint condition combines the lithology response matrix, injection parameters and noise level to form the control boundary conditions for subsequent forward and inverse simulations. It limits the reasonable frequency perturbation range, sensitive formation area, expected wave velocity response range, etc., and provides a technical basis for designing an efficient seismic monitoring control set.

[0038] The construction of the above first constraint condition realizes the physical response mapping between geological attributes and wave velocity perturbation by constructing the lithology-perturbation response matrix and setting the first constraint condition for inversion. Thus, without relying on a large amount of on-site monitoring data, it can accurately evaluate the sensitivity of different lithologies to the wave velocity changes caused by the carbon dioxide infiltration process, effectively improving the pertinence and scientificity of subsequent frequency perturbation simulation and seismic monitoring parameter design, and further enhancing the identification ability of the inversion model for the infiltration characteristics in complex storage formations, with significant prediction accuracy and engineering practical value.

[0039] S2. Based on the first constraint condition, perturb the seismic wave frequency and randomly combine different source types and acquisition methods to generate several seismic monitoring control sets.

[0040] In this embodiment, based on the first constraint condition, the seismic wave frequency is perturbed, and combined with different source types and acquisition methods randomly combined to generate a number of seismic monitoring control sets, specifically: Calculate the first initial frequency based on geological data combined with empirical formulas; Obtain the optimal frequency bandwidth range by combining formation scattering characteristics and signal-to-noise ratio analysis, and the formation scattering characteristics and signal-to-noise ratio analysis include satisfying the maximum signal-to-noise ratio and covering the preset geological interface response frequency band; Adjust the first initial frequency to capture the dynamic wave velocity change caused by infiltration, and use the frequency point matching the time constant of the infiltration process as the second initial frequency; Randomly perturb the second initial frequency under the condition of satisfying the first constraint condition to generate a number of frequencies to be screened; Randomly combine a number of frequencies to be screened with different source types and acquisition methods to generate a number of seismic monitoring control sets.

[0041] It should be noted that geological data includes logging data, lithologic profiles, interlayer thickness, wave velocity curves, etc., which are used to reflect the physical structure and medium characteristics of the formation. By combining existing empirical formulas (such as the frequency-wave velocity relationship derived based on the formation shear modulus and density), the initial frequency that can effectively penetrate the target formation and generate significant reflection signals, that is, the first initial frequency, can be preliminarily estimated, providing a reference value for subsequent frequency perturbation.

[0042] The formation scattering characteristics refer to the phenomenon of seismic wave propagation energy dispersion caused by geological interfaces, inhomogeneous bodies or fractures, etc. High-frequency signals are more affected by scattering and absorption, and low-frequency signals have insufficient resolution. Signal-to-noise ratio analysis combines the background noise, instrument response and geological interface reflection characteristics in the actual exploration environment to screen out the frequency bandwidth interval with the maximum signal-to-noise ratio on the premise of maintaining sufficient resolution, so as to determine the subsequent perturbation range.

[0043] The infiltration behavior in the sealed formation usually has a certain dynamic time scale, which is reflected in the seismic response as the dynamic change of a specific frequency band. By introducing a time constant (such as the Darcy seepage or capillary front diffusion rate), the frequency point most sensitively matched to this dynamic process is determined as the second initial frequency to more accurately capture the wave velocity perturbation effect caused by the infiltration behavior.

[0044] Based on the second initial frequency, perturbations are introduced, and mathematical models such as normal distribution or uniform distribution are used to generate a number of frequencies to be screened, simulating the influence of different settings in actual seismic monitoring on the formation response, and enhancing the system's identification ability for the sensitive interval of frequency change.

[0045] The source types, such as explosive sources, hammer sources, or Vibroseis sources, and the acquisition methods, including multi-component detection, linear arrays, planar arrays, etc., are parameterized and cross-combined with the frequencies to be screened to form multiple sets of seismic monitoring control sets. Each set represents a specific monitoring strategy and is independently executed in the forward simulation to screen out the parameter combination with the optimal response to the wave velocity perturbation caused by infiltration.

