Methods, apparatus, equipment, media and procedures for evaluating geological carbon sequestration potential

Abstract

This application discloses a method, apparatus, equipment, medium, and program product for evaluating geological carbon sequestration potential. The method includes: preprocessing an initial geological carbon sequestration training dataset to obtain a target geological carbon sequestration training dataset; constructing a transfer optimization model based on the target geological carbon sequestration training dataset; inputting the target geological carbon sequestration inference dataset of the research area of ​​a depleted marine oil and gas field into the transfer optimization model for multi-field coupled simulation to obtain target geological carbon sequestration evolution data; calculating the initial total geological carbon sequestration based on the target geological carbon sequestration evolution data; performing marine area correction on the initial total geological carbon sequestration to obtain the target total geological carbon sequestration; the target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​a depleted marine oil and gas field. This application can improve the reuse rate of oil and gas data and significantly reduce the amount of training data to reduce the cost of evaluating the geological carbon sequestration potential of depleted oil and gas fields, and can be widely applied in the field of intelligent geological carbon sequestration technology.

Description

Technical Field

[0001] This application relates to the field of intelligent geological carbon sequestration technology, and in particular to methods, apparatus, equipment, media and program products for evaluating geological carbon sequestration potential. Background Technology

[0002] Currently, carbon dioxide (CO2) Carbon capture, utilization and storage (CCUS) is a key technology for achieving global climate goals. Among them, marine geological carbon sequestration involves injecting carbon dioxide captured in industrial or oil and gas development into geological bodies such as seabed saline water layers and depleted oil and gas reservoirs in a supercritical state, thereby achieving long-term isolation from the atmosphere.

[0003] However, currently, in the context of depleted gas reservoirs... Research on methods for calculating geological storage potential is still incomplete and has significant limitations. In traditional depleted gas reservoirs... Geological storage potential calculation methods are typically used in gas reservoirs. Based on the theoretical reserves, considering Dissolution in formation water and It undergoes mineralization reactions with water and rocks to establish Effective buried reserves calculation method, and for Research on methods for evaluating storage potential mainly focuses on saline aquifers, deep-sea sediments, and oil and gas field development. While some technological advancements have been made in understanding the geological potential for gas storage, there are still limitations in addressing the challenges posed by depleted gas reservoirs. However, the evaluation of geological storage potential is relatively weak, especially in traditional depleted gas reservoirs. The methods for calculating geological storage potential are prone to resulting in low data reuse rates and high evaluation costs.

[0004] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0005] The embodiments of this application aim to at least partially solve one of the technical problems in the related art. Therefore, the main objective of the embodiments of this application is to propose a method, apparatus, equipment, medium, and program product for evaluating geological carbon sequestration potential, which can improve the reuse rate of oil and gas data and significantly reduce training data to lower the cost of evaluating the geological carbon sequestration potential of depleted offshore oil and gas fields.

[0006] To achieve the above objectives, one aspect of this application proposes a method for evaluating geological carbon sequestration potential, the method comprising the following steps:

[0007] An initial geological carbon sequestration training dataset was obtained for the experimental area of ​​a depleted marine oil and gas field. The initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environmental data.

[0008] The initial geological carbon sequestration training dataset is preprocessed to obtain the target geological carbon sequestration training dataset;

[0009] Based on the target geological carbon sequestration training dataset, a transfer optimization model is constructed;

[0010] The target geological carbon sequestration inference dataset of the research area of ​​depleted oil and gas fields in the sea is input into the migration optimization model for multi-field coupled simulation processing to obtain the target geological carbon sequestration evolution data.

[0011] Based on the target geological carbon sequestration evolution data, calculate the initial total geological carbon sequestration in the study area of ​​the depleted marine oil and gas field;

[0012] The initial total geological carbon sequestration is adjusted for marine area to obtain the target total geological carbon sequestration; the target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted oil and gas field in the marine area.

[0013] To achieve the above objectives, another aspect of this application provides a geological carbon sequestration potential evaluation device, which includes the following modules:

[0014] The initial training data acquisition module is used to acquire the initial geological carbon sequestration training dataset of the experimental area of ​​the depleted marine oil and gas field; the initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data and engineering environment data;

[0015] A geological carbon sequestration data preprocessing module is used to preprocess the initial geological carbon sequestration training dataset to obtain the target geological carbon sequestration training dataset.

[0016] The migration optimization model construction module is used to construct a migration optimization model based on the target geological carbon sequestration training dataset.

[0017] The multi-field coupled simulation processing module is used to input the target geological carbon sequestration inference dataset of the research area of ​​the depleted oil and gas field in the sea into the migration optimization model for multi-field coupled simulation processing to obtain the target geological carbon sequestration evolution data.

[0018] The initial geological carbon sequestration total calculation module is used to calculate the initial geological carbon sequestration total of the research area of ​​the depleted marine oil and gas field based on the target geological carbon sequestration evolution data.

[0019] The target geological carbon sequestration total calculation module is used to perform marine area correction on the initial geological carbon sequestration total to obtain the target geological carbon sequestration total; the target geological carbon sequestration total is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted oil and gas field in the marine area.

[0020] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0021] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0022] To achieve the above objectives, another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0023] The embodiments of this application include at least the following beneficial effects: This application provides a method, apparatus, equipment, medium, and program product for evaluating geological carbon sequestration potential. This scheme obtains an initial geological carbon sequestration training dataset for an experimental area of ​​a depleted offshore oil and gas field. The initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environment data. The initial geological carbon sequestration training dataset is preprocessed to obtain a target geological carbon sequestration training dataset. A migration optimization model is constructed based on the target geological carbon sequestration training dataset. The target geological carbon sequestration inference dataset of the research area of ​​the depleted offshore oil and gas field is input into the migration optimization model for multi-field coupled simulation processing to obtain target geological carbon sequestration evolution data. Based on the target geological carbon sequestration evolution data, the initial total geological carbon sequestration of the research area of ​​the depleted offshore oil and gas field is calculated. The initial total geological carbon sequestration is corrected for offshore area to obtain the target total geological carbon sequestration. The target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted offshore oil and gas field. This application embodiment combines transfer learning technology with multi-source data fusion technology to transfer reservoir property data from the exploration stage of depleted marine oil and gas fields to a carbon sequestration capacity evaluation model. This can quantify the geological carbon sequestration capacity, thereby assessing the geological carbon sequestration potential of depleted marine oil and gas fields based on the geological carbon sequestration capacity. This can improve the reuse rate of oil and gas data and significantly reduce the amount of training data, thereby reducing the cost of evaluating the geological carbon sequestration potential of depleted marine oil and gas fields. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the steps of a method for evaluating geological carbon sequestration potential provided in an embodiment of this application;

[0025] Figure 2 This is a schematic diagram of the structure of a multi-source data integration module provided in an embodiment of this application;

[0026] Figure 3 yes Figure 1 The flowchart of step S102 in the document;

[0027] Figure 4 This is a schematic diagram of the structure of a data preprocessing module provided in an embodiment of this application;

[0028] Figure 5 This is a schematic diagram of a process for constructing a migration optimization model provided in an embodiment of this application;

[0029] Figure 6 yes Figure 1 The flowchart of step S105 in the process;

[0030] Figure 7 This is a flowchart illustrating a method for evaluating geological carbon sequestration potential provided in an embodiment of this application;

[0031] Figure 8 This is a schematic diagram of a process for dynamically updating and correcting the archived capacity provided in an embodiment of this application;

[0032] Figure 9 This is a schematic diagram of the structure of a geological carbon sequestration potential evaluation device provided in an embodiment of this application;

[0033] Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0035] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0036] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0038] Currently, carbon dioxide (CO2) Carbon capture, utilization, and storage (CCUS) is a key technology for achieving global climate goals. Marine geological carbon sequestration involves injecting carbon dioxide captured during industrial or oil and gas development into supercritical states in seabed brackish water layers, depleted oil and gas reservoirs, and other geological bodies, achieving long-term isolation from the atmosphere. Compared to terrestrial carbon sequestration, marine geological carbon sequestration has significant advantages in that it is far from human activity areas and covered by seawater, offering both high safety and low environmental risk. Depleted oil and gas reservoirs, with their clearly defined storage space and mature geological parameters, are the preferred candidates for verifying techno-economic feasibility.

