Prediction method for underground rock fracture risk in carbon dioxide sequestration project

Abstract

The invention provides a method for predicting an underground rock fracture risk in a carbon dioxide sequestration project, and the method comprises the steps: obtaining carbon dioxide time sequence pressure data of a plurality of monitoring points in a carbon dioxide sequestration region; performing feature analysis on the carbon dioxide time sequence pressure data to obtain spatial features and time features; fusing the spatial features and the time features to obtain space-time fusion features; geological parameters and rock fracture historical data of the carbon dioxide sequestration area are obtained; performing characteristic engineering processing on the geological parameters and the rock fracture historical data to obtain geomechanical characteristics; inputting the space-time fusion features and the geomechanical features into a pre-trained rock fracture prediction model to obtain rock fracture prediction data; wherein the rock fracture prediction model is a CNN-LSTM hybrid model. According to the technical scheme, the underground rock fracture risk in the carbon dioxide sequestration project can be accurately predicted.

Description

Technical Field

[0001] This application relates to the field of carbon dioxide sequestration technology, and in particular to a method, apparatus, equipment and storage medium for predicting the risk of underground rock fracturing in carbon dioxide sequestration projects. Background Technology

[0002] Carbon dioxide geological storage has received widespread attention in recent years as a key technology for mitigating global climate change. During the storage process, the stability of underground rock is crucial; rock fracturing can lead to carbon dioxide leakage, posing a threat to the environment and engineering safety.

[0003] In related technologies, traditional rock fracture risk assessment methods often rely on complex physical models and limited monitoring data, and their accuracy and timeliness need to be improved. Summary of the Invention

[0004] This application aims to at least partially address one of the technical problems in the related art.

[0005] In a first aspect, this application proposes a method for predicting the risk of underground rock fracturing in carbon dioxide sequestration projects. The method includes: acquiring time-series carbon dioxide pressure data from multiple monitoring points within the carbon dioxide sequestration area; performing feature analysis on the time-series carbon dioxide pressure data to obtain spatial and temporal features; fusing the spatial and temporal features to obtain spatiotemporal fusion features; acquiring geological parameters and historical rock fracturing data of the carbon dioxide sequestration area; performing feature engineering on the geological parameters and historical rock fracturing data to obtain geomechanical features; and inputting the spatiotemporal fusion features and the geomechanical features into a pre-trained rock fracturing prediction model to obtain rock fracturing prediction data; wherein the rock fracturing prediction model is a CNN-LSTM hybrid model.

[0006] In one implementation, the step of performing feature analysis on the carbon dioxide time-series pressure data to obtain spatial features includes: performing spatial autocorrelation analysis on the time-series pressure data to obtain a spatial correlation strength map; inputting the spatial correlation strength map into a pre-trained graph convolutional network to initialize spatial correlation features; performing spatial gridding and time series reconstruction on the time-series pressure data to obtain a pressure field tensor; and using multi-scale convolutional kernels to extract features from the pressure difference tensor to obtain local spatial features.

[0007] In one implementation, feature analysis is performed on the carbon dioxide time-series pressure data to obtain time features, including: performing a Fourier transform on the time-series pressure data to obtain initial time features; and performing wavelet decomposition on the initial time features to obtain the time features.

[0008] In one implementation, fusing the spatial features and the temporal features to obtain spatiotemporal fusion features includes: aligning the feature dimensions of the spatial features and the temporal features; obtaining the spatial residual and temporal residual between the spatial features and the temporal features; mapping the spatial residual to the temporal features to obtain mapped temporal features, and setting the temporal residual to the spatial features to obtain mapped spatial features; and fusing the mapped temporal features and the mapped spatial features to obtain the spatiotemporal fusion features.

[0009] In one implementation, the geological parameters include at least one of the following: rock type, permeability, porosity, and geostress.

[0010] In one implementation, the rock fracture history data includes at least one of the following: rock fracture location, rock fracture time, fracture scale, and fracture type.

[0011] Secondly, this application proposes a device for predicting the risk of underground rock fracturing in carbon dioxide sequestration projects. The device includes: a first acquisition module for acquiring time-series carbon dioxide pressure data from multiple monitoring points within the carbon dioxide sequestration area; a first processing module for performing feature analysis on the time-series carbon dioxide pressure data to obtain spatial and temporal features; a second processing module for fusing the spatial and temporal features to obtain spatiotemporal fusion features; a second acquisition module for acquiring geological parameters and historical rock fracturing data of the carbon dioxide sequestration area; a third processing module for performing feature engineering processing on the geological parameters and historical rock fracturing data to obtain geomechanical features; and a fourth processing module for inputting the spatiotemporal fusion features and the geomechanical features into a pre-trained rock fracturing prediction model to obtain rock fracturing prediction data; wherein the rock fracturing prediction model is a CNN-LSTM hybrid model.