[0046] This application can accurately determine the initial frequency based on geological data and empirical formulas, and optimize the frequency bandwidth by combining formation scattering characteristics and signal-to-noise ratio analysis, effectively capturing the dynamic wave velocity change characteristics during the infiltration process. Further, through frequency perturbation and random combination of multiple source types and acquisition methods, diverse seismic monitoring control sets are constructed, enhancing the adaptability and robustness of the monitoring scheme. This method not only improves the sensitivity and resolution of seismic data to wave velocity perturbations in the storage area but also provides a scientific and reasonable parameter basis for subsequent forward simulation and inversion analysis, significantly enhancing the accuracy and reliability of gas permeability monitoring.

[0047] S3. Construct a three-dimensional geological model of the storage area and a simulation model of wave velocity change, and perform forward simulation based on each seismic monitoring control set to obtain wave velocity change response data under different frequency combinations.

[0048] In this embodiment, constructing a three-dimensional geological model of the storage area and a simulation model of wave velocity change specifically includes: Based on the drilling, logging, and seismic data of the target formation, construct a three-dimensional geological model of the storage area; Mesh the geological model; Map the pore pressure change and saturation change caused by carbon dioxide injection into the physical response of wave velocity change; Based on the finite difference method, perform forward simulation on the wave velocity change to obtain the wave field response under different seismic wave frequencies, source types, and acquisition parameter combinations, and obtain the spatial distribution characteristics of wave velocity perturbation; Through the analysis of the simulation results, extract the corresponding relationship between wave velocity change and carbon dioxide infiltration characteristics as the model output result.

[0049] Furthermore, perform forward simulation based on each seismic monitoring control set to obtain wave velocity change response data under different frequency combinations, specifically: Perform forward simulation on each seismic monitoring control set based on the geological model and the simulation model of wave velocity change, extract the characteristics of the simulation results and associate them with the formation position and time to obtain wave velocity change response data.

[0050] It should be noted that by using the finite element or finite difference method, the three-dimensional geological model is divided into regular or irregular grid cells to form a computational grid. The grid division needs to balance the computational accuracy and efficiency to ensure that it can accurately reflect the spatial variation of the geological body and at the same time support the discrete calculation of the wave field in numerical simulation.

[0051] Map the pore pressure change and saturation change caused by carbon dioxide injection into the physical response of wave velocity change: Based on hydrodynamics and rock physics theories, establish the physical relationship between pore pressure and rock elastic parameters (such as the longitudinal wave velocity Vp and shear wave velocity Vs), and convert the pore pressure increase and saturation change caused by the carbon dioxide injection process into the change of local wave velocity. This mapping process is based on poroelastic theory, Gassmann equation or similar rock physics models to quantitatively express the influence of the seepage dynamics in the storage area on wave velocity.

[0052] Through the analysis of the simulation results, extract the corresponding relationship between wave velocity change and carbon dioxide seepage characteristics as the model output result: Analyze the wave field response data obtained from the forward simulation, identify the wave velocity perturbation patterns and their spatio-temporal distributions under different seismic parameter combinations, and establish the corresponding relationship between wave velocity change and carbon dioxide seepage process (such as seepage rate, seepage path and seepage range). This corresponding relationship provides a key physical basis and quantitative reference for subsequent wave velocity change inversion and permeability monitoring.

[0053] Perform forward simulation on each seismic monitoring control set based on the geological model and the wave velocity change simulation model, extract the characteristics of the simulation results and associate them with the formation position and time to obtain the wave velocity change response data: For each seismic monitoring control set (including the combination of frequency perturbation, source type and acquisition method) generated in step S2, carry out forward simulation based on the above three-dimensional geological model and wave velocity change simulation model. By extracting the key wave field characteristics (such as amplitude change, spectrum change, wave velocity perturbation intensity, etc. in time-space) in each simulation result and corresponding them to the specific formation position and time point, a comprehensive and dynamic wave velocity change response data set is formed, laying a data foundation for subsequent evaluation model training and inversion analysis.