[0039] However, currently, in the context of depleted gas reservoirs... Research on methods for calculating geological storage potential is still incomplete and has significant limitations. In traditional depleted gas reservoirs... Geological storage potential calculation methods are typically used in gas reservoirs. Based on the theoretical reserves, considering Dissolution in formation water and It undergoes mineralization reactions with water and rocks to establish Effective buried reserves calculation method, and for Research on methods for evaluating storage potential mainly focuses on saline aquifers, deep-sea sediments, and oil and gas field development. While some technological advancements have been made in understanding the geological potential for gas storage, there are still limitations in addressing the challenges posed by depleted gas reservoirs. However, the evaluation of geological storage potential is relatively weak, and at the same time, the assessment of depleted gas reservoirs is also lacking. Methods for evaluating geological storage potential are also relatively scarce. For example, in traditional depleted gas reservoirs... Methods for calculating geological storage potential often rely on a single geological model, resulting in low data reuse rates and high evaluation costs.

[0040] In view of this, embodiments of this application provide a method, apparatus, equipment, medium, and program product for evaluating geological carbon sequestration potential. This scheme obtains an initial geological carbon sequestration training dataset for an experimental area of ​​a depleted offshore oil and gas field. The initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environment data. The initial geological carbon sequestration training dataset is preprocessed to obtain a target geological carbon sequestration training dataset. A migration optimization model is constructed based on the target geological carbon sequestration training dataset. The target geological carbon sequestration inference dataset for the research area of ​​the depleted offshore oil and gas field is input into the migration optimization model for multi-field coupled simulation processing to obtain target geological carbon sequestration evolution data. Based on the target geological carbon sequestration evolution data, the initial total geological carbon sequestration for the research area of ​​the depleted offshore oil and gas field is calculated. The initial total geological carbon sequestration is corrected for offshore conditions to obtain the target total geological carbon sequestration. The target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted offshore oil and gas field. This application embodiment combines transfer learning technology with multi-source data fusion technology to transfer reservoir property data from the exploration stage of depleted marine oil and gas fields to a carbon sequestration capacity evaluation model. This can quantify the geological carbon sequestration capacity, thereby assessing the geological carbon sequestration potential of depleted marine oil and gas fields based on the geological carbon sequestration capacity. This can improve the reuse rate of oil and gas data and significantly reduce the amount of training data, thereby reducing the cost of evaluating the geological carbon sequestration potential of depleted marine oil and gas fields.

[0041] The geological carbon sequestration potential evaluation method provided in this application relates to the field of intelligent geological carbon sequestration technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the geological carbon sequestration potential evaluation method, but is not limited to the above forms.

[0042] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0043] Please see Figure 1 , Figure 1 This is an optional flowchart of the geological carbon sequestration potential evaluation method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0044] Step S101: Obtain the initial geological carbon sequestration training dataset for the experimental area of ​​the depleted oil and gas field in the sea; the initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environment data;

[0045] Among them, depleted marine oil and gas fields refer to offshore oil and gas fields that have been fully exploited or are close to being abandoned, and the experimental area of ​​depleted marine oil and gas fields refers to the area that provides experimental data for building migration optimization models.

[0046] Optionally, the initial geological carbon sequestration training dataset is used for subsequent model adversarial training. The initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environment data.

[0047] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a multi-source data integration module provided in an embodiment of this application, such as... Figure 2 As shown, Figure 2In the diagram, number 200 represents the multi-source data integration module, number 201 represents the source domain data integration unit, number 202 represents the target domain data integration unit, and number 203 represents the engineering environment data integration unit. Specifically, source domain data integration unit 201 integrates source domain data, which mainly includes oil and gas reservoir data and geophysical exploration data; target domain data integration unit 202 integrates target domain data, which mainly includes sealing site data; and engineering environment data integration unit 203 integrates engineering environment data.

[0048] The integration of source domain data (oil and gas reservoirs) mainly involves extracting reservoir physical property parameters (i.e., source domain oil and gas data) such as reservoir thickness, area, porosity (e.g., 12%-40%), permeability (e.g., 1-500mD), and sand-soil ratio from depleted oil and gas fields to construct a pre-training dataset. In addition, the integration of source domain data also includes organizing geophysical exploration data (seismic profiles, well logging curves) for the calculation and correction of reservoir physical property data, as well as the acquisition or correction of relevant parameters for subsequent storage capacity calculation.

[0049] The integration of target domain data (storage sites) mainly involves the reservoir parameters of the deep saline aquifer in the target sea area (thickness, area, salinity (e.g., 3-35 g / L), pressure gradient (e.g., 0.8-1.2 MPa), and geothermal gradient, etc.). In other words, the target domain storage data is integrated, and a virtual scene covering the target domain with a water depth of n meters (e.g., 500-3000 meters) is generated through Monte Carlo simulation for subsequent model adversarial training.

[0050] The integration of engineering environmental data mainly involves collecting well network configuration, injection pressure history, oil production, caprock lithology, fault distribution, and groundwater salinity in depleted oil and gas fields. This data is used for establishing oil and gas migration coupling function models, dynamically updating carbon sequestration capacity, and calculating marine area corrections.

[0051] Step S102: Preprocess the initial geological carbon sequestration training dataset to obtain the target geological carbon sequestration training dataset;

[0052] Please see Figure 3 , Figure 3 yes Figure 1 In the flowchart of step S102, in some embodiments, step S102 may include, but is not limited to, steps S301 to S303:

[0053] Step S301: According to the normalization formula, the first initial geological carbon sequestration training data that is smaller than the preset dimensional span threshold in the initial geological carbon sequestration training dataset is normalized to obtain the first target geological carbon sequestration training data.

[0054] Step S302: According to the standardization formula, the second initial geological carbon sequestration training data in the initial geological carbon sequestration training dataset that is greater than the preset dimensional span threshold is standardized to obtain the second target geological carbon sequestration training data.

[0055] Step S303: Construct the target geological carbon sequestration training dataset based on the first target geological carbon sequestration training data and the second target geological carbon sequestration training data.

[0056] The first initial geological carbon sequestration training data refers to a dataset with a dimension span smaller than a preset threshold, i.e., a dataset with a small dimension span; the second initial geological carbon sequestration training data refers to a dataset with a dimension span larger than the preset threshold, i.e., a dataset with a large dimension span. The preset dimension span threshold can be set according to actual conditions, and this embodiment does not impose any restrictions on it.

[0057] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a data preprocessing module provided in an embodiment of this application, as shown below. Figure 4 As shown, Figure 4 In the diagram, number 400 represents the data preprocessing module, number 401 represents the data normalization processing unit, and number 402 represents the data standardization processing unit. Specifically, data normalization processing unit 401 is used to normalize datasets with small dimension spans; data standardization processing unit 402 is used to standardize datasets with large dimension spans.

[0058] In step S102, the core parameters such as reservoir thickness, area, porosity, permeability and sand-to-soil ratio corresponding to the target domain and source domain are preprocessed to eliminate dimensional differences. The preprocessed core parameters are collected to form a dataset for subsequent model adversarial training to build a transfer optimization model.

[0059] Specifically, based on the size distribution characteristics of the collected data (initial geological carbon sequestration training dataset), the "maximum-minimum method" normalization formula is applied point-by-point to calculate the first target geological carbon sequestration training data for small datasets with varying spans (i.e., the first initial geological carbon sequestration training data). The calculation expression for the normalization formula is as follows:

[0060]

[0061] in, Represents the actual value of a specific data point in the original dataset; This represents the maximum value in the original dataset; This represents the minimum value in the original dataset; This represents the normalized value of a specific data point in the original dataset.

