Carbon dioxide burying rate prediction model generation and prediction method, device and equipment

Abstract

The invention provides a carbon dioxide burying rate prediction model generation and prediction method, device and equipment, and relates to the technical field of geological sequestration, and the prediction model generation method comprises the steps: generating an initial sample data set; determining a target influence parameter from the plurality of geological parameters according to the initial sample data set; based on the target influence parameter, performing data removal on the initial sample data set, and determining a target sample data set; establishing an initial carbon dioxide embedding rate prediction model; training the initial carbon dioxide embedding rate prediction model based on the target sample data set to obtain a target carbon dioxide embedding rate prediction model meeting a preset model condition; wherein the preset model condition comprises at least one of a first constraint condition corresponding to the decision coefficient and a second constraint condition corresponding to the learning curve. According to the invention, the problem of low prediction accuracy of an existing carbon dioxide embedding rate prediction method can be effectively solved.

Description

Technical Field

[0001] This application relates to the field of geological storage technology, specifically to a method and apparatus for generating a carbon dioxide burial rate prediction model, an electronic device, and a machine-readable storage medium. Background Technology

[0002] Geological carbon dioxide storage is an important means of reducing greenhouse gas emissions. Currently, research on carbon dioxide storage rates mainly focuses on two methods: indoor physical simulation experiments and numerical simulations. However, geological storage is influenced by numerous factors, which vary significantly in their degree and patterns of influence, making indoor experiments costly and difficult. Meanwhile, numerical simulations are complex, and their results are affected by the accuracy of basic experimental parameters, resulting in significant limitations. Furthermore, it is difficult to establish reasonable models in complex geological environments.

[0003] Among related technologies, improving the accuracy of carbon dioxide burial rate prediction is an urgent problem to be solved in order to ensure the long-term safety of carbon dioxide burial, prevent environmental risks such as leakage, and protect the safety of ecosystems and humans. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, and equipment for generating and predicting a carbon dioxide burial rate prediction model, in order to solve the problem of low prediction accuracy in existing carbon dioxide burial rate prediction methods.

[0005] To achieve the above objectives, the first aspect of this application provides a method for generating a carbon dioxide burial rate prediction model, the method comprising: Generate an initial sample dataset; wherein the initial sample dataset includes multiple initial sample data subsets, each initial sample data subset including carbon dioxide burial-related values, burial rate values, and parameter values ​​of multiple geological parameters; Based on the initial sample dataset, target influence parameters are determined from multiple geological parameters; Based on the target impact parameters, data is removed from the initial sample dataset to determine the target sample dataset; wherein, the target sample dataset includes multiple target sample data subsets, and each target sample data subset includes carbon dioxide storage-related values, storage rate values, and parameter values ​​of the target impact parameters; Establish an initial carbon dioxide burial rate prediction model; The initial carbon dioxide burial rate prediction model is trained based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets the preset model conditions; wherein, the preset model conditions include at least one of the first constraint condition corresponding to the coefficient of determination and the second constraint condition corresponding to the learning curve.

[0006] In this embodiment of the application, the target influence parameter is determined from multiple geological parameters based on the initial sample dataset, including: Based on the initial sample dataset, generate a set of parameter values ​​corresponding to each geological parameter. For each geological parameter, based on the set of parameter values ​​corresponding to the geological parameter and the burial rate values ​​in each initial sample data subset, the correlation value corresponding to the geological parameter under each preset correlation coefficient algorithm is determined, and based on each correlation value, the comprehensive evaluation index value corresponding to the geological parameter is determined. The target influencing parameters are determined based on the comprehensive evaluation index values ​​corresponding to each geological parameter.

[0007] In this embodiment of the application, the initial carbon dioxide burial rate prediction model is trained based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets preset model conditions, including: Determine the kernel function of the initial carbon dioxide burial rate prediction model; Optimize the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained; Based on the target sample dataset, the target model to be trained is trained to obtain the undetermined model; Determine whether the undetermined model meets the preset model conditions. If yes, then determine the undetermined model as the target carbon dioxide burial rate prediction model. If no, return to the step of optimizing the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained, until the undetermined model meets the preset model conditions.

[0008] In this embodiment of the application, determining the kernel function of the initial carbon dioxide burial rate prediction model includes: For each preset kernel function, the model to be trained corresponding to the preset kernel function is determined based on the initial carbon dioxide storage rate prediction model. Based on the target sample dataset, determine the prediction accuracy value of each model to be trained; The kernel function corresponding to the training model with the highest prediction accuracy value among multiple training models is determined as the kernel function of the initial carbon dioxide storage rate prediction model. Optimizing the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained includes: The model parameters in the initial carbon dioxide burial rate prediction model are optimized using a grid search method to obtain the target model to be trained.

[0009] In this embodiment of the application, determining whether the model to be determined satisfies the preset model conditions includes: Based on the target sample dataset, a test dataset is determined; wherein the test dataset includes multiple subsets of the target sample data; For each target sample data subset in the test dataset, the carbon dioxide storage-related values ​​and target impact parameter values ​​in the target sample data subset are input into the undetermined model, and the undetermined model outputs the predicted value corresponding to the storage rate value in the target sample data subset. Based on each embedded rate value and its corresponding predicted value in the test dataset, determine the determination coefficient and learning curve of the model to be determined; If the determination coefficient of the undetermined model satisfies the first constraint condition and / or the learning curve satisfies the second constraint condition, then the undetermined model is determined to satisfy the preset model condition.