[0054] It should be noted that the existing gas permeability monitoring technologies for carbon dioxide geological storage usually collect and invert seismic data in a single frequency band, ignoring the wave velocity response differences of the formation structure at different depths and the multi-scale characteristics of the seepage process. The single-frequency band inversion lacks the coupling and matching mechanism between the formation depth change, seepage dynamics and signal characteristics, resulting in the possibility that the collected seismic signals may not fully reflect the complex seepage behavior in the storage area, and thus the inversion results may have deviations or information omissions. This limitation directly affects the timeliness and accuracy of the safety assessment and risk warning of carbon dioxide storage, increasing the potential risks of geological storage operations.

[0055] In this step, through forward simulation based on multi-frequency combination and simulation of dynamic wave velocity changes, a comprehensive response capture of the formation structures at different depths in the storage area and the multi-scale infiltration process is achieved. Combining diverse combinations of multi-frequencies, source types, and acquisition methods can more effectively extract the spatio-temporal distribution characteristics of wave velocity perturbations, significantly improving the accuracy and stability of wave velocity change inversion. This method promotes the dynamic matching between infiltration dynamic characteristics and seismic signals, enhances the monitoring system's ability to identify the migration behavior of carbon dioxide, improves the reliability of storage safety assessment and the timeliness of risk warning, and helps to achieve more scientific and precise geological storage management.

[0056] S4. Obtain the distribution of wave velocity perturbation characteristics caused by carbon dioxide injection under different seismic monitoring control sets, and combine the wave velocity change response data to weighted-train the set evaluation model.

[0057] In this embodiment, obtaining the distribution of wave velocity perturbation characteristics caused by carbon dioxide injection under different seismic monitoring control sets specifically includes: In the forward simulation results corresponding to each control set, identify the time-series wave velocity perturbation images in the carbon dioxide injection area and its adjacent areas, and extract the perturbation characteristic distribution; The perturbation characteristic distribution includes the perturbation morphology boundary in the injection influence area, forming a spatial mask for the perturbation area; The volume change rate, perturbation boundary advancement speed and direction of the perturbation area at different times; The wave velocity gradient distribution and uniformity index inside the perturbation area; The matching degree between the main propagation path representing the perturbation front and the preset infiltration direction; Construct the perturbation characteristic distribution into a perturbation characteristic vector.

[0058] Furthermore, combining the wave velocity change response data to weighted-train the set evaluation model specifically includes: Use the gradient boosting tree algorithm to construct the set evaluation model; The model uses the combined features composed of the perturbation characteristic vector and the corresponding wave velocity change response data as input to construct training samples; The training samples include the perturbation characteristic vector and the wave velocity change response data; According to the inversion requirements of the target formation, set the set evaluation index as the label value for supervised learning; The label value includes the effective perturbation identification ability characteristics evaluated based on the spatial matching degree between the perturbation distribution and the target storage layer; The time response accuracy characteristics evaluated based on the synchronization between the first appearance time of the perturbation characteristics and the injection; The wave velocity perturbation structure clarity characteristics evaluated based on the interpretability of the perturbation profile in the key area; Based on the input and supervised learning labels, a gradient boosting tree model is used for regression training to output a set evaluation model.

[0059] It should be noted that the recognition of the time series wave velocity perturbation image Based on the forward simulation results corresponding to each seismic monitoring control set, using seismic wave field data with high spatio-temporal resolution, the evolution process of wave velocity perturbation in the injection area and its adjacent areas is identified. By extracting the wave velocity change images at consecutive time nodes, a time series wave velocity perturbation image is formed to reflect the formation dynamic response caused by carbon dioxide injection.

[0060] Formation of perturbation morphology boundary and spatial mask: Through image processing and edge detection algorithms (such as Canny operator, region growing method, etc.), the perturbation boundary of the injection affected area is extracted to form a spatial mask of the perturbation area. The spatial mask is used to define the geographical scope of wave velocity perturbation, facilitating subsequent quantification of perturbation characteristics and spatial matching analysis.