[0062] For datasets with large spans in dimensions such as permeability (i.e., the second initial geological carbon sequestration training data), the Z-score standardization formula is used to calculate the second target geological carbon sequestration training data. The calculation expression for the standardization formula is as follows:

[0063]

[0064]

[0065]

[0066] in, Represents the actual value of a specific data point in the original dataset; This represents the average value of all samples in the dataset. This represents the standard deviation of all samples in the dataset; Represents the first in the dataset One data point; Indicates the total amount of data; This represents the standardized value of a specific data point in the original dataset.

[0067] Step S103: Construct a migration optimization model based on the target geological carbon sequestration training dataset;

[0068] In some embodiments, step S103 may include: labeling the target geological carbon sequestration training dataset to obtain a cross-domain label dataset; performing stratified sampling on the cross-domain label dataset to obtain a cross-domain adversarial training dataset; wherein the cross-domain adversarial training dataset includes a training set, a validation set, and a test set; performing adversarial training on the initial generative adversarial network based on the training set and the validation set; terminating the training of the initial generative adversarial network when the distribution difference between the generated data of the initial generative adversarial network and the target domain data reaches a preset convergence threshold, thereby obtaining the target generative adversarial network; and performing adversarial verification on the target generative adversarial network based on the training set and the validation set. The target generative adversarial network (GAN) is optimized, and a binary classification model is trained based on the training and validation sets. The trained binary classification model is used as the target discriminator of the optimized GAN. The test set is input into the optimized GAN, and the target domain probability is output by the target discriminator. The optimized GAN is then judged based on the target domain probability to determine whether it meets the adversarial training requirements. If the optimized GAN meets the adversarial training requirements, the physical property parameters of the carbon-encapsulated storage layer of the target domain generated by the target generator in the GAN are obtained. A transfer optimization model is constructed based on the physical property parameters of the carbon-encapsulated storage layer of the target domain using a multi-field coupling optimization algorithm.

[0069] Please see Figure 5 Please see Figure 5 This is a schematic diagram of a process for constructing a migration optimization model provided in an embodiment of this application, such as... Figure 5 As shown, the process of constructing the migration optimization model includes, but is not limited to, the following steps S501-S506:

[0070] Step S501, Adversarial Feature Construction: Data sets including reservoir thickness, porosity, and sand-to-soil ratio of oil and gas reservoirs (source domain) and storage sites (target domain) are added with "domain labels" (0 for source domain, 1 for target domain) as input features for adversarial training. The dataset is then stratified into training, validation, and test sets (70% / 15% / 15% of the total dataset). The training set serves as the primary data for model learning, the validation set is mainly used for parameter tuning and training monitoring, and the test set is used to evaluate model performance.

[0071] The purpose of constructing adversarial features is to use source domain data to generate virtual target domain data, and then use a discriminator to distinguish between virtual and real data, thereby achieving the training objective.

[0072] Step S502, adversarial training:

[0073] (1) A fixed generator is used to generate virtual target domain data by inputting source domain (oil and gas reservoir) parameters. The discriminator is trained to distinguish between real target domain data and generated virtual target domain data. By distinguishing between real target domain data (storage site, such as measured permeability of an oil field) and generated data (simulated permeability distribution), the generator is forced to learn the geological feature distribution of the target domain (storage site).

[0074] Understandably, the generator generates virtual target domain data based on oil and gas reservoir (source domain) data, and then compares the virtual target domain data with the storage site (target domain) data to learn the geological feature distribution of the storage site (target domain).

[0075] (2) Fix the discriminator and optimize the generator so that the discriminator cannot distinguish the data source. The loss function is as follows:

[0076]

[0077] in, Represents the generator loss function; Represents the mathematical expectation based on the source domain distribution; This represents the logarithm of the probability that the generated data is judged as "real" by the discriminator. This represents the parameters sampled from the source domain (simulated or historical data); This represents the virtual target domain data generated by the generator based on the source domain parameters; This represents the output probability of the discriminator (0 for generated data, 1 for real data).

[0078] When the discriminator determines the probability of generating data When the value stabilizes around 0.5 (indistinguishable from the source) and the distribution of geological parameters shows no significant difference in statistical tests, it is considered converged. In the specific implementation, the above steps are repeated until convergence. The generator adjusts the mapping of source domain parameters (such as sand-to-land ratio) to make the distribution of generated data (virtual target domain sand-to-land ratio) approximate the distribution of the real target domain, until the discriminator can no longer distinguish them.

[0079] In assessing the storage potential of depleted oil and gas fields, the generator task is to generate parameters (such as porosity, permeability, and sand-to-soil ratio) consistent with the geological distribution of the target domain. For example: Input The sand-soil ratio of mature oil and gas fields in the source region (i.e. =Sand source), generator output That is, the sand-to-soil ratio of the simulated target storage area ( =Virtual sand ratio), discriminator output That is, the discriminator estimates the probability of whether the virtual sand-land ratio comes from the target domain, and outputs the probability by optimizing the target. (Virtual sand-to-land ratio) ≈ 0.5; and gradually reduce the difference in geological parameters between the source domain and the target domain to make the generated data consistent with the actual sand-to-land ratio distribution of the target domain.

[0080] Step S503, Adversarial Validation: Perform distribution difference detection, train a binary classification model (LightGBM) using core parameters such as porosity, sand-to-soil ratio and permeability of the source and target reservoirs, predict whether a sample belongs to the source or target domain, and output the probability that a sample belongs to the target domain. The predicted probability of each sample using the binary classification model as a discriminator is the discrimination score.

[0081] In this embodiment, the binary classification model constructed in step S503 replaces the discriminator that originally participated in training and optimization in step S502. The binary classification model is used to solve the problem of data distribution differences and optimize the subsequent transfer optimization model.

[0082] In the process of training a binary classification model, training set, validation set and test set are required. The training set is used to update the model parameters of the binary classification model, the validation set guides hyperparameter optimization and training control, and the test set is only used for the final model performance evaluation and does not participate in the training process.

[0083] The binary classification model is a specific application of LightGBM in a binary classification scenario. It outputs the probability that a sample belongs to the positive class (range [0,1]), and determines the final class by using a threshold. The objective function is 'binary', using log loss (binary_logloss) as the optimization objective. The output format is: predict_proba() returns the positive class probability, and predict() returns the 0 / 1 class label. The formula for the probability value output by the predict() function is as follows:

[0084]

[0085] in, This represents the predicted probability value, i.e., the model based on features. The calculated conditional probability that the sample belongs to the positive class (y=1) is: y represents the target variable. When y=1, the sample is determined to belong to the target domain, i.e. the storage potential area; X represents the input feature matrix, which contains the feature data of the sample to be predicted, such as reservoir thickness, porosity, sand-soil ratio, permeability, etc.

[0086] Optionally, the output results and discrimination suggestions of the binary classification model (LightGBM) in this embodiment are shown in Table 1 below (which can be set according to actual conditions):

[0087] Table 1: Suggestions for Judging Results of Binary Classification Models

[0088]

[0089] In the specific implementation, the output results of the probability value formula are used to identify and improve the generator forged samples, while the discriminator is optimized for subsequent model construction and computation.

[0090] Step S504, Score sorting and threshold traversal, calculate TPR / FPR: Sort all samples from high to low according to their discrimination scores to generate a sorted list. Use each discrimination score as a classification threshold to calculate the corresponding True Positive Rate (TPR) and False Positive Rate (FPR):

[0091]

[0092]

[0093] Using the formulas for calculating the True Positive Rate (TPR) and False Positive Rate (FPR), a series of (FPR, TPR) points are generated. Each point corresponds to the classification performance at a certain threshold. Connecting these points forms the ROC curve (Receiver Operating Characteristic Curve), and the point closest to the top left corner of the curve corresponds to the optimal classification threshold.