[0012] In one implementation, the first processing module can be used to: perform spatial autocorrelation analysis on the time-series pressure data to obtain a spatial correlation strength map; input the spatial correlation strength map into a pre-trained graph convolutional network to initialize spatial correlation features; perform spatial gridding and time series reconstruction on the time-series pressure data to obtain a pressure field tensor; and use multi-scale convolutional kernels to extract features from the pressure difference tensor to obtain local spatial features.

[0013] In one implementation, the first processing module can be used to: perform Fourier transform on the time-series pressure data to obtain initial time features; and perform wavelet decomposition on the initial time features to obtain the time features.

[0014] In one implementation, the second processing module can be used to: align the spatial features and the temporal features in terms of feature dimensions; obtain the spatial residual and temporal residual between the spatial features and the temporal features; map the spatial residual to the temporal features to obtain the mapped temporal features, and set the temporal residual to the spatial features to obtain the mapped spatial features; perform feature fusion on the mapped temporal features and the mapped spatial features to obtain the spatiotemporal fusion features.

[0015] In one implementation, the geological parameters include at least one of the following: rock type, permeability, porosity, and geostress.

[0016] In one implementation, the rock fracture history data includes at least one of the following: rock fracture location, rock fracture time, fracture scale, and fracture type.

[0017] Thirdly, this application proposes an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method for predicting the risk of underground rock fracturing in a carbon dioxide sequestration project as described in the first aspect.

[0018] Fourthly, this application proposes a computer-readable storage medium for storing instructions that, when executed, cause the method described in the first aspect to be implemented.

[0019] Fifthly, this application proposes a computer program product, including a computer program that, when executed by a processor, implements the steps of the method for predicting the risk of underground rock fracturing in a carbon dioxide sequestration project as described in the first aspect.

[0020] The method, apparatus, equipment, and storage medium provided in this application for predicting the risk of underground rock fracturing in carbon dioxide storage projects can process time-series pressure data from multiple monitoring points within the carbon dioxide storage area to obtain spatiotemporal fusion characteristics, and process geological parameters and historical rock fracturing data to obtain geomechanical characteristics. These spatiotemporal fusion characteristics and geomechanical characteristics are then input into a pre-trained rock fracturing prediction model to predict rock fracturing risk. This enables accurate prediction of underground rock fracturing risk in carbon dioxide storage projects.

[0021] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0022] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0023] Figure 1 This is a flowchart illustrating a method for predicting the risk of underground rock fracturing in a carbon dioxide sequestration project, as provided in an embodiment of this application.

[0024] Figure 2 This is a schematic diagram of the structure of a device for predicting the risk of underground rock fracturing in a carbon dioxide sequestration project, provided in an embodiment of this application.

[0025] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The following describes, with reference to the accompanying drawings, a method and apparatus for predicting the risk of underground rock fracturing in a carbon dioxide sequestration project according to an embodiment of this application.

[0028] Figure 1 This is a flowchart illustrating a method for predicting the risk of underground rock fracturing in a carbon dioxide sequestration project, as provided in an embodiment of this application. Figure 1 As shown, the method may include, but is not limited to, the following steps:

[0029] S101: Obtain time-series pressure data of carbon dioxide from multiple monitoring points within the carbon dioxide storage area.

[0030] For example, multiple monitoring points are set up in the carbon dioxide storage area, and a pressure sensor is set up at each monitoring point, so as to obtain the carbon dioxide time-series pressure data of each monitoring point in the carbon dioxide storage area within a preset time period through the pressure sensor.

[0031] S102: Perform feature analysis on the carbon dioxide time-series pressure data to obtain spatial and temporal features.

[0032] For example, time features and spatial features are extracted from the time-series pressure data corresponding to each monitoring point to obtain the time features and spatial features corresponding to each monitoring point.

[0033] In one implementation, the above-mentioned feature analysis of the carbon dioxide time-series pressure data to obtain spatial features may include the following steps A1 to A4:

[0034] A1: Perform spatial autocorrelation analysis on the time-series pressure data to obtain a spatial correlation strength map.