[0010] A second aspect of this application provides a method for predicting carbon dioxide burial rates, the method comprising: Obtain the target dataset to be predicted; wherein, the target dataset to be predicted includes carbon dioxide storage-related values ​​and parameter values ​​of target impact parameters; The target dataset to be predicted is input into the target carbon dioxide burial rate prediction model generated by the method described in the first aspect above, and the burial rate value corresponding to the target dataset to be predicted is obtained.

[0011] A third aspect of this application provides a device for generating a carbon dioxide burial rate prediction model, the device comprising: The first dataset generation module is used to generate an initial sample dataset; wherein, the initial sample dataset includes multiple initial sample data subsets, and each initial sample data subset includes carbon dioxide burial-related values, burial rate values, and parameter values ​​of multiple geological parameters; The influence parameter determination module is used to determine the target influence parameter from multiple geological parameters based on the initial sample dataset; The second dataset generation module is used to remove data from the initial sample dataset based on the target impact parameters to determine the target sample dataset; wherein, the target sample dataset includes multiple target sample data subsets, and each target sample data subset includes carbon dioxide storage-related values, storage rate values ​​and parameter values ​​of the target impact parameters; The model building module is used to build an initial carbon dioxide burial rate prediction model; The model generation module is used to train the initial carbon dioxide burial rate prediction model based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets preset model conditions; wherein, the preset model conditions include at least one of a first constraint condition corresponding to the coefficient of determination and a second constraint condition corresponding to the learning curve.

[0012] A fourth aspect of this application provides a carbon dioxide burial rate prediction device, the device comprising: The data acquisition module is used to acquire the target dataset to be predicted; wherein, the target dataset to be predicted includes carbon dioxide storage-related values ​​and parameter values ​​of target impact parameters; The burial rate prediction module is used to input the target dataset to be predicted into the target carbon dioxide burial rate prediction model generated by the method described in the first aspect above, and obtain the burial rate value corresponding to the target dataset to be predicted.

[0013] The fifth aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the carbon dioxide burial rate prediction model generation method described in the first aspect and / or the carbon dioxide burial rate prediction method described in the second aspect.

[0014] The sixth aspect of this application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the carbon dioxide burial rate prediction model generation method described in the first aspect and / or the carbon dioxide burial rate prediction method described in the second aspect.

[0015] The carbon dioxide burial rate prediction model generation and prediction method, apparatus and equipment provided in this application firstly determine the target influencing parameters that have a significant impact on the carbon dioxide burial rate from a large number of geological parameters based on an initial sample dataset, thereby generating a target sample dataset. Then, a model is constructed and trained based on the target sample dataset, and finally a target carbon dioxide burial rate prediction model that meets the preset model conditions is obtained, ensuring that the target carbon dioxide burial rate prediction model has high prediction accuracy and reliability.

[0016] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 The schematic diagram illustrates the process of generating a carbon dioxide burial rate prediction model according to an embodiment of this application. Figure 2 The schematic diagram illustrates a flow chart of the carbon dioxide burial rate prediction method according to an embodiment of this application; Figure 3 This schematic diagram illustrates the structural block diagram of a carbon dioxide burial rate prediction model generation device according to an embodiment of this application. Figure 4 This schematic diagram illustrates the structural block diagram of a carbon dioxide burial rate prediction device according to an embodiment of this application; Figure 5 The diagram illustrates the internal structure of a computer device according to an embodiment of this application.

[0018] Explanation of reference numerals in the attached figures A01 - Processor; A02 - Network Interface; A03 - Internal Memory; A04 - Display Screen; A05 - Input Device; A06 - Non-volatile Storage Media; B01 - Operating System; B02 - Computer Program. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0020] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0021] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0022] In view of the low prediction accuracy of carbon dioxide burial rate prediction methods in related technologies, this application provides a carbon dioxide burial rate prediction model generation and prediction method, apparatus, equipment and medium. The following, in conjunction with the accompanying drawings, provides a detailed description of the carbon dioxide burial rate prediction model generation and prediction method, apparatus, equipment and medium provided in this application through specific embodiments and implementation methods.

[0023] Figure 1 The illustration shows a schematic flowchart of the carbon dioxide burial rate prediction model generation method according to an embodiment of this application. Figure 1 As shown in one embodiment of this application, a method for generating a carbon dioxide burial rate prediction model is provided, the method including the following steps.

[0024] Step A200: Generate the initial sample dataset.

[0025] The initial sample dataset includes multiple initial sample data subsets, each of which includes carbon dioxide burial-related values, burial rate values, and parameter values ​​of multiple geological parameters.

[0026] In this embodiment of the application, before step A200, the method may further include the following steps in order to generate the initial sample dataset.

[0027] Step A110: Obtain the original sample dataset.

[0028] The original sample dataset includes multiple original sample data subsets, each of which may include carbon dioxide burial-related values, burial rate values, and parameter values ​​of multiple geological parameters.

[0029] Specifically, carbon dioxide storage-related values ​​refer to the values ​​of relevant storage parameters during carbon dioxide storage. These parameters may include the total amount of carbon dioxide injected, the time of carbon dioxide injection, and the carbon dioxide injection rate. The carbon dioxide injection rate is equal to the total amount of carbon dioxide injected divided by the time of carbon dioxide injection. The storage rate value refers to the numerical value of the carbon dioxide storage rate. Geological parameters are used to describe the underground rock and soil mass and its characteristics. These parameters may include formation pressure, formation temperature, depth from the ground surface, etc.

[0030] In this embodiment of the application, the original sample dataset can be obtained by acquiring relevant data such as carbon dioxide sequestration geological projects. This application does not specifically limit the method of acquiring the original sample dataset.

[0031] Step A120: Filter the original sample dataset to obtain the dataset to be processed.