[0094] Step S505, calculate the AUC index: Calculate the area under the ROC curve, i.e., the AUC index, using the trapezoidal rule or the Man-Whitney U statistic. The formula for calculating AUC using the Man-Whitney U statistic method is as follows:

[0095]

[0096] in, Indicates the number of samples in the source domain; Indicates the number of samples in the target domain; This represents the discriminator's prediction score for the source domain sample; This represents the discriminator's prediction score for the target domain sample; This is an indicator function; it takes the value 1 if the condition is met, and 0 otherwise.

[0097] Specifically, the discriminator is trained to calculate the AUC metric. If AUC > 0.85, it indicates a significant difference in distribution between the source and target domains, requiring parameter transfer. If AUC < 0.6, the distribution difference is considered to have significantly decreased. The generator of the Generative Adversarial Network (GAN) is used to adjust the parameter distribution of the source domain, making the discriminator's AUC approach 0.5 (i.e., unable to distinguish between the source and target domains), thus achieving the adversarial training objective and obtaining the carbon dioxide sealed storage layer physical property parameters of the target domain (i.e., the sealed storage site). The carbon dioxide sealed storage layer physical property parameters of the target domain (i.e., the sealed storage site) refer to a set of physical property parameter data containing some virtual data generated by using the generalization ability of the adversarial model to fill the gaps in the target domain.

[0098] Step S506: Construct and optimize the migration optimization model based on the multi-field coupling optimization algorithm: First, based on the reservoir physical property parameters of the target domain (i.e., the storage site) obtained in the above process (steps S501-S505), including thickness, porosity, permeability and sand-to-soil ratio, an initial migration optimization model is constructed in combination with the multi-field coupling optimization algorithm; then, the dynamic migration law is captured based on the unsteady flow model to optimize the multi-field coupling optimization algorithm, wherein optimizing the multi-field coupling optimization algorithm is equivalent to optimizing the migration optimization model.

[0099] Specifically, a multi-field coupling optimization algorithm is applied to simulate... The flow path and distribution range after injection, the physical field model type and main framework include: (1) thermal field (T), temperature influence (1) Density, viscosity, solubility; (2) Flow field (H): Darcy's law and equation of state for multiphase flow; (3) Mechanical field (M): effective stress change controls crack opening or rock deformation feedback permeability; (4) Chemical field: mineral reaction kinetics (e.g., calcite conversion rate ΔM / Δt).

[0100] Table 2 shows the mainstream simulators that can be used for carbon dioxide sequestration in depleted oil and gas fields. These simulators can be used not only for simulation... The flow path and distribution range after injection can also be used as a parameter solver for subsequent optimization algorithms.

[0101] Table 2: Comparison of Multiphysics Coupled Simulators for Geological Carbon Sequestration

[0102]

[0103] By combining an unsteady flow model to capture dynamic migration patterns, a multi-field fully coupled algorithm for THMC is optimized. The calculation formula for capturing dynamic migration patterns is as follows:

[0104]

[0105] in, express Saturation (in pores) (phase volume ratio) Represents a time variable; Relative penetration rate ( (Ratio of relative penetration to absolute penetration) Indicates pressure gradient (driving force) (The main cause of migration) Indicates the migration coefficient (characterizing storage efficiency). Represents the oil and gas migration coupling function (residual hydrocarbon pairs) (enhancing effect of transport); Mathematical operators representing partial differentials.

[0106] The above calculation formula for capturing dynamic migration patterns simulates and describes Saturation ( The non-steady-state variation of ) with time (t), where the main parameters are obtained or calibrated as follows:

[0107] (1) Output is obtained through well logging resistivity inversion, nuclear magnetic resonance (NMR), or numerical simulation;

[0108] (2) Calibration was performed using core displacement experiments (steady-state / non-steady-state method) and the Brooks-Corey model;

[0109] (3) Real-time monitoring via bottom hole pressure gauge + in-situ stress inversion (hydraulic fracturing method).

[0110] (4) The calculation formula is: Where Φ represents porosity, which is obtained through core mercury injection experiments or numerical simulation of well logging curves; The tortuosity factor is calculated using CT scans and a tortuosity algorithm. This represents the diffusion coefficient, obtained through high-temperature and high-pressure diffusion coefficient experiments.

[0111] (5) The function formula is: Among them, α (migration sensitivity coefficient) is obtained by core displacement experiments or by inversion based on historical production data (error <5%). (Residual oil and gas concentration gradient) is obtained by fitting residual saturation to historical production data, or by inverting the saturation distribution from well logging data (resistivity / NMR); β (density coupling coefficient) is obtained by testing fluid density using a PVT instrument (Pressure-Volume-Temperature Analyzer, a specialized experimental device for measuring the physical properties of substances such as oil, natural gas, and polymer materials under high pressure and variable temperature conditions), and corrected using the temperature-pressure formula; Δρ (density difference), i.e. The density difference was obtained through a PVT test experiment.

[0112] Step S104: Input the target geological carbon sequestration inference dataset of the research area of ​​the depleted oil and gas field in the sea into the migration optimization model for multi-field coupling simulation processing to obtain the target geological carbon sequestration evolution data.

[0113] In some embodiments, step S104 may include: inputting the target geological carbon sequestration inference dataset of the research area of ​​the depleted marine oil and gas field into the migration optimization model, and using a multi-field coupling optimization algorithm to perform multi-field coupling simulation processing on the target geological carbon sequestration inference dataset to obtain target geological carbon sequestration evolution data.

[0114] Among them, the research area of ​​depleted marine oil and gas fields refers to the area where the geological carbon sequestration potential needs to be assessed. Specifically, it is to determine whether the research area of ​​depleted marine oil and gas fields can be used for geological carbon sequestration based on the results of the geological carbon sequestration potential.

[0115] In the specific implementation, after the migration optimization model is constructed in step S104, the target geological carbon sequestration inference dataset of the research area of ​​the depleted marine oil and gas field is input into the migration optimization model. The multi-field coupling optimization algorithm is used to perform multi-field coupling simulation processing on the target geological carbon sequestration inference dataset to obtain the target geological carbon sequestration evolution data.

[0116] The target geological carbon sequestration evolution data includes solutions from multi-field coupled simulations. Flow field, chemical field, and stress field during geological storage process, dynamically adjusting porosity Permeability K, migration coefficient Parameters, and output distributed.

[0117] Step S105: Calculate the initial total geological carbon sequestration in the research area of ​​the depleted marine oil and gas field based on the target geological carbon sequestration evolution data.

[0118] Please see Figure 6 , Figure 6 yes Figure 1 In the flowchart of step S105, in some embodiments, step S105 may include, but is not limited to, steps S601 to S605:

[0119] Step S601: Calculate the reservoir structural parameters in the target geological carbon sequestration evolution data according to the structural sequestration calculation formula to obtain the structural sequestration capacity of the research area of ​​the depleted marine oil and gas field.

[0120] Step S602: Calculate the fluid dissolution and migration parameters in the target geological carbon sequestration evolution data according to the dissolution and sequestration calculation formula to obtain the dissolution and sequestration capacity of the research area of ​​the depleted marine oil and gas field.

[0121] Step S603: Calculate the multiphase seepage constraint parameters in the target geological carbon sequestration evolution data according to the residual sequestration calculation formula to obtain the residual sequestration capacity of the research area of ​​the depleted marine oil and gas field.

[0122] Step S604: Calculate the mineral transformation reaction parameters in the target geological carbon sequestration evolution data according to the mineralization sequestration calculation formula to obtain the mineralization sequestration capacity of the research area of ​​the depleted marine oil and gas field.

[0123] Step S605: Calculate the structural storage capacity, the dissolved storage capacity, the residual storage capacity, and the mineralized storage capacity according to the initial total storage formula to obtain the initial total geological carbon storage.