[0035] For example, each monitoring point is treated as a grid cell, and the time-series stress data of each monitoring point are aligned with the grid. The spatial weight of each grid with the surrounding grids is calculated. The correlation strength between each grid and its neighborhood is calculated based on the spatial weight. Finally, the correlation strength values ​​of each grid are mapped to a spatial correlation strength map.

[0036] A2: Input the spatial association strength map into a pre-trained graph convolutional network to generate initial spatial association features.

[0037] For example, the association strength value of each grid in the spatial association strength map is used as the initial node feature of the graph convolutional network, and the spatial proximity relationship between grids is used as the edge structure of the graph. High-order spatial association patterns are extracted through multi-layer convolution operations, and the initial spatial association features are output.

[0038] A3: Spatial gridding and time series reconstruction are performed on the time-series pressure data to obtain the pressure field tensor.

[0039] For example, the time-series pressure data of each monitoring point are arranged according to spatial coordinates and combined into a three-dimensional tensor of "number of spatial rows × number of spatial columns × number of time steps" as the pressure field tensor, while retaining spatial distribution and temporal evolution information.

[0040] A4: Multi-scale convolution kernels are used to extract features from the pressure difference tensor to obtain local spatial features.

[0041] In one implementation, feature analysis is performed on the carbon dioxide time-series pressure data to obtain time features, including: performing a Fourier transform on the time-series pressure data to obtain initial time features; and performing wavelet decomposition on the initial time features to obtain the time features.

[0042] For example, Fourier transform is performed on the time-series pressure data to convert it from the time domain to the frequency domain, and the amplitude and phase of different frequency components are obtained, which are used as the initial time features. Then, wavelet decomposition is performed on the initial time features to separate the high-frequency instantaneous pressure disturbance and the low-frequency long-term trend components at different time scales as time features.

[0043] Step S103: Fuse the spatial features and the temporal features to obtain spatiotemporal fusion features.

[0044] For example, the temporal and spatial features corresponding to each monitoring point are fused to obtain the spatiotemporal fusion features corresponding to each monitoring point.

[0045] In some embodiments, fusing the spatial features and the temporal features to obtain spatiotemporal fused features may include steps B1 to B4:

[0046] B1: Align the spatial features and the temporal features according to their feature dimensions.

[0047] For example, feature compression is performed on spatial and temporal features to make their feature dimensions the same, thereby achieving feature dimension alignment.

[0048] B2: Obtain the spatial residual and temporal residual between the spatial feature and the temporal feature.

[0049] For example, the spatial features are linearly transformed to obtain the predicted temporal features, and the deviation between the predicted temporal features and the temporal features is obtained as the spatial residual; and the temporal features are linearly transformed to obtain the predicted spatial features, and the deviation between the predicted spatial features and the spatial features is obtained as the temporal residual.

[0050] In some embodiments, the spatial features include spatial correlation features and local spatial features. Then, the spatial correlation features are linearly transformed to generate predicted local spatial features and the difference between the predicted local spatial features and the actual local spatial features is calculated to obtain a first spatial residual. The spatial features are linearly transformed to generate predicted temporal features and the difference between the predicted temporal features and the actual temporal features is calculated to obtain a second spatial residual. The temporal features are linearly transformed to generate predicted spatial features and the difference between the predicted spatial features and the actual spatial features is calculated to obtain a temporal residual.

[0051] B3: Map the spatial residual to the temporal feature to obtain the mapped temporal feature, and set the temporal residual to the spatial feature to obtain the mapped spatial feature.

[0052] For example, the spatial residual dimension is first adjusted to match the temporal feature, and then the element-wise addition is used to superimpose it onto the original temporal feature, so that the temporal feature incorporates the spatial local deviation information to obtain the mapped temporal feature; the temporal residual is mapped to the spatial feature to obtain the mapped spatial feature, and the temporal residual dimension is first adjusted to match the spatial feature, and then superimposed according to the corresponding position of the spatial grid to obtain the mapped spatial feature.

[0053] B4: Perform feature fusion on the mapped time features and the mapped space features to obtain the spatiotemporal fusion features.

[0054] For example, the spatiotemporal fusion feature is obtained by fusing the mapping time feature and the mapping space feature using a combination of splicing and weighted summation, and then processed by an activation function.

[0055] S104: Obtain the geological parameters and historical rock fracture data of the carbon dioxide storage area.

[0056] In the embodiments of this application, the geological parameters mentioned above include at least one of the following: rock type; permeability; porosity; and geostress.

