CN-121999899-A - Method, device and equipment for generating and predicting carbon dioxide buried rate prediction model

CN121999899ACN 121999899 ACN121999899 ACN 121999899ACN-121999899-A

Abstract

The application provides a method, a device and equipment for generating and predicting a carbon dioxide embedding rate prediction model, and relates to the technical field of geological sealing, wherein the method for generating the prediction model comprises the steps of generating an initial sample data set; the method comprises the steps of determining target influence parameters from a plurality of geological parameters according to an initial sample data set, removing data from the initial sample data set based on the target influence parameters to determine 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 preset model conditions, wherein the preset model conditions comprise at least one of first constraint conditions corresponding to decision coefficients and second constraint conditions corresponding to learning curves. The method can effectively solve the problem of low prediction accuracy of the existing carbon dioxide buried rate prediction method.

Inventors

ZHOU BING
LUN ZENGMIN
ZHOU XIA
LI TIANYI
GE QIAOYU
YIN XIA

Assignees

中国石油化工股份有限公司
中国石油化工股份有限公司石油勘探开发研究院

Dates

Publication Date: 20260508
Application Date: 20241107

Claims (10)

1. A method for generating a prediction model of a carbon dioxide sequestration rate, the method comprising: Generating an initial sample data set, wherein the initial sample data set comprises a plurality of initial sample data subsets, and each initial sample data subset comprises a carbon dioxide sequestration related value, a sequestration speed value and parameter values of a plurality of geological parameters; determining target influence parameters from a plurality of geological parameters according to the initial sample data set; Performing data removal on the initial sample data set based on target influence parameters to determine a target sample data set, wherein the target sample data set comprises a plurality of target sample data subsets, and each target sample data subset comprises a carbon dioxide sequestration related value, a sequestration speed value and a parameter value of the target influence parameters; 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 preset model conditions, wherein the preset model conditions comprise at least one of first constraint conditions corresponding to decision coefficients and second constraint conditions corresponding to learning curves.
2. The method of claim 1, wherein determining a target influencing parameter from a plurality of geological parameters from the initial sample dataset comprises: generating a parameter value set corresponding to each geological parameter according to the initial sample data set; For each geological parameter, determining a correlation value corresponding to the geological parameter under each preset correlation coefficient algorithm according to a parameter value set corresponding to the geological parameter and a buried rate value in each initial sample data subset, and determining a comprehensive evaluation index value corresponding to the geological parameter according to each correlation value; and determining target influence parameters according to the comprehensive evaluation index values corresponding to the geological parameters.
3. The method of claim 1, wherein training the initial carbon dioxide sequestration rate prediction model based on the target sample dataset to obtain a target carbon dioxide sequestration rate prediction model satisfying a preset model condition comprises: determining a kernel function of the initial carbon dioxide sequestration rate prediction model; optimizing model parameters in the initial carbon dioxide embedding rate prediction model to obtain a target model to be trained; training the target model to be trained based on the target sample data set to obtain a model to be determined; and determining whether the undetermined model meets the preset model conditions, if so, determining the undetermined model as a target carbon dioxide embedding rate prediction model, and if not, returning to the step of optimizing model parameters in the initial carbon dioxide embedding rate prediction model to obtain a target model to be trained until the undetermined model meets the preset model conditions.
4. The method of claim 3, wherein determining a kernel function of the initial carbon dioxide sequestration rate prediction model comprises: For each preset kernel function, determining a model to be trained corresponding to the preset kernel function based on the initial carbon dioxide embedding rate prediction model; determining the prediction accuracy value of each model to be trained according to the target sample data set; Determining a preset kernel function corresponding to a to-be-trained model with the maximum prediction accuracy value in a plurality of to-be-trained models as the kernel function of the initial carbon dioxide embedding rate prediction model; Optimizing model parameters in the initial carbon dioxide embedding rate prediction model to obtain a target model to be trained, wherein the method comprises the following steps: and optimizing model parameters in the initial carbon dioxide embedding rate prediction model by a grid search method to obtain a target model to be trained.
5.A method according to claim 3, wherein determining whether the predetermined model satisfies the predetermined model condition comprises: Determining a test data set according to the target sample data set, wherein the test data set comprises a plurality of target sample data subsets; For each target sample data subset in the test data set, inputting a carbon dioxide buried related value and a parameter value of a target influence parameter in the target sample data subset into the pending model, and outputting a predicted value corresponding to a buried rate value in the target sample data subset through the pending model; Determining a decision coefficient and a learning curve of the undetermined model according to each buried rate value and a corresponding predicted value in the test data set; and if the decision coefficient of the undetermined model meets the first constraint condition and/or the learning curve meets the second constraint condition, determining that the undetermined model meets the preset model condition.
6. A method for predicting a carbon dioxide sequestration rate, the method comprising: Acquiring a target data set to be predicted, wherein the target data set to be predicted comprises a carbon dioxide buried related value and a parameter value of a target influence parameter; Inputting the target data set to be predicted into a target carbon dioxide embedding rate prediction model generated by adopting the method of any one of claims 1 to 5 to obtain an embedding rate value corresponding to the target data set to be predicted.
7. A carbon dioxide sequestration rate prediction model generation device, the device comprising: The system comprises a first data set generation module, a second data set generation module and a data processing module, wherein the first data set generation module is used for generating an initial sample data set, the initial sample data set comprises a plurality of initial sample data subsets, and each initial sample data subset comprises a carbon dioxide embedding related value, an embedding speed value and parameter values of a plurality of geological parameters; The influence parameter determining module is used for determining target influence parameters from a plurality of geological parameters according to the initial sample data set; The system comprises a first data set generation module, a second data set generation module and a storage module, wherein the first data set generation module is used for carrying out data removal on an initial sample data set based on target influence parameters to determine a target sample data set, the target sample data set comprises a plurality of target sample data subsets, and each target sample data subset comprises a carbon dioxide storage related value, a storage speed value and a parameter value of the target influence parameters; The model building module is used for building an initial carbon dioxide buried rate prediction model; The model generation module is used for 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 preset model conditions, wherein the preset model conditions comprise at least one of first constraint conditions corresponding to decision coefficients and second constraint conditions corresponding to learning curves.
8. A carbon dioxide sequestration rate prediction device, the device comprising: The system comprises a data acquisition module, a data prediction module and a data prediction module, wherein the data acquisition module is used for acquiring a target data set to be predicted, and the target data set to be predicted comprises a carbon dioxide buried related value and a parameter value of a target influence parameter; the embedding rate prediction module is configured to input the target to-be-predicted data set into a target carbon dioxide embedding rate prediction model generated by using the method according to any one of claims 1 to 5, so as to obtain an embedding rate value corresponding to the target to-be-predicted data set.
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 the processor implements the carbon dioxide sequestration rate prediction model generation method of any one of claims 1 to 5 and/or the carbon dioxide sequestration rate prediction method of claim 6 when executing the computer program.
10. A machine-readable storage medium having instructions stored thereon, which when executed by a processor, cause the processor to be configured to perform the carbon dioxide sequestration rate prediction model generation method according to any one of claims 1 to 5 and/or the carbon dioxide sequestration rate prediction method according to claim 6.