[0124] Alternatively, depleted oil and gas fields Geological carbon sequestration mainly involves four mechanisms: tectonic sequestration, dissolution sequestration, residual sequestration, and mineralization sequestration. Multi-field coupled simulation is used to solve the above equations (calculation formulas capturing dynamic migration patterns). Flow field, chemical field, and stress field during geological storage process, dynamically adjusting porosity Permeability K, migration coefficient Parameters, and output The distribution is used to calculate the contribution of the four types of sequestration mechanisms. Specifically, the four sequestration mechanisms and their calculation formulas are as follows:

[0125] (1) Construction of sealing: Under the protection of caprock, the gas is stored in a supercritical state within structural traps, similar to the hydrocarbon reservoir formation mechanism; the structural storage injection period is 0–30 years, with a storage contribution rate of 60%–80%. Specifically, the formula for calculating the structural storage capacity is as follows:

[0126]

[0127] in, Indicates the storage capacity; Indicates porosity; express Saturation; This indicates the reservoir volume.

[0128] (2) Dissolution and sealing: Dissolution occurs in formation water and is inhibited by salinity and temperature; the dissolution-sequestration injection period is 50–200 years, with a sequestration contribution rate of 5%–15%. Specifically, the formula for calculating the dissolution-sequestration capacity is as follows:

[0129]

[0130] in, Indicates the capacity for dissolution and sealing; Indicates the migration coefficient (characterizing storage efficiency). express Solubility, obtained using the Spycher model: ; Indicates porosity; The water saturation level is represented by the inversion of well logging data, indicating a depleted gas reservoir. ≈0.85; This represents a micro-element of the dissolved and sealed volume.

[0131] (3) Residual Sequestration: Captured in the pores by capillary forces, with residual gas saturation ( Related to this; the residual storage injection period is 10-50 years, and the storage contribution rate is 20-30%. Specifically, the calculation formula for the residual storage capacity is as follows:

[0132]

[0133] in, Indicates the remaining storage capacity; Indicates porosity; Residual gas saturation is indicated by core displacement-suction experiments, reflecting... The proportion of gas retained in the corners or blind ends of pores after displacement is typically 0.05-0.25 in depleted gas reservoirs. The reservoir volume is represented by the reservoir thickness and area; Indicates reservoir temperature and pressure conditions Density, calculated using the supercritical equation of state. And it was corrected using PVT experiments.

[0134] (4) Mineralization sequestration: It reacts with minerals to form carbonates (such as calcite), with a mineral sequestration period of >200 years and a sequestration contribution rate of <5%. Specifically, the formula for calculating the mineral sequestration capacity is as follows:

[0135]

[0136] in, Indicates the chemical sequestration capacity of mineral reactions; This indicates the increase in the mass of calcite, reflecting... The amount of carbonate minerals produced by the mineralization reaction is obtained through geochemical simulation or experimental determination. Calcium ( The molar volume of minerals, representing the space occupied by a unit mass of mineral (theoretical value ≈ 36.9). / mol).

[0137] This application embodiment also involves dynamically updating and calculating the total amount of sediment. When conditions such as water depth, pressure, and salinity change, or when results from seismic activity, well logging monitoring and inversion, physical property testing, and artificial catalyst acceleration indicate changes in porosity, permeability, etc. When parameters such as saturation and mineralization rate change, the changes will be fed back to the above series of equations (the calculation formulas for the four storage mechanisms) to dynamically update the storage capacity and calculate the total storage volume.

[0138]

[0139] in, express Total storage capacity; Indicates the storage capacity; Indicates the capacity for dissolution and sealing; Indicates the remaining storage capacity; This indicates the mineralization storage capacity.

[0140] Step S106: The initial total geological carbon sequestration is corrected for marine area to obtain the target total geological carbon sequestration; the target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted oil and gas field in the marine area.

[0141] In some embodiments, step S106 may include: adjusting the initial total geological carbon sequestration amount by marine area according to the marine area adjustment formula to obtain the target total geological carbon sequestration amount; wherein, the marine area adjustment formula is:

[0142]

[0143] in, It is expressed as the total target geological carbon sequestration (i.e., the sequestration capacity after marine area correction). This is expressed as the initial total geological carbon sequestration (i.e. ), Indicated as water depth, It is expressed as the sand content saturation.

[0144] In this embodiment of the application, the data obtained after correction using the marine area correction formula is... This refers to the geological carbon sequestration capacity of depleted oil and gas fields in the sea. The geological carbon sequestration capacity of depleted oil and gas fields in the sea is used to characterize the geological carbon sequestration potential of the research area of ​​depleted oil and gas fields in the sea.

[0145] Steps S101 to S106 of this embodiment involve obtaining an initial geological carbon sequestration training dataset for the experimental area of ​​a depleted offshore oil and gas field. This initial dataset includes source region oil and gas data, target region sequestration data, and engineering environment data. The initial dataset is preprocessed to obtain a target geological carbon sequestration training dataset. A migration optimization model is constructed based on this dataset. The target geological carbon sequestration inference dataset for the depleted offshore oil and gas field is input into the migration optimization model for multi-field coupled simulation processing to obtain target geological carbon sequestration evolution data. Based on this evolution data, the initial total geological carbon sequestration for the depleted offshore oil and gas field is calculated. The initial total geological carbon sequestration is then corrected for offshore conditions to obtain the target total geological carbon sequestration. This target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the depleted offshore oil and gas field. This application embodiment combines transfer learning technology with multi-source data fusion technology to transfer reservoir property data from the exploration stage of depleted marine oil and gas fields to a carbon sequestration capacity evaluation model. This can quantify the geological carbon sequestration capacity, thereby assessing the geological carbon sequestration potential of depleted marine oil and gas fields based on the geological carbon sequestration capacity. This can improve the reuse rate of oil and gas data and significantly reduce the amount of training data, thereby reducing the cost of evaluating the geological carbon sequestration potential of depleted marine oil and gas fields.

[0146] To explain in detail the principle of the technical solution of the present invention, the overall process of the present invention will be described below with reference to some specific embodiments. It is easy to understand that the following is an explanation of the technical principle of the present invention and should not be regarded as a limitation of the present invention.

[0147] Please see Figure 7 , Figure 7 This is a flowchart illustrating a method for evaluating geological carbon sequestration potential provided in an embodiment of this application. Figure 7 The overall implementation process of the geological carbon sequestration potential assessment method is demonstrated. Specifically, the overall implementation process of the geological carbon sequestration potential assessment method includes, but is not limited to, steps S701-S705:

[0148] Step S701: Integration and preprocessing of multi-source heterogeneous training data;

[0149] In practical implementation, the multi-source heterogeneous training data includes source domain data (oil and gas reservoirs), target domain data (storage sites), and engineering environmental data. The integration of source domain data (oil and gas reservoirs) mainly involves extracting reservoir physical property parameters (i.e., source domain oil and gas data) such as reservoir thickness, area, porosity (e.g., 12%-40%), permeability (e.g., 1-500 mD), and sand-soil ratio from depleted oil and gas fields to construct a pre-training dataset. Additionally, the integration of source domain data also includes organizing geophysical exploration data (seismic profiles, well logging curves) for the calculation and correction of reservoir physical property data, as well as the acquisition or correction of parameters related to subsequent storage capacity calculations. For the integration of target domain data (storage sites), the reservoir parameters of the deep saline aquifers in the main target marine area (thickness, area, salinity (e.g., 3-35 g / L), pressure gradient (e.g., 0.8-1.2 MPa), and geothermal gradient, etc.) are mainly collected. This involves integrating the target domain storage data and generating a virtual scenario covering the target domain at a water depth of n meters (e.g., 500-3000 meters) through Monte Carlo simulation for subsequent model adversarial training. For the integration of engineering environmental data, the main data collected include well network configuration, injection pressure history, oil production, caprock lithology, fault distribution, and groundwater salinity in depleted oil and gas fields. This data is used for establishing oil and gas migration coupling function models, dynamically updating carbon sequestration capacity, and calculating marine area corrections.