[0032] The dataset to be processed includes multiple subsets of data to be processed, each subset including carbon dioxide burial-related values, burial rate values, and parameter values ​​of multiple geological parameters.

[0033] Specifically, when a subset of the original sample data contains missing data (i.e., lacks necessary values ​​related to carbon dioxide burial, burial rate, or geological parameters), this subset of the original sample data will not be selected as the subset of data to be processed; that is, the initial subset of sample data will not be generated based on this subset of the original sample data. Furthermore, when an extreme result exists in the original subset of sample data due to inaccurate statistical data, this subset of the original sample data will also not be selected as the subset of data to be processed.

[0034] This application embodiment filters the original sample dataset after obtaining it to avoid affecting the accuracy and generalization ability of the model due to the source data.

[0035] It is worth mentioning that, in the embodiments of this application, the carbon dioxide storage-related values ​​in the data subset to be processed include the value of carbon dioxide injection rate, and / or the value of total carbon dioxide injection amount and carbon dioxide injection time, and each data subset to be processed includes the value of the same storage-related parameters, so as to ensure that the storage rate value can be predicted by the model in the future.

[0036] Step A130: Perform data normalization on each subset of data to be processed to obtain the initial sample dataset.

[0037] Specifically, data normalization refers to converting data with different dimensions and units into a unified standard format. In this application embodiment, data normalization is performed to improve the efficiency and effectiveness of subsequent determination of target influence parameters and model generation.

[0038] Step A300: Based on the initial sample dataset, determine the target influence parameter from multiple geological parameters.

[0039] In this embodiment of the application, step A300 may include the following steps.

[0040] Step A310: Based on the initial sample dataset, generate a set of parameter values ​​corresponding to each geological parameter.

[0041] The set of parameter values ​​corresponding to the geological parameters includes the parameter values ​​of the geological parameters in each subset of the initial sample data.

[0042] Step A320: For each geological parameter, based on the set of parameter values ​​corresponding to the geological parameter and the burial rate values ​​in each initial sample data subset, determine the correlation value corresponding to the geological parameter under each preset correlation coefficient algorithm, and determine the comprehensive evaluation index value corresponding to the geological parameter based on each correlation value.

[0043] Specifically, in the set of parameter values ​​corresponding to geological parameters, each parameter value has its corresponding burial rate value, which is the burial rate value in the initial sample data subset to which it belongs.

[0044] In this application embodiment, the preset correlation coefficient algorithms specifically include Spearman's rank correlation coefficient (Spearman), Pearson correlation coefficient (Pearson), Kendall's correlation coefficient (Kendall), and Grey Relational Analysis (Grey). Each preset correlation coefficient algorithm is used to measure the relationship between geological parameters and burial rate. The magnitude of the correlation value calculated by each algorithm represents the degree of influence of geological parameters on carbon dioxide burial rate under the corresponding preset correlation coefficient algorithm.

[0045] Since determining the correlation value of a geological parameter under various preset correlation coefficient algorithms is an existing method given the known set of parameter values ​​corresponding to the geological parameter and the burial rate values ​​in each initial sample data subset, it will not be elaborated further.

[0046] For each geological parameter, after determining the correlation value corresponding to the geological parameter under various preset correlation coefficient algorithms, this application embodiment uses a specific method to determine the comprehensive evaluation index value corresponding to the geological parameter based on each correlation value: summing the correlation values ​​and taking the average value, yields the comprehensive evaluation index value corresponding to the geological parameter. The magnitude of the comprehensive evaluation index value corresponding to the geological parameter reflects the overall influence of the geological parameter on the carbon dioxide burial rate; that is, the larger the comprehensive evaluation index value corresponding to the geological parameter, the greater the influence of the geological parameter on the carbon dioxide burial rate.

[0047] Step A330: Determine the target influencing parameters based on the comprehensive evaluation index values ​​corresponding to each geological parameter.

[0048] In this embodiment of the application, after obtaining the comprehensive evaluation index values ​​corresponding to each geological parameter, step A330 sorts the comprehensive evaluation index values ​​in descending order, and determines the geological parameters corresponding to the first N comprehensive evaluation index values ​​as target influence parameters. Here, N is a positive integer, and its value can be specifically set according to the actual situation. For example, N can be selected as 7, meaning the number of target influence parameters is 7.

[0049] This application embodiment uses four analysis methods—Spearman correlation coefficient, Pearson correlation coefficient, Kendall correlation coefficient, and grey relational analysis—to analyze various geological parameters, determine the comprehensive evaluation index value of each geological parameter, and screen out target influencing parameters. This can reduce the amount of calculation in the subsequent modeling process, reduce the difficulty of model optimization, and enhance the model's predictive ability.

[0050] Step A400: Based on the target influence parameters, perform data removal on the initial sample dataset to determine the target sample dataset.

[0051] The target sample dataset includes multiple target sample data subsets, each of which includes carbon dioxide burial-related values, burial rate values, and target influence parameter values.

[0052] Specifically, the initial sample data subset contains numerous data points, and each initial sample data subset includes values ​​of geological parameters other than the target influence parameter. In this embodiment, data removal of the initial sample dataset refers to deleting all data from each initial sample data subset except for the carbon dioxide burial-related values, burial rate values, and the parameter values ​​of the target influence parameter, to obtain the target sample data subset corresponding to that initial sample data subset.

[0053] This application embodiment removes data from the initial sample dataset to reduce the adverse effects of other geological parameters that have little or no impact on the burial rate value on the model's prediction performance during subsequent model training.

[0054] Step A500: Establish an initial carbon dioxide burial rate prediction model.