[0061] Volume change rate and boundary advancement speed of the perturbation area: Calculate the spatial volume of the perturbation area at different times, analyze the change rate of the perturbation volume, and reveal the dynamic expansion trend of the carbon dioxide penetration effect. At the same time, by tracking the movement of the perturbation boundary over time, calculate the speed and direction of boundary advancement to characterize the propagation characteristics of the penetration wavefront.

[0062] Inside the perturbation area, calculate the spatial gradient of wave velocity to reflect the local intensity and directionality of wave velocity change, and combine statistical analysis (such as variance, entropy value) to evaluate the uniformity of wave velocity perturbation, characterizing the homogeneous or heterogeneous characteristics of the penetration process.

[0063] Based on the prior injection scheme and geological structure, determine the preset main penetration direction, calculate the main propagation path of the perturbation front through path analysis methods (such as the shortest path algorithm or streamline tracing), and conduct a matching degree evaluation with the preset penetration direction to verify the rationality of wave velocity perturbation and the compliance of the penetration process.

[0064] Combine the above-mentioned multiple quantitative characteristics such as perturbation morphology boundary, volume change rate, boundary velocity, wave velocity gradient, uniformity index, and propagation path matching degree into a perturbation feature vector, which is used as the input data for the subsequent machine learning model to completely describe the wave velocity perturbation information caused by carbon dioxide injection.

[0065] It should be noted that the gradient boosting tree algorithm is used for the training of the set evaluation model. GBT is an ensemble learning method that gradually reduces the prediction error through multiple rounds of iteration, has good non-linear modeling ability and robustness, and is suitable for processing complex multi-dimensional seismic monitoring data and perturbation characteristics.

[0066] Fuse the perturbation feature vector with the corresponding wave velocity change response data to form a combined feature input. The wave velocity change response data provides supplementary information on the spatio-temporal wave field characteristics, enhancing the model's learning ability for multi-dimensional formation responses.

[0067] According to the inversion requirements of the target formation, design reasonable ensemble evaluation indicators as the labels (target variables) for model training to guide the model to learn the advantages and disadvantages of different seismic monitoring control ensembles. The labels cover three aspects: effective perturbation identification ability, time response accuracy, and clarity of wave velocity perturbation structure, ensuring that the comprehensive evaluation indicators output by the model are highly representative and practical.

[0068] Effective perturbation identification ability feature: Based on the degree of spatial matching between the perturbation distribution and the target storage layer, evaluate the ensemble's ability to capture wave velocity perturbations in key permeable regions.

[0069] Time response accuracy feature: Using the synchronization between the time when the perturbation feature first appears and the actual injection time as an indicator, reflect the real-time response ability of the monitoring system to the permeation process.

[0070] Clarity of wave velocity perturbation structure feature: Evaluate the interpretability of the perturbation contour in the key area, reflecting the spatial resolution and signal-to-noise ratio of the seismic wave velocity perturbation image.

[0071] Using the labeled training samples, adopt the gradient boosting tree model for regression training, gradually optimize the model parameters, so that the model can accurately predict the ensemble evaluation value corresponding to the input features. Finally, the model outputs a scalar evaluation value, which is used to comprehensively evaluate the advantages and disadvantages of different seismic monitoring control ensembles and assist in screening the optimal acquisition scheme.

[0072] In this embodiment, by constructing a perturbation feature vector and a combined feature that combines wave velocity change response data, and using the gradient boosting tree algorithm to establish an ensemble evaluation model, the quantitative evaluation and intelligent screening of different seismic monitoring control ensembles are realized. This method can effectively capture the spatial dynamic characteristics and multi-dimensional wave velocity responses during the permeation process, improve the scientificity and accuracy of the monitoring scheme, and provide solid data support and decision-making basis for ensuring the safety of carbon dioxide geological storage.

