Early warning method, system and terminal based on fault instability evaluation model

Abstract

The invention discloses an early warning method, system and terminal based on a fault instability evaluation model, and the method comprises the steps: obtaining crack data collected by a plurality of sensors, carrying out the crack propagation analysis of a rock according to the crack data, and obtaining the comprehensive crack multi-modal data; inputting the fracture multi-modal data into a fault instability evaluation model, and coupling the fracture multi-modal data by the fault instability evaluation model to generate predicted fracture parameters; and when the predicted crack parameter reaches a target early warning threshold, matching a corresponding early warning level according to the predicted crack parameter to obtain a target early warning level, and taking a corresponding response measure according to the target early warning level. According to sensor data, cracks are analyzed, multi-modal data are generated, the instability risk is evaluated, the early warning level is matched, corresponding response measures are taken, fracturing related parameters are rapidly adjusted, valuable data resources are provided, and potential risks such as resource consumption or geological disaster aggravation caused by excessive fracturing are avoided.

Description

Technical Field

[0001] The present invention relates to the field of crack detection technology, and in particular to an early warning method, system, terminal and computer-readable storage medium based on a fault instability assessment model. Background Art

[0002] In recent years, in the process of oil and gas well exploitation, in order to maximize production, hydraulic fracturing methods are usually used to create and expand cracks to promote oil and gas flow; however, existing technologies often do not adequately monitor the stability of underground faults during exploitation, and focus more on post-disaster remediation, while lacking early assessment and early warning of potential disasters; for example, existing technologies usually reduce the injection flow rate or even stop construction after detecting small or medium-sized earthquake events, but the losses caused at this time are irreversible. Potential dangers such as resource consumption or aggravated geological disasters due to excessive fracturing have become issues that need to be urgently addressed.

[0003] Therefore, the prior art still needs to be improved and developed. Summary of the invention

[0004] The main purpose of the present invention is to provide an early warning method, system and computer-readable storage medium based on a fault instability assessment model, aiming to solve the problems of insufficient monitoring of underground fault stability and lack of early assessment and early warning of potential disasters in the prior art.

[0005] To achieve the above object, the present invention provides an early warning method based on a fault instability assessment model, the early warning method based on a fault instability assessment model comprising the following steps:

[0006] Acquire fracture data collected by multiple sensors, perform fracture propagation analysis on rocks based on the fracture data, and obtain comprehensive fracture multimodal data;

[0007] Inputting the fracture multimodal data into a fault instability assessment model, wherein the fault instability assessment model couples the fracture multimodal data to generate predicted fracture parameters;

[0008] When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level, and corresponding response measures are taken according to the target warning level.

[0009] Optionally, the early warning method based on the fault instability assessment model, wherein the step of acquiring the crack data collected by multiple sensors and performing crack extension analysis on the rock according to the crack data to obtain comprehensive crack multimodal data, specifically includes:

[0010] Acquire strain sensors, temperature sensors, acoustic emission sensors and displacement sensors to collect crack data on the rock surface, and perform crack extension analysis on the rock based on the crack data to obtain stress, strain, temperature information and crack geometry parameters of the crack structure;

[0011] Comprehensive fracture multimodal data are generated according to the stress, the strain, the temperature information and the fracture geometric parameters.

[0012] Optionally, the early warning method based on the fault instability assessment model, wherein the multimodal data of the cracks is input into the fault instability assessment model, and the fault instability assessment model couples the multimodal data of the cracks to generate predicted crack parameters, specifically includes:

[0013] Acquire structured values ​​of the multimodal data, and input the structured values ​​and the crack geometric parameters into a fault instability assessment model;

[0014] The fault instability assessment model fuses the structural values ​​and the crack geometric parameters through a coding layer to generate fusion feature information;

[0015] The fault instability assessment model maps the data distribution of the fused feature information through a transformation layer to generate predicted crack parameters.

[0016] Optionally, the early warning method based on the fault instability assessment model, wherein the multimodal data of the cracks is input into the fault instability assessment model, the fault instability assessment model couples the multimodal data of the cracks to generate predicted crack parameters, and then further comprises:

[0017] Performing difference calculation based on the predicted crack parameters and the parameters of the fault instability assessment model to generate a loss value;

[0018] The loss value is transmitted back to the fault instability assessment model to optimize parameters and obtain optimal parameters.