[0055] In this embodiment of the application, the initial carbon dioxide burial rate prediction model adopts the Support Vector Machine (SVM) model.

[0056] Specifically, an SVM model is established, and through support vector regression, a function that approximates linear regression in the high-dimensional case is created. The expression of this function can be represented as:

[0057] in, Indicates the predicted value. Represents a weight vector. Represents the input vector. Let represent the input vector after the high-dimensional mapping, and b represent the bias value.

[0058] This application embodiment selects a kernel function for support vector regression, transforms it to a high-dimensional space using the kernel function, and uses a linear expression in the high-dimensional space to predict the burial rate value.

[0059] Of course, it is understood that the initial carbon dioxide burial rate prediction model can also employ other existing machine learning models, and this application does not specifically limit the selection of the initial carbon dioxide burial rate prediction model.

[0060] The embodiments of this application establish an initial carbon dioxide burial rate prediction model based on machine learning, and then train and optimize the model to facilitate rapid prediction of carbon dioxide burial rate based on the final model, with high prediction accuracy.

[0061] Step A600: Train the initial carbon dioxide burial rate prediction model based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets the preset model conditions.

[0062] The preset model conditions include at least one of the first constraint corresponding to the coefficient of determination and the second constraint corresponding to the learning curve.

[0063] In this embodiment of the application, step A600 may include the following steps.

[0064] Step A610: Determine the kernel function of the initial carbon dioxide burial rate prediction model.

[0065] In one specific embodiment, step A610 includes the following steps: Step A611: For each preset kernel function, determine the training model corresponding to the preset kernel function based on the initial carbon dioxide storage rate prediction model.

[0066] Specifically, preset kernel functions include linear kernel functions, polynomial kernel functions, Gaussian kernel functions, and so on. This specific implementation method allows for the creation of different trainable models in machine learning software by using different preset kernel functions.

[0067] Step A612: Determine the prediction accuracy value of each model to be trained based on the target sample dataset.

[0068] Specifically, for each model to be trained, the carbon dioxide storage-related values ​​and target influence parameters from each subset of the target sample data are sequentially used as inputs. Based on the outputs of the model (i.e., the predicted carbon dioxide storage rate) and the storage rate values ​​(i.e., the actual carbon dioxide storage rate) in each subset of the target sample data, the prediction accuracy of the model can be determined. It can be understood that there is a one-to-one correspondence between the output of the model and the subset of the target sample data; that is, the output of the model corresponds one-to-one with the storage rate values ​​in the target sample dataset. Under these circumstances, the prediction accuracy of each model to be trained can be calculated.

[0069] Step A613: Determine the preset kernel function corresponding to the training model with the highest prediction accuracy value among multiple training models as the kernel function of the initial carbon dioxide burial rate prediction model.

[0070] Specifically, the higher the prediction accuracy value, the better and more stable the prediction effect of the corresponding training model. Based on this, this specific implementation determines the preset kernel function corresponding to the training model with the highest prediction accuracy value as the kernel function of the initial carbon dioxide burial rate prediction model, so as to improve the generation rate and prediction effect of the target carbon dioxide burial rate prediction model.

[0071] Step A620: Optimize the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained.

[0072] In one specific embodiment, step A620 includes the following steps: Step A621: Optimize the model parameters in the initial carbon dioxide burial rate prediction model using a grid search method to obtain the target model to be trained.

[0073] Specifically, step A621 optimizes the model parameters of the initial carbon dioxide burial rate prediction model using a grid search method, that is, determines the optimal values ​​of the penalty parameter C and the kernel function parameter gamma.

[0074] Step A630: Based on the target sample dataset, train the target model to be trained to obtain the undetermined model.

[0075] In one specific implementation, before training the target model to be trained, the target sample dataset is first divided into a training dataset and a test dataset, wherein both the training dataset and the test dataset include multiple subsets of target sample data.

[0076] In this specific embodiment, step A630 specifically includes: using the carbon dioxide storage-related values ​​and target influence parameters in the training dataset as inputs to the target model to be trained, using the storage rate values ​​in the training dataset as outputs to the target model to be trained, and training the target model to obtain a model to be determined.

[0077] Step A640: Determine whether the undetermined model meets the preset model conditions. If yes, then determine the undetermined model as the target carbon dioxide burial rate prediction model. If no, return to step A620 until the undetermined model meets the preset model conditions.

[0078] In one specific implementation, the coefficient of determination (COP) is used. The prediction accuracy of the undetermined model is evaluated by the learning curve and the first constraint condition corresponding to the coefficient of determination. The second constraint condition corresponding to the learning curve is used to improve the prediction effect of the final target carbon dioxide storage rate prediction model.

[0079] The coefficient of determination can be used to evaluate the predictive accuracy of a model. It reflects the degree to which the independent variable explains the variation of the dependent variable. Its value ranges from 0 to 1. The closer it is to 1, the higher the predictive accuracy of the model.

[0080] During the training process of the target model using the training dataset, as the amount of model input increases, if... The closer the value of the learning curve is to 1, the better the model's prediction fit, that is, the better the model's prediction performance.

[0081] In this specific embodiment, step A640, determining whether the model to be determined meets the preset model conditions, includes the following steps: Step 641: Determine the test dataset based on the target sample dataset.

[0082] The test dataset includes multiple subsets of target sample data.

[0083] Specifically, the target sample dataset can be divided into a training dataset and a test dataset using a proportional division method (e.g., 6:4 or 7:3). It is understood that if the target sample dataset has already been divided into a training dataset and a test dataset in step 630, then step 640 can directly use the test dataset, i.e., step 641 is unnecessary.