Description

Method, device and equipment for generating and predicting carbon dioxide buried rate prediction model Technical Field The application relates to the technical field of geological sequestration, in particular to a method and a device for generating a carbon dioxide sequestration rate prediction model, a method and a device for predicting the carbon dioxide sequestration rate, electronic equipment and a machine-readable storage medium. Background Carbon dioxide geological sequestration technology is an important means of reducing greenhouse gas emissions. At present, research on carbon dioxide sequestration rate is mainly focused on two methods of indoor physical simulation experiments and numerical simulation. However, geological storage is affected by numerous factors, which have significant differences in the extent and laws of influence, making laboratory experiments costly and difficult. Meanwhile, the numerical simulation process is complex, the result is influenced by the accuracy of basic experimental parameters, the limitation is large, and a reasonable model is difficult to build under the condition of complex stratum environment. In the related art, in order to ensure the long-term safety of carbon dioxide sequestration, prevent environmental risks such as leakage, and ensure the safety of ecosystems and human beings, improving the accuracy of carbon dioxide sequestration rate prediction is a problem to be solved urgently. Disclosure of Invention The embodiment of the application aims to provide a method, a device and equipment for generating and predicting a carbon dioxide embedding rate prediction model, so as to solve the problem of low prediction accuracy of the carbon dioxide embedding rate prediction method in the prior art. In order to achieve the above object, a first aspect of the present application provides a method for generating a prediction model of a carbon dioxide sequestration rate, the method comprising: Generating an initial sample data set, wherein the initial sample data set comprises a plurality of initial sample data subsets, and each initial sample data subset comprises a carbon dioxide sequestration related value, a sequestration speed value and parameter values of a plurality of geological parameters; determining target influence parameters from a plurality of geological parameters according to the initial sample data set; Performing data removal on the initial sample data set based on target influence parameters to determine a target sample data set, wherein the target sample data set comprises a plurality of target sample data subsets, and each target sample data subset comprises a carbon dioxide sequestration related value, a sequestration speed value and a parameter value of the target influence parameters; 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 preset model conditions, wherein the preset model conditions comprise at least one of first constraint conditions corresponding to decision coefficients and second constraint conditions corresponding to learning curves. In an embodiment of the present application, determining, from the initial sample dataset, a target influencing parameter from a plurality of geological parameters includes: generating a parameter value set corresponding to each geological parameter according to the initial sample data set; For each geological parameter, determining a correlation value corresponding to the geological parameter under each preset correlation coefficient algorithm according to a parameter value set corresponding to the geological parameter and a buried rate value in each initial sample data subset, and determining a comprehensive evaluation index value corresponding to the geological parameter according to each correlation value; and determining target influence parameters according to the comprehensive evaluation index values corresponding to the geological parameters. In the embodiment of the present application, training the initial carbon dioxide sequestration rate prediction model based on the target sample data set to obtain a target carbon dioxide sequestration rate prediction model satisfying a preset model condition, including: determining a kernel function of the initial carbon dioxide sequestration rate prediction model; optimizing model parameters in the initial carbon dioxide embedding rate prediction model to obtain a target model to be trained; training the target model to be trained based on the target sample data set to obtain a model to be determined; and determining whether the undetermined model meets the preset model conditions, if so, determining the undetermined model as a target carbon dioxide embedding rate prediction model, and if not, returning to the step of optimizing model parameters in the ini