[0073] In the prior art, seismic inversion for geological carbon dioxide sequestration mostly monitors and inverses the wave velocity change using a single frequency band, without fully considering the complex changes of the formation structure in the vertical direction and the multi-scale seepage characteristics during the carbon dioxide seepage process. This single-frequency-band inversion method lacks a dynamic matching mechanism for the different response characteristics of each depth of the formation and the frequency characteristics of seismic signals, resulting in prone deviation of the inversion results and inability to comprehensively reflect the spatial and temporal dynamic changes of the seepage process. In addition, ignoring frequency diversity may also cause the loss of key wave velocity perturbation information, reducing the evaluation accuracy of the safety of the sequestration area and the timeliness of risk warning. Therefore, there is an urgent need to develop a comprehensive inversion method that combines multi-frequency combinations and a seismic monitoring control set of multi-parameters to improve the capture ability and identification accuracy of the dynamic response of the formation caused by carbon dioxide injection.

[0074] In this step, a seismic monitoring control set with multi-frequency combinations and random combinations of different seismic source types and acquisition methods is constructed, and forward simulation is carried out based on a three-dimensional geological model and a wave velocity change simulation model to systematically obtain wave velocity change response data and wave velocity perturbation characteristic distributions under multi-frequency and multi-parameter conditions. This method realizes the dynamic matching of the formation structures at different depths and seepage scale characteristics, significantly improving the spatial resolution and time response accuracy of the seismic inversion results. It effectively avoids information omission and error accumulation caused by single-band inversion, enhancing the sensitivity and identification ability for the carbon dioxide seepage process. Furthermore, it improves the accuracy of the safety assessment of the geological sequestration area and the timeliness of risk warning, providing a more reliable technical support for the scientific decision-making of the sequestration monitoring system.

[0075] S5. Construct a second set according to the evaluation result of the set evaluation model and the corresponding seismic monitoring control set.

[0076] Each second set includes a seismic monitoring control set and an output value of the corresponding set evaluation model.

[0077] S6. Fit a multi-dimensional surface according to the second set, and screen the optimal seismic monitoring control set for seismic data acquisition and inversion.

[0078] In this embodiment, fitting a multi-dimensional surface according to the second set and screening the optimal seismic monitoring control set for seismic data acquisition and inversion specifically includes: Mapping the second set to a multi-dimensional coordinate system, and fitting the mapped points of several second sets into a surface by the curve fitting method; Derive the surface, and analyze to obtain the seismic monitoring control set corresponding to the most stable peak for seismic data acquisition and inversion.

[0079] It should be noted that by calculating the partial derivatives (i.e., the gradients in each dimension) of the multi-dimensional surface obtained by fitting, the rate of change information of each point on the surface can be obtained. The stationary point is the point where the gradient is close to zero, indicating that the change of the surface at this point tends to be gentle, without a significant upward or downward trend. Further, by calculating the second derivative (i.e., the Hessian matrix) of this point to judge its nature, it is ensured to be a local maximum (i.e., the surface is concave in all directions at this point), so as to confirm that this is a stable and optimal peak point. The seismic monitoring control set corresponding to this point is selected as the optimal solution to ensure the optimal and stable comprehensive performance of the evaluation index, reduce the uncertainty caused by the index fluctuation, and improve the effect and accuracy of seismic data acquisition and inversion.

[0080] Example 2 Figure 2 The gas permeability monitoring system based on carbon dioxide geological storage of the present invention is given, including a constraint module, a control set module, a geology module, a set evaluation module, and a screening module; The constraint module is used to obtain the geological data of the target formation and determine the first constraint conditions required for inversion; The control set module is used to perturb the seismic wave frequency based on the first constraint conditions, and randomly combine different source types and acquisition methods to generate a number of seismic monitoring control sets; The geology module is used to construct a three-dimensional geological model of the storage area and a wave velocity change simulation model, and perform forward simulation based on each seismic monitoring control set to obtain the wave velocity change response data under different frequency combinations; The set evaluation module is used to obtain the distribution of wave velocity perturbation characteristics caused by carbon dioxide injection under different seismic monitoring control sets, and combine the wave velocity change response data to weighted train the set evaluation model; The screening module is used to construct a second set according to the evaluation results of the set evaluation model and the corresponding seismic monitoring control sets; fit a multi-dimensional surface according to the second set, and screen the optimal seismic monitoring control set for seismic data acquisition and inversion.