[0084] Step 642: For each target sample data subset in the test dataset, input the carbon dioxide burial-related values ​​and target influence parameter values ​​in the target sample data subset into the undetermined model, and output the predicted value corresponding to the burial rate value in the target sample data subset through the undetermined model.

[0085] Step 643: Determine the determination coefficient and learning curve of the undetermined model based on each embedded rate value and its corresponding predicted value in the test dataset.

[0086] Step 644: If the determination coefficient of the undetermined model satisfies the first constraint condition and / or the learning curve satisfies the second constraint condition, then the undetermined model is determined to satisfy the preset model condition.

[0087] In this specific implementation, the first constraint is set to a determination coefficient approaching 1, and the second constraint is set to a learning curve approaching 0.

[0088] It is understandable that when the determination coefficient and learning curve of the undetermined model both satisfy their respective constraints, it indicates that the undetermined model has high prediction accuracy.

[0089] In another specific embodiment, step A640, determining whether the model to be determined meets the preset model conditions, includes the following steps: Step 641: Determine the test dataset based on the target sample dataset.

[0090] Step 642: For each target sample data subset in the test dataset, input the carbon dioxide burial-related values ​​and target influence parameter values ​​in the target sample data subset into the undetermined model, and output the predicted value corresponding to the burial rate value in the target sample data subset through the undetermined model.

[0091] Step 643: Determine the first determination coefficient and the first learning curve based on each embedding rate value and its corresponding predicted value in the test dataset; and determine the second determination coefficient and the second learning curve based on each predicted value output by the model and its corresponding embedding rate value during the training process of the target model to be trained.

[0092] In this specific implementation, since the undetermined model is generated based on the target model to be trained, the first determination coefficient, the first learning curve, the second determination coefficient, and the second learning curve are all used as evaluation index values ​​of the undetermined model.

[0093] Step 644: If both the first determination coefficient and the second determination coefficient satisfy the first constraint condition, and both the first learning curve and the second learning curve satisfy the second constraint condition, then the undetermined model is determined to satisfy the preset model condition.

[0094] In this embodiment of the application, when the undetermined model meets the preset model conditions, it indicates that the undetermined model has met the expected requirements. At this time, the undetermined model can be determined as the target carbon dioxide burial rate prediction model to complete the prediction of the carbon dioxide burial rate. Otherwise, the model parameters need to be further optimized to improve the prediction effect of the undetermined model.

[0095] Unlike verifying model reliability solely through prediction accuracy, this application's embodiments evaluate the undetermined model using the coefficient of determination and learning curve to determine whether the model needs further optimization, thereby improving model reliability.

[0096] The following specific example illustrates the method for generating a carbon dioxide burial rate prediction model provided in the embodiments of this application.

[0097] Data from 219 carbon dioxide sequestration geological projects were collected and organized, resulting in 219 raw sample data subsets. Each raw sample data subset includes burial data such as total carbon dioxide injection volume, injection time, and injection rate, as well as geological parameters such as formation pressure, formation temperature, and depth below the surface. Preprocessing (including filtering and data normalization) was performed on each raw sample data subset to obtain the initial sample dataset.

[0098] The correlation values ​​of each geological parameter were determined using Spearman's correlation coefficient, Pearson's correlation coefficient, Kendall's correlation coefficient, and grey relational analysis. These values ​​were then summed and averaged to obtain the comprehensive evaluation index value for each geological parameter. The comprehensive evaluation index values ​​were ranked by correlation to screen for controlling factors, ultimately identifying seven target influencing parameters: pressure (P), temperature (T), depth above the surface, reservoir thickness, porosity (Φ), permeability (K), and total dissolved solids (TDS). The correlation values ​​of each target influencing parameter in this example are shown in Table 1 below.

[0099] Table 1 - Relevant values ​​of target influence parameters Evaluation methods P T depth thickness Φ K TDS Spearman 0.10 0.02 0.18 0.32 0.22 0.15 -0.18 Grey 0.79 0.77 0.77 0.83 0.78 0.86 0.84 Kendall 0.06 0.01 0.13 0.23 0.17 0.12 -0.13 Pearson 0.09 0.11 0.07 0.17 0.16 0.56 -0.11 Comprehensive evaluation index value 0.26 0.23 0.29 0.39 0.33 0.42 0.11 Relevance ranking 5 6 4 2 3 1 7 Based on the determined target impact parameters, data removal is performed on the initial sample dataset to determine the target sample dataset. Each subset of target sample data in the target sample dataset includes not only carbon dioxide burial-related values ​​and burial rate values, but also the parameter values ​​of seven geological parameters: pressure P, temperature T, depth from the ground, reservoir thickness, porosity Φ, permeability K, and total dissolved solids content TDS.

[0100] An SVM model was built, and its prediction accuracy was determined under different kernel functions. Since the SVM model (i.e., the model to be trained) based on the Gaussian kernel function (linear, polynomial, and Gaussian kernel functions) exhibits higher and more stable prediction accuracy, the Gaussian kernel function was chosen as the kernel function for the SVM model.

[0101] The optimal values ​​of the penalty parameter C and the parameter gamma of the Gaussian kernel function are determined using a grid search method, resulting in the target model to be trained. The range of the penalty parameter C is as follows: ,in, It is an integer, and The range of values ​​for the parameter gamma is: ,in, It is an integer, and .

[0102] The determined target sample dataset is divided into a training dataset and a test dataset. The target model to be trained is trained using the training dataset to obtain the undetermined model.