[0019] Optionally, the early warning method based on the fault instability assessment model, wherein the reverse transmission of the loss value to the fault instability assessment model to optimize parameters to obtain optimal parameters, specifically includes:

[0020] Reversely transmitting the loss value to the fault instability assessment model;

[0021] The predicted crack parameters are compared with the actual crack parameters to obtain an evaluation result, and the parameters of the fault instability evaluation model are optimized according to the evaluation result and the loss value to obtain optimal parameters.

[0022] Optionally, the step of inputting the fracture multimodal data into a fault instability assessment model, wherein the fault instability assessment model couples the fracture multimodal data to generate predicted fracture parameters, further comprising:

[0023] Capturing time based on a recurrent neural network and a long short-term memory network to generate time series data, and processing the time series data according to preset rules to obtain time series prediction rules;

[0024] Inputting the time series prediction rule into the fault instability assessment model for training to obtain the trained fault instability assessment model;

[0025] The trained fault instability assessment model is used to predict the crack multimodal data.

[0026] Optionally, in the early warning method based on the fault instability assessment model, the target early warning level includes a mild early warning, a moderate early warning and a severe early warning;

[0027] When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level, and corresponding response measures are taken according to the target warning level, specifically including:

[0028] Dynamically adjust the warning threshold based on the adaptive algorithm to obtain the target warning threshold;

[0029] When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level;

[0030] When the target warning level is a mild warning, relevant personnel will be notified to strengthen daily inspections and data monitoring in the monitoring area;

[0031] When the target warning level is a moderate warning, relevant personnel are notified to conduct surveys and assessments of the monitoring area;

[0032] When the target warning level is a severe warning, emergency reinforcement measures are taken or the use of the rock area is suspended to ensure the safety of personnel and structures.

[0033] In addition, to achieve the above-mentioned purpose, the present invention further provides an early warning system based on a fault instability assessment model, wherein the early warning system based on a fault instability assessment model:

[0034] A fracture multimodal data acquisition module is used to acquire fracture data collected by multiple sensors, and to perform fracture extension analysis on the rock according to the fracture data to obtain comprehensive fracture multimodal data;

[0035] A crack parameter prediction module, used for inputting the crack multimodal data into a fault instability assessment model, wherein the fault instability assessment model couples the crack multimodal data to generate predicted crack parameters;

[0036] The crack parameter warning module is used to match the corresponding warning level according to the predicted crack parameter when the predicted crack parameter reaches the target warning threshold, obtain the target warning level, and take corresponding response measures according to the target warning level.

[0037] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores an early warning program based on a fault instability assessment model, and when the early warning program based on the fault instability assessment model is executed by a processor, the steps of the early warning method based on the fault instability assessment model as described above are implemented.

[0038] In the present invention, the crack data collected by multiple sensors are obtained, and the crack extension analysis of the rock is performed according to the crack data to obtain comprehensive crack multimodal data; the crack multimodal data is input into the fault instability assessment model, and the fault instability assessment model couples the crack multimodal data to generate predicted crack parameters; when the predicted crack parameters reach the target warning threshold, the corresponding warning level is matched according to the predicted crack parameters to obtain the target warning level, and corresponding response measures are taken according to the target warning level. The present invention analyzes cracks according to sensor data, generates multimodal data, assesses the instability risk and matches the warning level to take corresponding response measures, so as to quickly adjust the fracturing related parameters to provide valuable data resources, so as to avoid potential dangers such as excessive fracturing leading to resource consumption or aggravation of geological disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a flow chart of a preferred embodiment of the early warning method based on the fault instability assessment model of the present invention;

[0040] Figure 2 It is an extended physical experiment flow chart of a preferred embodiment of the early warning method based on the fault instability assessment model of the present invention;

[0041] Figure 3 It is a principle diagram of a fault instability assessment model of a preferred embodiment of the early warning method based on the fault instability assessment model of the present invention;