[0150] After integrating the data, based on the size and distribution characteristics of the collected data, the "maximum-minimum method" normalization formula is applied to calculate the data point by point for small datasets with large spans, while the Z-score normalization formula is used to calculate the data for large datasets with large spans such as penetration rate. Finally, the preprocessed multi-source heterogeneous training data is obtained.

[0151] Step S702: Construct a transfer optimization model based on the preprocessed multi-source heterogeneous training data;

[0152] In practice, the overall process of constructing the migration optimization model is as follows:

[0153] The first step is to construct adversarial features: the data set of oil and gas reservoirs (source domain) and storage sites (target domain) such as reservoir thickness, porosity and sand-soil ratio are added with "domain labels" (0 represents source domain and 1 represents target domain) and used as input features for adversarial training. The data set is then stratified and divided into training set / validation set / test set (the ratio of training set / validation set / test set is 70% / 15% / 15%, respectively).

[0154] The second step, combat training:

[0155] (1) A fixed generator is used to generate virtual target domain data by inputting source domain (oil and gas reservoir) parameters. The discriminator is trained to distinguish between real target domain data and generated virtual target domain data. By distinguishing between real target domain data (storage site, such as measured permeability of an oil field) and generated data (simulated permeability distribution), the generator is forced to learn the geological feature distribution of the target domain (storage site).

[0156] (2) Fix the discriminator and optimize the generator so that the discriminator cannot distinguish the data source. When the discriminator's probability of judging the generated data stabilizes at around 0.5 (cannot distinguish the source), and the distribution of geological parameters shows no significant difference in statistical tests, it is considered converged. In the specific implementation, the above steps are repeated until convergence. The generator adjusts the mapping of source domain parameters (such as sand-to-soil ratio) so that the distribution of generated data (virtual target domain sand-to-soil ratio) approximates the distribution of the real target domain until the discriminator cannot distinguish them. In the assessment of the storage potential of depleted oil and gas fields, the generator's task is to generate parameters (such as porosity, permeability, and sand-to-soil ratio) consistent with the geological distribution of the target domain.

[0157] The third step is adversarial validation: Distribution difference detection is performed. A binary classification model (LightGBM) is trained using core parameters such as porosity, sand-to-soil ratio, and permeability of the source and target reservoirs. This model predicts whether a sample belongs to the source or target domain, outputting the probability that the sample belongs to the target domain. The predicted probability for each sample, using the binary classification model as a discriminator, is the discrimination score. In the specific implementation, the probability value formula is used to output the results to identify and improve the generator's fake samples, while simultaneously optimizing the discriminator for subsequent model construction and computation.

[0158] The fourth step involves ranking the scores and traversing thresholds to calculate the TPR / FPR: All samples are sorted from highest to lowest according to their discriminant scores, generating a sorted list. Each discriminant score is used as a classification threshold to calculate the corresponding True Positive Rate (TPR) and False Positive Rate (FPR). Using the formulas for calculating TPR and FPR, a series of (FPR, TPR) points are generated. Each point corresponds to the classification performance at a given threshold. Connecting these points forms the ROC curve (Receiver Operating Characteristic Curve), and the point closest to the top left corner of the curve corresponds to the optimal classification threshold.

[0159] The fifth step is to calculate the AUC index: The area under the ROC curve, or AUC, is calculated using the trapezoidal rule or the Man-Whitney U statistic. Specifically, the discriminator is trained to calculate the AUC index. If AUC > 0.85, it indicates a significant difference in the distribution between the source and target domains, requiring parameter transfer. If AUC < 0.6, the distribution difference is considered to have significantly decreased. The generator of a Generative Adversarial Network (GAN) is used to adjust the parameter distribution of the source domain, making the discriminator's AUC approach 0.5 (i.e., unable to distinguish between the source and target domains). This achieves the adversarial training objective and obtains the physical properties of the carbon dioxide storage layer in the target domain (i.e., the storage site).

[0160] Step 6: Construct and optimize the migration optimization model based on the multi-field coupling optimization algorithm: First, based on the reservoir physical property parameters of the target domain (i.e., the storage site) obtained in the above process (steps S501-S505), including thickness, porosity, permeability and sand-to-soil ratio, an initial migration optimization model is constructed in combination with the multi-field coupling optimization algorithm; then, the dynamic migration law is captured based on the unsteady flow model to optimize the multi-field coupling optimization algorithm, wherein optimizing the multi-field coupling optimization algorithm is equivalent to optimizing the migration optimization model.

[0161] Specifically, a multi-field coupling optimization algorithm is applied to simulate... The flow path and distribution range after injection, the physical field model type and main framework include: (1) thermal field (T), temperature influence (1) Density, viscosity, solubility; (2) Flow field (H): Darcy's law and equation of state for multiphase flow; (3) Mechanical field (M): effective stress change controls crack opening or rock deformation feedback permeability; (4) Chemical field: mineral reaction kinetics (e.g., calcite conversion rate ΔM / Δt). Combine unsteady flow model to capture dynamic migration law and realize THMC multi-field fully coupled algorithm optimization.

[0162] Step S703: Input the multi-source heterogeneous inference data to be predicted into the migration optimization model to obtain geological carbon sequestration evolution data;

[0163] The multi-source heterogeneous inference data to be predicted is similar to the data collected in step S701 and the data preprocessing content, and will not be described in detail here in the embodiments of this application.

[0164] Specifically, the multi-source heterogeneous inference data to be predicted is input into the trained transfer optimization model to obtain geological carbon sequestration evolution data. The geological carbon sequestration evolution data can include, but is not limited to, multi-field coupled simulation solutions. Flow field, chemical field, and stress field during geological storage process, dynamically adjusting porosity Permeability K, migration coefficient Parameters, and output distributed.

[0165] Step S704: Quantitatively evaluate the geological carbon sequestration potential based on geological carbon sequestration evolution data;

[0166] Alternatively, depleted oil and gas fields Geological carbon sequestration mainly involves four mechanisms: tectonic sequestration, dissolution sequestration, residual sequestration, and mineralization sequestration. This is solved using multi-field coupled simulation. Flow field, chemical field, and stress field during geological storage process, dynamically adjusting porosity Permeability K, migration coefficient Parameters, and output Distribution, used to calculate the contribution of the four types of sealing mechanisms.

[0167] Step S705: Dynamically update and correct the archive capacity.

[0168] Please see Figure 8 , Figure 8 This is a schematic diagram illustrating a process for dynamically updating and correcting the archived capacity, as provided in an embodiment of this application. Figure 8 As shown, step S705 includes, but is not limited to, steps S801-S802:

[0169] Step S801: Dynamically update and calculate the total amount of data to be sealed;

[0170] Specifically, when conditions such as water depth, pressure, and salinity change, or when results from earthquakes, well logging monitoring and inversion, physical property testing, and artificial catalyst acceleration indicate changes in porosity, permeability, etc. When parameters such as saturation and mineralization rate change, the changes are fed back into the calculation formulas of the four storage mechanisms to dynamically update the storage capacity and calculate the initial total geological carbon storage.

[0171] Step S802: The marine area correction formula dynamically adjusts the storage capacity.

[0172] Specifically, the initial total geological carbon sequestration is corrected for marine area based on the marine area correction formula to obtain the target total geological carbon sequestration. The target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted oil and gas field in the marine area.

[0173] It should be noted that this embodiment is only a brief illustrative description of the overall process of the geological carbon sequestration potential evaluation method. Detailed descriptions of each step can be found in the relevant content of the foregoing embodiments, and will not be repeated here. It is understood that the present invention does not limit this.