[0103] To more comprehensively evaluate the prediction model, i.e., the undetermined model, the coefficient of determination (COD) and learning curves were used to assess the overall prediction accuracy of the undetermined model. Specifically, the COD for the test dataset (i.e., the first COD) was 0.820, and the COD for the training dataset (i.e., the second COD) was 0.99. The learning curves for the test dataset (i.e., the first learning curve) were all close to 0, while the learning curves for the training dataset (i.e., the second learning curve) gradually approached 0 during training. Based on this, it is considered that the undetermined model can be directly used for predicting carbon dioxide storage rates, and that the undetermined model has good prediction accuracy.

[0104] As can be seen, the carbon dioxide burial rate prediction model generation method provided in this application first determines the target influencing parameters that have a significant impact on the carbon dioxide burial rate from a large number of geological parameters based on the initial sample dataset, thereby generating a target sample dataset. Then, a model is constructed and trained based on the target sample dataset, and finally a target carbon dioxide burial rate prediction model that meets the preset model conditions is obtained, ensuring that the target carbon dioxide burial rate prediction model has high prediction accuracy and reliability.

[0105] The carbon dioxide burial rate prediction model generation method provided in this application improves the accuracy and efficiency of carbon dioxide burial potential assessment by screening target influencing parameters and training the model. Based on the target carbon dioxide burial rate prediction model, carbon dioxide burial rate prediction is performed, providing theoretical support and technical guidance for carbon dioxide burial engineering.

[0106] Figure 1 This is a flowchart illustrating a method for generating a carbon dioxide burial rate prediction model in one embodiment. It should be understood that, although... Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise explicitly stated herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0107] Figure 2 A schematic flowchart illustrating the carbon dioxide burial rate prediction method according to an embodiment of this application is shown. Figure 2As shown in one embodiment of this application, a method for predicting carbon dioxide burial rate is provided, which may include the following steps.

[0108] Step B200: Obtain the target dataset to be predicted.

[0109] Step B400: Input the target dataset to be predicted into the target carbon dioxide burial rate prediction model obtained by the method described in the above embodiments to obtain the burial rate value corresponding to the target dataset to be predicted.

[0110] The target dataset to be predicted includes carbon dioxide storage-related values ​​and target impact parameters.

[0111] Specifically, the carbon dioxide burial-related values ​​included in the target dataset to be predicted correspond to the carbon dioxide burial-related values ​​included in the target sample dataset used to train the target carbon dioxide burial rate prediction model, so as to ensure that the target carbon dioxide burial rate prediction model can accurately predict the burial rate value corresponding to the target dataset to be predicted.

[0112] For example, when the relevant storage parameters in the target sample dataset include the total amount of carbon dioxide injected and the carbon dioxide injection time, the carbon dioxide storage-related values ​​in the target dataset to be predicted are also the values ​​of the total amount of carbon dioxide injected and the carbon dioxide injection time.

[0113] Optionally, in this embodiment of the application, after step B200, the method may further include the following steps.

[0114] Step B300: Perform data normalization processing on the target dataset to be predicted to obtain a new dataset to be predicted.

[0115] In this case, the new dataset to be predicted is replaced with the target dataset to be predicted, so that the target carbon dioxide burial rate prediction model can make predictions based on the new dataset to be predicted, thereby obtaining the burial rate value corresponding to the target dataset to be predicted.

[0116] The embodiments of this application can improve the prediction accuracy of the model by performing data normalization processing on the target dataset to be predicted.

[0117] As can be seen, the carbon dioxide burial rate prediction method provided in this application can directly obtain the burial rate value corresponding to the target carbon dioxide burial rate by inputting the target dataset into the target carbon dioxide burial rate prediction model after determining the target dataset. It is efficient and has a high prediction accuracy.

[0118] Figure 3A schematic diagram illustrating the structural block diagram of a carbon dioxide burial rate prediction model generation device according to an embodiment of this application is shown. Figure 3 As shown in one embodiment of this application, a carbon dioxide burial rate prediction model generation device is provided, which may include the following functional modules.

[0119] The first dataset generation module is used to generate an initial sample dataset. This initial sample dataset includes multiple subsets of initial sample data, each subset containing carbon dioxide burial-related values, burial rate values, and parameter values ​​for multiple geological parameters.

[0120] The influence parameter determination module is used to determine the target influence parameter from multiple geological parameters based on the initial sample dataset; The second dataset generation module is used to remove data from the initial sample dataset based on the target impact parameters to determine the target sample dataset. The target sample dataset includes multiple target sample data subsets, each of which includes carbon dioxide burial-related values, burial rate values, and parameter values ​​of the target impact parameters.

[0121] The model building module is used to build an initial carbon dioxide burial rate prediction model.

[0122] The model generation module is used to train the initial carbon dioxide burial rate prediction model based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets preset model conditions. The preset model conditions include at least one of a first constraint corresponding to the coefficient of determination and a second constraint corresponding to the learning curve.

[0123] In this embodiment of the application, the influencing parameter determination module may include: The parameter value set generation unit is used to generate a parameter value set corresponding to each geological parameter based on the initial sample dataset.

[0124] The evaluation index value determination unit is used to determine the correlation value of each geological parameter under each preset correlation coefficient algorithm based on the parameter value set corresponding to the geological parameter and the burial rate value in each initial sample data subset, and to determine the comprehensive evaluation index value corresponding to the geological parameter based on each correlation value.

[0125] The influencing parameter determination unit is used to determine the target influencing parameters based on the comprehensive evaluation index values ​​corresponding to each geological parameter.

[0126] In this embodiment of the application, the model generation module may include: The kernel function determination unit is used to determine the kernel function of the initial carbon dioxide burial rate prediction model.

[0127] The target model to be trained unit is used to optimize the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained.