Claims

1. A method for monitoring gas permeability based on geological sequestration of carbon dioxide, characterized in that, It includes the following steps: Obtain the geological data of the target formation and determine the first constraint conditions required for inversion; Based on the first constraint conditions, perturb the seismic wave frequencies and randomly combine different source types and acquisition methods to generate a number of seismic monitoring control sets; Construct a three-dimensional geological model of the storage area and a wave velocity change simulation model, and perform forward simulation based on each seismic monitoring control set to obtain wave velocity change response data under different frequency combinations; Obtain the distribution of wave velocity perturbation characteristics caused by carbon dioxide injection under different seismic monitoring control sets, and combine the wave velocity change response data to weightedly train the set evaluation model; Construct a second set according to the evaluation results of the set evaluation model and the corresponding seismic monitoring control sets; Fit a multi-dimensional surface according to the second set and screen the optimal seismic monitoring control set for seismic data acquisition and inversion.

2. The gas permeability monitoring method based on carbon dioxide geological storage according to claim 1, characterized in that The obtaining of the geological data of the target formation and determining the first constraint conditions required for inversion is specifically: Based on the core test results and empirical models, obtain the lithology types and their corresponding pore permeability characteristics; Based on the core test fitting method, convert the lithology types and their corresponding pore permeability characteristics into sensitivity factors of wave velocity perturbation, form a lithology-perturbation response matrix, and quantitatively express the degree of wave velocity change caused by the penetration of carbon dioxide by different lithologies; According to the lithology-perturbation response matrix, combine the obtained injection history and the pressure gradient, injection rate, and formation noise level in the predicted working conditions to set the first constraint conditions for inversion.

3. The gas permeability monitoring method based on carbon dioxide geological storage according to claim 2, wherein The perturbing of the seismic wave frequencies based on the first constraint conditions and randomly combining different source types and acquisition methods to generate a number of seismic monitoring control sets is specifically: Calculate the first initial frequency based on the geological data combined with an empirical formula; Obtain the optimal frequency bandwidth range by combining formation scattering characteristics and signal-to-noise ratio analysis, and the formation scattering characteristics and signal-to-noise ratio analysis include satisfying the maximum signal-to-noise ratio and covering the preset geological interface response frequency band; Adjust the first initial frequency to capture the dynamic wave velocity change caused by penetration, and use the frequency points matching the time constant of the penetration process as the second initial frequency; Randomly perturb the second initial frequency under the condition of satisfying the first constraint conditions to generate a number of frequencies to be screened; Randomly combine the number of frequencies to be screened with different source types and acquisition methods to generate a number of seismic monitoring control sets.

4. The gas permeability monitoring method based on carbon dioxide geological sequestration according to claim 3, wherein The constructing of the three-dimensional geological model of the storage area and the wave velocity change simulation model is specifically: Based on the drilling, logging, and seismic data of the target formation, construct a three-dimensional geological model of the storage area; Mesh the geological model; Map the pore pressure change and saturation change caused by carbon dioxide injection into the physical response of wave velocity change; Based on the finite difference method, perform forward simulation on the wave velocity change to obtain the wave field response under different combinations of seismic wave frequencies, source types, and acquisition parameters, and obtain the spatial distribution characteristics of wave velocity perturbation; Through the analysis of the simulation results, extract the corresponding relationship between the wave velocity change and the carbon dioxide penetration characteristics as the model output result.