[0057] In the embodiments of this application, the aforementioned historical rock fracture data includes at least one of the following: rock fracture location and rock fracture time.

[0058] S105: Perform feature engineering processing on the geological parameters and the rock fracture history data to obtain geomechanical characteristics.

[0059] For example, the geological parameters are first standardized and discretized to extract static geological features; then, the frequency and interval of the rupture history data are statistically analyzed according to time windows, and dynamic rupture features are generated by combining spatial location; finally, the two types of features are spliced ​​together and highly correlated variables are selected to obtain geomechanical features.

[0060] S106: Input the spatiotemporal fusion features and the geomechanical features into the pre-trained rock fracture prediction model to obtain rock fracture prediction data.

[0061] The rock fracture prediction model is a CNN-LSTM hybrid model.

[0062] Understandably, CNN excels at extracting spatial features, capturing spatial patterns such as local pressure abrupt changes and regional gradient anomalies from data like pressure field distribution; while LSTM focuses on mining time series patterns, capturing temporal features such as the dynamic evolution and periodic fluctuations of pressure over time. Combining the two preserves both local details and macroscopic correlations in the spatial dimension, while integrating trend changes in the temporal dimension, comprehensively depicting the complex relationship between rock fracturing risk and spatiotemporal coupling characteristics, and capturing spatiotemporal anomaly signals before fracturing more accurately than a single model.

[0063] In some embodiments, the above-described fracture prediction model is used to train a rock fracture prediction model based on sample training data; the cross-entropy loss function is used as the output with the fracture risk probability of each monitoring point; the parameters are optimized through backpropagation and the hyperparameters are adjusted in conjunction with the validation set for training.

[0064] By implementing the embodiments of this application, time-series pressure data from multiple monitoring points within a carbon dioxide storage area can be processed to obtain spatiotemporal fusion characteristics, and geological parameters and historical rock fracture data can be processed to obtain geomechanical characteristics. These spatiotemporal fusion characteristics and geomechanical characteristics are then input into a pre-trained rock fracture prediction model to predict rock fracture risk. This enables accurate prediction of underground rock fracture risk in carbon dioxide storage projects.

[0065] Please see Figure 2 , Figure 2 This is a schematic diagram of a device for predicting the risk of underground rock fracturing in a carbon dioxide sequestration project, provided in an embodiment of this application. Figure 2 As shown, the device 200 includes: a first acquisition module for acquiring time-series carbon dioxide pressure data from multiple monitoring points within a carbon dioxide sequestration area; a first processing module for performing feature analysis on the time-series carbon dioxide pressure data to obtain spatial and temporal features; a second processing module for fusing the spatial and temporal features to obtain spatiotemporal fusion features; a second acquisition module for acquiring geological parameters and historical rock fracture data of the carbon dioxide sequestration area; a third processing module for performing feature engineering processing on the geological parameters and historical rock fracture data to obtain geomechanical features; and a fourth processing module for inputting the spatiotemporal fusion features and the geomechanical features into a pre-trained rock fracture prediction model to obtain rock fracture prediction data; wherein the rock fracture prediction model is a CNN-LSTM hybrid model.

[0066] In one implementation, the first processing module can be used to: perform spatial autocorrelation analysis on the time-series pressure data to obtain a spatial correlation strength map; input the spatial correlation strength map into a pre-trained graph convolutional network to initialize spatial correlation features; perform spatial gridding and time series reconstruction on the time-series pressure data to obtain a pressure field tensor; and use multi-scale convolutional kernels to extract features from the pressure difference tensor to obtain local spatial features.

[0067] In one implementation, the first processing module can be used to: perform Fourier transform on the time-series pressure data to obtain initial time features; and perform wavelet decomposition on the initial time features to obtain the time features.

[0068] In one implementation, the second processing module can be used to: align the spatial features and the temporal features in terms of feature dimensions; obtain the spatial residual and temporal residual between the spatial features and the temporal features; map the spatial residual to the temporal features to obtain the mapped temporal features, and set the temporal residual to the spatial features to obtain the mapped spatial features; perform feature fusion on the mapped temporal features and the mapped spatial features to obtain the spatiotemporal fusion features.

[0069] In one implementation, the geological parameters include at least one of the following: rock type, permeability, porosity, and geostress.

[0070] In one implementation, the rock fracture history data includes at least one of the following: rock fracture location, rock fracture time, fracture scale, and fracture type.