[0128] The undetermined model determination unit is used to train the target model to be trained based on the target sample dataset to obtain the undetermined model.

[0129] The target model generation unit is used to determine whether the candidate model meets the preset model conditions. If yes, the candidate model is determined as the target carbon dioxide storage rate prediction model. If no, the process returns to the target training model determination unit until the candidate model meets the preset model conditions.

[0130] In this embodiment of the application, the kernel function determination unit may include: The model to be trained determination subunit is used to determine the model to be trained corresponding to each preset kernel function based on the initial carbon dioxide burial rate prediction model.

[0131] The prediction accuracy value determination subunit is used to determine the prediction accuracy value of each model to be trained based on the target sample dataset.

[0132] The kernel function determination sub-unit is used to determine the preset kernel function corresponding to the training model with the highest prediction accuracy value among multiple training models as the kernel function of the initial carbon dioxide burial rate prediction model.

[0133] In this embodiment of the application, the target model to be trained determination unit is specifically used for: The model parameters in the initial carbon dioxide burial rate prediction model are optimized using a grid search method to obtain the target model to be trained.

[0134] In this embodiment of the application, the target model generation unit may include: A test dataset determination subunit is used to determine the test dataset based on the target sample dataset. The test dataset includes multiple subsets of the target sample data.

[0135] The prediction value determination subunit is used to input the carbon dioxide burial-related values ​​and target influence parameter values ​​of each target sample data subset in the test dataset into the undetermined model, and output the prediction value corresponding to the burial rate value in the target sample data subset through the undetermined model.

[0136] The evaluation index value determination sub-unit is used to determine the determination coefficient and learning curve of the undetermined model based on each embedded rate value and its corresponding predicted value in the test dataset.

[0137] The model judgment subunit is used to determine that the model to be determined satisfies the preset model conditions if the determination coefficient of the model to be determined satisfies the first constraint condition and / or the learning curve satisfies the second constraint condition.

[0138] Since the carbon dioxide burial rate prediction model generation device provided in this application embodiment is a virtual device corresponding to the carbon dioxide burial rate prediction model generation method in the above embodiment, it can also solve the problem of low prediction accuracy of the carbon dioxide burial rate prediction method in the prior art.

[0139] Figure 4 A schematic block diagram of a carbon dioxide burial rate prediction device according to an embodiment of this application is shown. Figure 4 As shown in one embodiment of this application, a carbon dioxide burial rate prediction device is provided, which may include the following functional modules.

[0140] The data acquisition module is used to acquire the target dataset to be predicted. This dataset includes carbon dioxide storage-related values ​​and parameter values ​​of target impact parameters.

[0141] The burial rate prediction module is used to input the target dataset to be predicted into the target carbon dioxide burial rate prediction model obtained by the method described in the above embodiments, and to obtain the burial rate value corresponding to the target dataset to be predicted.

[0142] Since the carbon dioxide burial rate prediction device provided in this application embodiment is a virtual device corresponding to the carbon dioxide burial rate prediction method of the above embodiment, it can also solve the problem of low prediction accuracy of the carbon dioxide burial rate prediction method in the prior art.

[0143] This application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the carbon dioxide burial rate prediction model generation method and / or carbon dioxide burial rate prediction method described in the above embodiments.

[0144] The electronic device provided in this application embodiment includes a processor capable of running the carbon dioxide burial rate prediction model generation method of the aforementioned embodiment, thus also solving the problem of low prediction accuracy of the prior art carbon dioxide burial rate prediction method.

[0145] This application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the carbon dioxide burial rate prediction model generation method and / or carbon dioxide burial rate prediction method described in the above embodiments.

[0146] The machine-readable storage medium provided in this application embodiment stores instructions for causing a machine to execute the carbon dioxide burial rate prediction model generation method of the above embodiment, thus also solving the problem of low prediction accuracy of the carbon dioxide burial rate prediction method in the prior art.

[0147] Figure 5 The diagram schematically illustrates the internal structure of a computer device according to an embodiment of this application. Figure 5 As shown in one embodiment of this application, a computer device is provided, which can be a terminal. The computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05, and a memory (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A06. The network interface A02 is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor A01, it implements a method for generating a carbon dioxide storage rate prediction model and / or a method for predicting carbon dioxide storage rates. The display screen A04 of the computer device can be an LCD screen or an e-ink screen. The input device A05 of the computer device can be a touch layer covering the display screen, or a button, trackball, or touchpad set on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0148] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0149] In one embodiment, the carbon dioxide burial rate prediction model generation device provided in this application can be implemented as a computer program, which can be used in, for example... Figure 5The device operates on the computer shown. The computer's memory can store various program modules that make up the carbon dioxide storage rate prediction model generation apparatus. The computer program, composed of these program modules, causes the processor to execute the steps in the carbon dioxide storage rate prediction model generation method of the various embodiments of this application described in this specification.

[0150] Figure 5 The computer device shown can be used as follows Figure 3 The first dataset generation module in the carbon dioxide burial rate prediction model generation device shown executes step 200, the influencing parameter determination module executes step 300, the second dataset generation module executes step 400, the model building module executes step 500, and the model generation module executes step 600.

[0151] Similarly, in another embodiment, the carbon dioxide burial rate prediction device provided in this application can also be implemented as a computer program, which will not be described in detail here.