[0042] Figure 4 It is a structural diagram of a preferred embodiment of the early warning system based on the fault instability assessment model of the present invention;

[0043] Figure 5 It is a structural diagram of a preferred embodiment of the terminal of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0045] In the existing technology, the stability of underground faults is often insufficiently monitored during mining, and more attention is paid to post-disaster remediation, but there is a lack of early assessment and early warning of potential disasters. If there is no early warning, excessive fracturing will lead to resource consumption or aggravation of geological disasters. Therefore, there is a need for an early warning method based on a fault instability assessment model, which solves the problem of insufficient monitoring of underground fault stability and lack of early assessment and early warning of potential disasters in the existing technology.

[0046] The early warning method based on the fault instability assessment model described in the preferred embodiment of the present invention is as follows: Figure 1 As shown, the early warning method based on the fault instability assessment model includes the following steps:

[0047] Step S10: Acquire the fracture data collected by multiple sensors, perform fracture extension analysis on the rock according to the fracture data, and obtain comprehensive fracture multimodal data.

[0048] Specifically, strain sensors, temperature sensors, acoustic emission sensors and displacement sensors are used to collect crack data on the rock surface, and crack extension analysis is performed on the rock based on the crack data to obtain stress, strain, temperature information and crack geometric parameters of the crack structure. Strain sensors, such as resistance strain gauges and fiber Bragg grating strain sensors, are installed on the surface or inside of the rock structure to measure changes in structural strain caused by cracks. Temperature sensors are used to monitor ambient temperature and rock sample temperature distribution, compensate for the influence of temperature on the measurement results of other sensors, and analyze the correlation between temperature changes and crack development. Acoustic emission sensors monitor the internal state and mechanical state changes of rocks during stress through elastic waves emitted by rocks to obtain rock damage characteristics and crack evolution trends. Displacement sensors are used to obtain displacement information during crack extension and provide deformation of rock samples under various conditions such as strain, temperature and seepage. Comprehensive crack multimodal data are generated based on the stress, strain, temperature information and crack geometric parameters (physical experiments simulating crack extension of rock samples are carried out in a thermal-fluid-solid coupling environment. During the experiment, a variety of sensors are used to monitor and record rock states and changes to obtain comprehensive crack multimodal data).

[0049] like Figure 2As shown, the fracture data obtained by various types of sensors installed in the extended physical experiment are used, and the fracture extension analysis of the rock is carried out according to the fracture data to obtain the stress, strain, temperature information and fracture geometry parameters of the fracture structure. The fault instability assessment model (model design, model training and time series prediction functions) is used to make predictions, and the predicted fracture parameters (future fracture parameters) are generated. The sound, fluid, temperature data and future fracture parameters are monitored and fed back in real time, and the future fracture parameters are warned by setting the warning threshold and the warning release mechanism.

[0050] Step S20: input the fracture multimodal data into a fault instability assessment model, and the fault instability assessment model couples the fracture multimodal data to generate predicted fracture parameters.

[0051] Specifically, Figure 3As shown, based on the recurrent neural network and the long short-term memory network, time is captured to generate time series data (for example, acoustic emission, fluid pressure and temperature), and the time series data is processed according to the preset rules to obtain the time series prediction rules; the time series prediction rules are input into the fault instability assessment model for training to obtain the trained fault instability assessment model; the trained fault instability assessment model is used to predict the multimodal data of the crack (for example, the acoustic emission signal at a certain moment may be related to the signal a few minutes or hours ago. By capturing this temporal correlation, the acoustic emission signal of the next 5 minutes, 1 hour or more can be predicted. The above prediction can help understand the expansion of the crack and thus evaluate the stability of the crack); obtain the structured values ​​of the multimodal data (the structured values ​​include acoustic emission, fluid pressure, temperature, crack length and crack aperture), and input the structured values ​​and the crack geometric parameters into the fault instability assessment model, for example: for structured data (such as acoustic emission, fluid pressure, temperature, crack length, crack aperture), use the fully connected The fault instability assessment model uses a convolution layer as input, and a convolution layer as input for the crack extension image of the crack geometric parameters; the fault instability assessment model fuses the structured numerical value and the crack geometric parameters (crack extension image) through an encoding layer to generate fused feature information (unified feature); the fault instability assessment model maps the data distribution of the fused feature information through a transformation layer (the transformation layer consists of multiple reversible bijective functions, which map the input data from one distribution to another distribution, and calculate the Jacobian determinant of the transformation at the same time to adjust the probability density to ensure that the total probability before and after the transformation remains 1 to maintain the consistency of the probability. Adding a coupling layer between the transformation layers can further enhance the flexibility and expression ability of the model) to generate predicted crack parameters, wherein the coupling layer is a specific type of layer in the transformation layer. The coupling layer represents the input feature by segmenting it, applies an affine transformation to a part of it, and then re-merges the transformed part with the untransformed part. The input of the coupling layer is the intermediate result of the transformation layer, that is, the output of the previous transformation layer (which may be another coupling layer or other type of transformation layer).