[0071] It should be noted that the explanation of the aforementioned method embodiment for predicting the risk of underground rock fracture in carbon dioxide sequestration projects also applies to the device for predicting the risk of underground rock fracture in carbon dioxide sequestration projects in this embodiment, and will not be repeated here.

[0072] To implement the above embodiments, this application also proposes an electronic device. Please see [link to relevant documentation]. Figure 3 , Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of this application. For example... Figure 3 As shown, the electronic device 300 includes: a processor 301, and a memory 302 communicatively connected to the processor 301; the memory 302 stores computer execution instructions; the processor 301 executes the computer execution instructions stored in the memory to implement the method provided in the foregoing embodiments.

[0073] To implement the above embodiments, this application also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0074] To implement the above embodiments, this application also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0075] It should be noted that the acquisition, transmission, storage, use, and processing of data in this application comply with the relevant provisions of national laws and regulations and do not violate public order and good morals.

[0076] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0077] It is worth noting that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, they do not mean that the applicant has used or necessarily used the solution.

[0078] In the description of this application, unless otherwise stated, " / " means "or", for example, A / B can mean A or B; "and / or" in this document is merely a description of the relationship between related objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone.

[0079] In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0080] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0081] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0082] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0083] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0084] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0085] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0086] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for predicting the risk of underground rock fracturing in carbon dioxide sequestration projects, characterized in that, include: Acquire time-series pressure data of carbon dioxide from multiple monitoring points within the carbon dioxide storage area; Feature analysis was performed on the carbon dioxide time-series pressure data to obtain spatial and temporal features; The spatial features and the temporal features are fused to obtain spatiotemporal fused features; Obtain the geological parameters and historical rock fracture data of the carbon dioxide storage area; The geological parameters and the rock fracture history data are subjected to feature engineering processing to obtain geomechanical characteristics; The spatiotemporal fusion features and the geomechanical features are input into a pre-trained rock fracture prediction model to obtain rock fracture prediction data; wherein, the rock fracture prediction model is a CNN-LSTM hybrid model.

2. The method according to claim 1, characterized in that, The step of performing feature analysis on the carbon dioxide time-series pressure data to obtain spatial features includes: Spatial autocorrelation analysis was performed on the time-series pressure data to obtain a spatial correlation strength map; The spatial association strength map is input into a pre-trained graph convolutional network to generate initial spatial association features; The time-series pressure data is spatially gridded and time-series reconstructed to obtain the pressure field tensor; The pressure difference tensor is used to extract features by employing multi-scale convolution kernels to obtain local spatial features.

3. The method according to claim 1, characterized in that, Feature analysis is performed on the carbon dioxide time-series pressure data to obtain time features, including: Perform a Fourier transform on the time-series pressure data to obtain the initial time characteristics; The initial time features are decomposed using wavelet decomposition to obtain the time features.

4. The method according to claim 1, characterized in that, The process of fusing the spatial features and the temporal features to obtain spatiotemporal fusion features includes: Align the spatial features and the temporal features according to their feature dimensions; Obtain the spatial residual and temporal residual between the spatial feature and the temporal feature; The spatial residual is mapped to the temporal feature to obtain the mapped temporal feature, and the temporal residual is set to the spatial feature to obtain the mapped spatial feature; The time-mapping feature and the space-mapping feature are fused to obtain the spatiotemporal fusion feature.

5. The method according to claim 1, characterized in that, The geological parameters include at least one of the following: rock type, permeability, porosity, and geostress.

6. The method according to claim 1, characterized in that, The historical data on rock fracture includes at least one of the following: location of rock fracture, time of rock fracture, fracture scale, and fracture type.

7. A device for predicting the risk of underground rock fracturing in carbon dioxide sequestration projects, characterized in that, include: The first acquisition module is used to acquire time-series pressure data of carbon dioxide from multiple monitoring points within the carbon dioxide storage area. The first processing module is used to perform feature analysis on the carbon dioxide time-series pressure data to obtain spatial and temporal features. The second processing module is used to fuse the spatial features and the temporal features to obtain spatiotemporal fusion features; The second acquisition module is used to acquire geological parameters and historical rock fracture data of the carbon dioxide storage area; The third processing module is used to perform feature engineering processing on the geological parameters and the rock fracture history data to obtain geomechanical characteristics. The fourth processing module is used to input the spatiotemporal fusion features and the geomechanical features into a pre-trained rock fracture prediction model to obtain rock fracture prediction data; wherein, the rock fracture prediction model is a CNN-LSTM hybrid model.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 6.

Images