[0152] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0153] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0154] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0155] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0156] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0157] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0158] Computer-readable media include both permanent and non-permanent, removable and non-removable media, which can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0159] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0160] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for generating a carbon dioxide burial rate prediction model, characterized in that, The method includes: Generate an initial sample dataset; wherein the initial sample dataset includes multiple initial sample data subsets, each initial sample data subset including carbon dioxide burial-related values, burial rate values, and parameter values ​​of multiple geological parameters; Based on the initial sample dataset, target influence parameters are determined from multiple geological parameters; Based on the target impact parameters, data is removed from the initial sample dataset to determine the target sample dataset; wherein, the target sample dataset includes multiple target sample data subsets, and each target sample data subset includes carbon dioxide storage-related values, storage rate values, and parameter values ​​of the target impact parameters; Establish an initial carbon dioxide burial rate prediction model; The initial carbon dioxide burial rate prediction model is trained based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets the preset model conditions; wherein, the preset model conditions include at least one of the first constraint condition corresponding to the coefficient of determination and the second constraint condition corresponding to the learning curve.

2. The method according to claim 1, characterized in that, Based on the initial sample dataset, target influence parameters are determined from multiple geological parameters, including: Based on the initial sample dataset, generate a set of parameter values ​​corresponding to each geological parameter. For each geological parameter, based on the set of parameter values ​​corresponding to the geological parameter and the burial rate values ​​in each initial sample data subset, the correlation value corresponding to the geological parameter under each preset correlation coefficient algorithm is determined, and based on each correlation value, the comprehensive evaluation index value corresponding to the geological parameter is determined. The target influencing parameters are determined based on the comprehensive evaluation index values ​​corresponding to each geological parameter.

3. The method according to claim 1, characterized in that, The initial carbon dioxide burial rate prediction model is trained based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets preset model conditions, including: Determine the kernel function of the initial carbon dioxide burial rate prediction model; Optimize the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained; Based on the target sample dataset, the target model to be trained is trained to obtain the undetermined model; Determine whether the undetermined model meets the preset model conditions. If yes, then determine the undetermined model as the target carbon dioxide burial rate prediction model. If no, return to the step of optimizing the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained, until the undetermined model meets the preset model conditions.

4. The method according to claim 3, characterized in that, Determining the kernel function of the initial carbon dioxide burial rate prediction model includes: For each preset kernel function, the model to be trained corresponding to the preset kernel function is determined based on the initial carbon dioxide storage rate prediction model. Based on the target sample dataset, determine the prediction accuracy value of each model to be trained; The kernel function corresponding to the training model with the highest prediction accuracy value among multiple training models is determined as the kernel function of the initial carbon dioxide storage rate prediction model. Optimizing the model parameters in the initial carbon dioxide burial rate prediction model to obtain the target model to be trained includes: The model parameters in the initial carbon dioxide burial rate prediction model are optimized using a grid search method to obtain the target model to be trained.

5. The method according to claim 3, characterized in that, Determining whether the model to be determined satisfies the preset model conditions includes: Based on the target sample dataset, a test dataset is determined; wherein the test dataset includes multiple subsets of the target sample data; For each subset of target sample data in the test dataset, the carbon dioxide storage-related values ​​and target impact parameters in the subset of target sample data are input into the undetermined model, and the undetermined model outputs the predicted value corresponding to the storage rate value in the subset of target sample data. Based on each embedded rate value and its corresponding predicted value in the test dataset, determine the determination coefficient and learning curve of the model to be determined; If the determination coefficient of the undetermined model satisfies the first constraint condition and / or the learning curve satisfies the second constraint condition, then the undetermined model is determined to satisfy the preset model condition.

6. A method for predicting carbon dioxide burial rate, characterized in that, The method includes: Obtain the target dataset to be predicted; wherein, the target dataset to be predicted includes carbon dioxide storage-related values ​​and parameter values ​​of target impact parameters; The target dataset to be predicted is input into the target carbon dioxide burial rate prediction model generated by the method of any one of claims 1 to 5 to obtain the burial rate value corresponding to the target dataset to be predicted.

7. A device for generating a carbon dioxide burial rate prediction model, characterized in that, The device includes: The first dataset generation module is used to generate an initial sample dataset; wherein, the initial sample dataset includes multiple initial sample data subsets, and each initial sample data subset includes carbon dioxide burial-related values, burial rate values, and parameter values ​​of multiple geological parameters; The influence parameter determination module is used to determine the target influence parameter from multiple geological parameters based on the initial sample dataset; The second dataset generation module is used to remove data from the initial sample dataset based on the target impact parameters to determine the target sample dataset; wherein, the target sample dataset includes multiple target sample data subsets, and each target sample data subset includes carbon dioxide storage-related values, storage rate values ​​and parameter values ​​of the target impact parameters; The model building module is used to build an initial carbon dioxide burial rate prediction model; The model generation module is used to train the initial carbon dioxide burial rate prediction model based on the target sample dataset to obtain a target carbon dioxide burial rate prediction model that meets preset model conditions; the preset model conditions include at least one of a first constraint condition corresponding to the coefficient of determination and a second constraint condition corresponding to the learning curve.

8. A carbon dioxide burial rate prediction device, characterized in that, The device includes: The data acquisition module is used to acquire the target dataset to be predicted; wherein, the target dataset to be predicted includes carbon dioxide storage-related values ​​and parameter values ​​of target impact parameters; The burial rate prediction module is used to input the target dataset to be predicted into the target carbon dioxide burial rate prediction model generated by the method according to any one of claims 1 to 5, and obtain the burial rate value corresponding to the target dataset to be predicted.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the carbon dioxide burial rate prediction model generation method according to any one of claims 1 to 5 and / or the carbon dioxide burial rate prediction method according to claim 6.

10. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform the carbon dioxide burial rate prediction model generation method according to any one of claims 1 to 5 and / or the carbon dioxide burial rate prediction method according to claim 6.

Images