[0174] This application embodiment obtains an initial geological carbon sequestration training dataset for an experimental area of ​​a depleted offshore oil and gas field. The initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environment data. The initial geological carbon sequestration training dataset is preprocessed to obtain a target geological carbon sequestration training dataset. A migration optimization model is constructed based on the target geological carbon sequestration training dataset. The target geological carbon sequestration inference dataset for the research area of ​​the depleted offshore oil and gas field is input into the migration optimization model for multi-field coupled simulation processing to obtain target geological carbon sequestration evolution data. Based on the target geological carbon sequestration evolution data, the initial total geological carbon sequestration for the research area of ​​the depleted offshore oil and gas field is calculated. The initial total geological carbon sequestration is corrected for offshore conditions to obtain the target total geological carbon sequestration. The target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted offshore oil and gas field. This application embodiment combines transfer learning technology with multi-source data fusion technology to transfer reservoir property data from the exploration stage of depleted marine oil and gas fields to a carbon sequestration capacity evaluation model. This can quantify the geological carbon sequestration capacity, thereby assessing the geological carbon sequestration potential of depleted marine oil and gas fields based on the geological carbon sequestration capacity. This can improve the reuse rate of oil and gas data and significantly reduce the amount of training data, thereby reducing the cost of evaluating the geological carbon sequestration potential of depleted marine oil and gas fields.

[0175] In summary, the key advantages of the geological carbon sequestration potential evaluation method provided in this application are as follows:

[0176] (1) By using the transfer learning framework, reservoir physical parameters (porosity, permeability, sand-soil ratio, etc.) from the oil and gas field exploration stage are transferred to the carbon sequestration potential evaluation model, reusing oil and gas data and reducing evaluation costs. Key points include the dynamic weight adjustment mechanism of transfer learning, the construction of the transfer optimization model, and the adversarial training optimization process.

[0177] For example, relevant geological carbon sequestration potential assessment methods (such as traditional numerical simulation methods) usually require the independent collection of geological parameters (porosity, permeability, etc.) required for sequestration assessment. However, the embodiments of this application use a transfer learning framework to transfer reservoir physical property data (such as porosity, permeability, pressure gradient) from the exploration stage of depleted oil and gas fields to the carbon sequestration assessment model, which can improve the reuse rate of oil and gas data and significantly reduce the need for training data.

[0178] (2) Solving based on multi-field coupling simulation Flow field, chemical field, and stress field during geological storage process, dynamically adjusting porosity Permeability K, migration coefficient Parameters; output simultaneously Distribution is used to calculate the capacity of the four types of sealing mechanisms.

[0179] (3) When natural conditions or artificial causes cause changes in water depth, pressure, salinity, reservoir properties, mineralization rate, etc., it can provide feedback and dynamically update the storage capacity, and can also correct the storage capacity of depleted oil and gas fields in the sea area.

[0180] It is worth mentioning that (1) the evaluation method of establishing a full-parameter geological model based on reservoir numerical simulation software (such as ECLIPSE, COMSOL) and predicting the amount of reserves through historical data matching and parameter inversion is mainly applicable to depleted oil and gas fields with simple geological conditions and complete data (such as single sandstone reservoirs), but has poor adaptability to heterogeneous reservoirs (such as fractured limestone). (2) the evaluation method of building a reserve prediction model based on artificial intelligence optimization model, such as using supervised learning (ANN (Artificial Neural Network), SVM (Support Vector Machine)) or reinforcement learning, and relying on historical datasets for training, is applicable to small and medium-sized reserve projects (such as reserves <1Gt), but has insufficient predictive reliability for complex geological conditions (such as fault-developed areas).

[0181] However, this technical solution, by combining transfer learning and multi-field coupled optimization algorithms, correlates oil and gas reservoir physical parameters (reservoir thickness, porosity, sand-to-soil ratio) with the storage capacity calculation model. This allows for the quantification of geological storage capacity, leakage risk, and economic indicators, providing a basis for depleted oil and gas reservoirs. Geological carbon sequestration provides technical support and decision support for oil and gas field transformation. Therefore, the geological carbon sequestration potential evaluation method provided in this application significantly outperforms other related schemes in terms of data reusability, dynamic optimization capability, and economy, and is particularly suitable for complex scenarios such as heterogeneous reservoirs and fault-developed areas. Furthermore, to further reduce initial investment, it can be combined with… Oil displacement synergistic technology is used for processing; for more refined scientific research-level analysis, multi-field coupled simulation verification can be added. It is understood that the geological carbon sequestration potential evaluation method provided in this application embodiment is applicable to various scenarios. The above is merely a simple example, and the embodiments of this application will not be elaborated upon here. Furthermore, those skilled in the art can select corresponding auxiliary methods (such as the combination of examples) according to the actual scenario. The embodiments of this application do not limit the scope of optimization (such as oil displacement synergy technology).

[0182] Please see Figure 9 This application also provides a geological carbon sequestration potential evaluation device 900, which can implement the above-mentioned method. The device includes the following modules:

[0183] The initial training data acquisition module 901 is used to acquire the initial geological carbon sequestration training dataset of the experimental area of ​​the depleted marine oil and gas field; the initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data and engineering environment data.

[0184] The geological carbon sequestration data preprocessing module 902 is used to preprocess the initial geological carbon sequestration training dataset to obtain the target geological carbon sequestration training dataset.

[0185] The migration optimization model construction module 903 is used to construct a migration optimization model based on the target geological carbon sequestration training dataset.

[0186] The multi-field coupled simulation processing module 904 is used to input the target geological carbon sequestration inference dataset of the research area of ​​the depleted oil and gas field in the sea into the migration optimization model for multi-field coupled simulation processing to obtain the target geological carbon sequestration evolution data.

[0187] The initial geological carbon sequestration total calculation module 905 is used to calculate the initial geological carbon sequestration total of the research area of ​​the depleted marine oil and gas field based on the target geological carbon sequestration evolution data.

[0188] The target geological carbon sequestration total calculation module 906 is used to perform marine area correction on the initial geological carbon sequestration total to obtain the target geological carbon sequestration total; the target geological carbon sequestration total is used to characterize the geological carbon sequestration potential of the marine depleted oil and gas field research area.

[0189] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0190] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0191] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0192] Please see Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0193] The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0194] The memory 1002 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1002 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called and executed by the processor 1001 using the methods described in the embodiments of this application.

[0195] Input / output interface 1003 is used to implement information input and output;

[0196] The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0197] Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004);

[0198] The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0199] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0200] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0201] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0202] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0203] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0204] The geological carbon sequestration potential evaluation method, apparatus, equipment, medium, and program products provided in this application embodiment acquire an initial geological carbon sequestration training dataset of an experimental area of ​​a depleted offshore oil and gas field. The initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environmental data. The initial geological carbon sequestration training dataset is preprocessed to obtain a target geological carbon sequestration training dataset. A migration optimization model is constructed based on the target geological carbon sequestration training dataset. The target geological carbon sequestration inference dataset of the depleted offshore oil and gas field is input into the migration optimization model for multi-field coupled simulation processing to obtain target geological carbon sequestration evolution data. Based on the target geological carbon sequestration evolution data, the initial total geological carbon sequestration of the depleted offshore oil and gas field is calculated. The initial total geological carbon sequestration is corrected for offshore area conditions to obtain the target total geological carbon sequestration. The target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the depleted offshore oil and gas field. This application embodiment combines transfer learning technology with multi-source data fusion technology to transfer reservoir property data from the exploration stage of depleted marine oil and gas fields to a carbon sequestration capacity evaluation model. This can quantify the geological carbon sequestration capacity, thereby assessing the geological carbon sequestration potential of depleted marine oil and gas fields based on the geological carbon sequestration capacity. This can improve the reuse rate of oil and gas data and significantly reduce the amount of training data, thereby reducing the cost of evaluating the geological carbon sequestration potential of depleted marine oil and gas fields.