5. The gas permeability monitoring method based on carbon dioxide geological storage according to claim 4, wherein The performing of forward simulation based on each seismic monitoring control set to obtain wave velocity change response data under different frequency combinations is specifically: Forward modeling is performed on each seismic monitoring control set based on the geological model and the wave velocity change simulation model, the characteristics of the simulation results are extracted and associated with the formation position and time to obtain wave velocity change response data.

6. The gas permeability monitoring method based on carbon dioxide geological sequestration according to claim 5, characterized in that The obtaining of the wave velocity perturbation characteristic distribution caused by carbon dioxide injection under different seismic monitoring control sets is specifically as follows: In the forward simulation results corresponding to each control set, identify the time series wave velocity perturbation images in the carbon dioxide injection area and its adjacent areas, and extract the perturbation characteristic distribution; The perturbation characteristic distribution includes the perturbation form boundary of the injection influence area to form a spatial mask of the perturbation area; The volume change rate, perturbation boundary advancement speed and direction of the perturbation area at different times; The wave velocity gradient distribution and uniformity index inside the perturbation area; The matching degree between the main propagation path representing the perturbation front and the preset penetration direction; Construct the perturbation characteristic distribution into a perturbation characteristic vector.

7. The gas permeability monitoring method based on carbon dioxide geological sequestration according to claim 6, characterized in that, Combining the wave velocity change response data to weighted train the set evaluation model is specifically as follows: Use the gradient boosting tree algorithm to construct the set evaluation model; The model takes the combined features composed of the perturbation characteristic vector and the corresponding wave velocity change response data as input to construct training samples; The training samples include the perturbation characteristic vector and the wave velocity change response data; According to the inversion requirements of the target formation, set the set evaluation index as the label value of supervised learning; The label value includes the effective perturbation identification ability characteristic evaluated based on the spatial matching degree between the perturbation distribution and the target storage layer; The time response accuracy characteristic evaluated based on the synchronization of the first appearance time of the perturbation characteristic and the injection; The wave velocity perturbation structure clarity characteristic evaluated based on the interpretability of the perturbation profile in the key area; Based on the input and the label of supervised learning, use the gradient boosting tree model for regression training and output the set evaluation model.

8. The gas permeability monitoring method based on carbon dioxide geological sequestration according to claim 7, characterized in that The constructing of the second set according to the evaluation result of the set evaluation model and the corresponding seismic monitoring control set is specifically as follows: Each second set includes a seismic monitoring control set and the output value of a corresponding set evaluation model.

9. The gas permeability monitoring method based on carbon dioxide geological storage according to claim 8, characterized in that The fitting of the multi-dimensional surface according to the second set and screening the optimal seismic monitoring control set for seismic data acquisition and inversion is specifically as follows: Map the second set to the multi-dimensional coordinate system, and use the curve fitting method to fit the points mapped by several second sets into a surface; Derive the surface and analyze to obtain the seismic monitoring control set corresponding to the most stable peak for seismic data acquisition and inversion.

10. A system using the gas permeability monitoring method based on carbon dioxide geological storage as described in any one of claims 1-9, characterized in that, It includes a constraint module, a control set module, a geology module, a set evaluation module, and a screening module; The constraint module is used to obtain the geological data of the target formation and determine the first constraint conditions required for inversion; The control set module is used to perturb the seismic wave frequency based on the first constraint conditions, and randomly combine different seismic source types and acquisition methods to generate several seismic monitoring control sets; The geology module is used to construct a three-dimensional geological model of the storage area and a wave velocity change simulation model, and perform forward simulation based on each seismic monitoring control set to obtain wave velocity change response data under different frequency combinations; The set evaluation module is used to obtain the characteristic distribution of wave velocity perturbations caused by carbon dioxide injection under different seismic monitoring control sets, and combine the wave velocity change response data to weightedly train the set evaluation model; The screening module is used to construct a second set according to the evaluation results of the set evaluation model and the corresponding seismic monitoring control sets; fit a multi-dimensional surface according to the second set, and screen the optimal seismic monitoring control set for seismic data acquisition and inversion.

Images