[0205] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0206] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0207] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0208] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0209] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0210] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0211] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0212] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0213] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0214] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0215] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for evaluating geological carbon sequestration potential, characterized in that, The method includes the following steps: An initial geological carbon sequestration training dataset was obtained for the experimental area of ​​a depleted marine oil and gas field. The initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data, and engineering environmental data. The initial geological carbon sequestration training dataset is preprocessed to obtain the target geological carbon sequestration training dataset; Based on the target geological carbon sequestration training dataset, a transfer optimization model is constructed; The target geological carbon sequestration inference dataset of the research area of ​​depleted oil and gas fields in the sea is input into the migration optimization model for multi-field coupled simulation processing to obtain the target geological carbon sequestration evolution data. Based on the target geological carbon sequestration evolution data, calculate the initial total geological carbon sequestration in the study area of ​​the depleted marine oil and gas field; The initial total geological carbon sequestration is corrected for marine area to obtain the target total geological carbon sequestration; the target total geological carbon sequestration is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted oil and gas field in the marine area. The step of constructing a transfer optimization model based on the target geological carbon sequestration training dataset includes: The target geological carbon sequestration training dataset is labeled to obtain a cross-domain labeled dataset; The cross-domain label dataset is stratified and sampled to obtain a cross-domain adversarial training dataset; wherein the cross-domain adversarial training dataset includes a training set, a validation set, and a test set; Based on the training set and the validation set, the initial generative adversarial network is subjected to adversarial training. When the distribution difference between the generated data of the initial generative adversarial network and the target domain data reaches a preset convergence threshold, the training of the initial generative adversarial network is terminated, and the target generative adversarial network is obtained. Based on the training set and the validation set, the target generative adversarial network is subjected to adversarial verification to optimize the target generative adversarial network, and a binary classification model is trained based on the training set and the validation set. The trained binary classification model is used as the target discriminator of the optimized target generative adversarial network. The test set is input into the optimized target generative adversarial network, the target discriminator outputs the target domain probability, and the optimized target generative adversarial network is judged based on the target domain probability to determine whether it meets the adversarial training requirements. If the optimized target generative adversarial network meets the adversarial training requirements, then the target domain carbon-sealed storage layer physical property parameters generated by the target generator in the target generative adversarial network are obtained. The migration optimization model is constructed using a multi-field coupling optimization algorithm based on the physical property parameters of the target domain carbon-sealed storage layer.

2. The method according to claim 1, characterized in that, The preprocessing of the initial geological carbon sequestration training dataset to obtain the target geological carbon sequestration training dataset includes: According to the normalization formula, the first initial geological carbon sequestration training data that is smaller than the preset dimensional span threshold in the initial geological carbon sequestration training dataset is normalized to obtain the first target geological carbon sequestration training data. According to the standardization formula, the second initial geological carbon sequestration training data that is larger than the preset dimensional span threshold in the initial geological carbon sequestration training dataset is standardized to obtain the second target geological carbon sequestration training data. The target geological carbon sequestration training dataset is constructed based on the first target geological carbon sequestration training data and the second target geological carbon sequestration training data.

3. The method according to claim 1, characterized in that, The target geological carbon sequestration inference dataset of the research area of ​​depleted marine oil and gas fields is input into the migration optimization model for multi-field coupled simulation processing to obtain target geological carbon sequestration evolution data, including: The target geological carbon sequestration inference dataset of the research area of ​​the depleted oil and gas field in the sea area is input into the migration optimization model, and the target geological carbon sequestration inference dataset is processed by multi-field coupling optimization algorithm to obtain the target geological carbon sequestration evolution data.

4. The method according to claim 1, characterized in that, The initial total geological carbon sequestration includes tectonic sequestration capacity, dissolution sequestration capacity, residual sequestration capacity, and mineralization sequestration capacity. The calculation of the initial total geological carbon sequestration in the research area of ​​the depleted marine oil and gas field based on the target geological carbon sequestration evolution data includes: The reservoir structural parameters in the target geological carbon sequestration evolution data are calculated according to the structural sequestration calculation formula to obtain the structural sequestration capacity of the research area of ​​the depleted marine oil and gas field. The fluid dissolution and migration parameters in the target geological carbon sequestration evolution data are calculated based on the dissolution and sequestration calculation formula to obtain the dissolution and sequestration capacity of the research area of ​​the depleted oil and gas field in the sea area. The residual storage capacity of the research area of ​​the depleted marine oil and gas field is obtained by calculating the multiphase seepage constraint parameters in the target geological carbon storage evolution data according to the residual storage calculation formula. The mineral transformation reaction parameters in the target geological carbon sequestration evolution data are calculated based on the mineralization sequestration calculation formula to obtain the mineralization sequestration capacity of the research area of ​​the depleted oil and gas field in the marine area. The initial total geological carbon sequestration is obtained by calculating the tectonic sequestration capacity, the dissolution sequestration capacity, the residual sequestration capacity, and the mineralization sequestration capacity according to the initial total sequestration formula.

5. The method according to claim 1, characterized in that, The process of correcting the initial total geological carbon sequestration to obtain the target total geological carbon sequestration includes: The initial total geological carbon sequestration is adjusted for marine area using a marine area adjustment formula to obtain the target total geological carbon sequestration; wherein, the marine area adjustment formula is: in, This represents the total amount of target geological carbon sequestration. This represents the total initial geological carbon sequestration. Indicated as water depth, It is expressed as the sand content saturation.

6. A geological carbon sequestration potential evaluation device, characterized in that, The device includes the following modules: The initial training data acquisition module is used to acquire the initial geological carbon sequestration training dataset of the experimental area of ​​the depleted marine oil and gas field; the initial geological carbon sequestration training dataset includes source domain oil and gas data, target domain sequestration data and engineering environment data; The geological carbon sequestration data preprocessing module is used to preprocess the initial geological carbon sequestration training dataset to obtain the target geological carbon sequestration training dataset. The migration optimization model construction module is used to construct a migration optimization model based on the target geological carbon sequestration training dataset. The multi-field coupled simulation processing module is used to input the target geological carbon sequestration inference dataset of the research area of ​​the depleted oil and gas field in the sea into the migration optimization model for multi-field coupled simulation processing to obtain the target geological carbon sequestration evolution data. The initial geological carbon sequestration total calculation module is used to calculate the initial geological carbon sequestration total of the research area of ​​the depleted marine oil and gas field based on the target geological carbon sequestration evolution data. The target geological carbon sequestration total calculation module is used to perform marine area correction on the initial geological carbon sequestration total to obtain the target geological carbon sequestration total; the target geological carbon sequestration total is used to characterize the geological carbon sequestration potential of the research area of ​​the depleted oil and gas field in the marine area; The migration optimization model construction module is specifically used for: The target geological carbon sequestration training dataset is labeled to obtain a cross-domain labeled dataset; The cross-domain label dataset is stratified and sampled to obtain a cross-domain adversarial training dataset; wherein the cross-domain adversarial training dataset includes a training set, a validation set, and a test set; Based on the training set and the validation set, the initial generative adversarial network is subjected to adversarial training. When the distribution difference between the generated data of the initial generative adversarial network and the target domain data reaches a preset convergence threshold, the training of the initial generative adversarial network is terminated, and the target generative adversarial network is obtained. Based on the training set and the validation set, the target generative adversarial network is subjected to adversarial verification to optimize the target generative adversarial network, and a binary classification model is trained based on the training set and the validation set. The trained binary classification model is used as the target discriminator of the optimized target generative adversarial network. The test set is input into the optimized target generative adversarial network, the target discriminator outputs the target domain probability, and the optimized target generative adversarial network is judged based on the target domain probability to determine whether it meets the adversarial training requirements. If the optimized target generative adversarial network meets the adversarial training requirements, then the target domain carbon-sealed storage layer physical property parameters generated by the target generator in the target generative adversarial network are obtained. The migration optimization model is constructed using a multi-field coupling optimization algorithm based on the physical property parameters of the target domain carbon-sealed storage layer.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 5.

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

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 5.

Images