[0052] Furthermore, a difference calculation is performed based on the predicted crack parameters and the parameters of the fault instability assessment model to generate a loss value; the loss value is transmitted back to the fault instability assessment model to optimize the parameters and obtain the optimal parameters (in each training cycle, the model calculates the output through forward propagation, and the loss is calculated based on the difference between the output and the true value of the target parameter, and then back propagation is performed to optimize the parameters, and the model performance is regularly monitored on the validation set to prevent overfitting), and the loss value is transmitted back to the fault instability assessment model; the predicted crack parameters are compared with the actual crack parameters (the real sample is the real value of the crack parameter corresponding to the input feature) to obtain an evaluation result, and the parameters of the fault instability assessment model are optimized based on the evaluation result and the loss value to obtain the optimal parameters (based on the difference between the predicted crack parameters and the true value, and the model hyperparameters are continuously adjusted according to the evaluation result).

[0053] The process of model training is to divide the processed data into training set, validation set and test set, based on the order of time series. For example, the earliest multimodal data is used as the training set, the middle data is used as the validation set, and the latest data is used as the test set. This ensures that the model will not see future (test set) data during training, and maintains the temporal nature of the data, that is, multimodal data in the corresponding time. Furthermore, during the training process, the negative log-likelihood loss function is used to evaluate the similarity between the samples generated by the model and the real samples, and optimization algorithms such as Adam or SGD are selected for parameter optimization (using the gradient calculated by backpropagation and the selected optimization algorithm to update the parameters of the model, and then the optimization algorithm will adjust the parameter values ​​according to the gradient and some other parameters, such as: learning rate). After multiple cycles of iteration, until the preset termination condition is reached, the model performance is finally evaluated using the test set.

[0054] Step S30: When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level, and corresponding response measures are taken according to the target warning level.

[0055] Specifically, the target warning level includes mild warning, moderate warning and severe warning; the warning threshold is dynamically adjusted based on an adaptive algorithm (for example, if the displacement change of the crack exceeds the preset allowable change, the monitoring system will adaptively adjust the data collection interval to monitor the crack changes more frequently) to obtain the target warning threshold; when the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level; when the target warning level is mild warning, relevant personnel are notified to strengthen daily inspections and data monitoring of the monitoring area; when the target warning level is moderate warning, relevant personnel are notified to conduct surveys and assessments of the monitoring area; when the target warning level is severe warning, emergency reinforcement measures are taken for the rock area or the use is suspended to ensure the safety of personnel and structures.

[0056] When the crack assessment results reach the warning threshold, the system sends warning information to relevant personnel (such as engineering management personnel, maintenance personnel, technical experts, etc.) through SMS, email, system pop-up windows and other means at the same time; the SMS content is concise and clear, including key crack information and warning levels; the email details the crack situation, assessment basis and recommended measures; the system pop-up window is highlighted on the monitoring system interface to remind on-site operators to pay attention in time.

[0057] Beneficial effects of the present invention:

[0058] (1) The present invention improves the comprehensiveness and accuracy of data by integrating data from multiple sensors. Compared with the data collection method of a single sensor, this multimodal data fusion technology can more comprehensively reflect the actual situation of the structure, thereby improving the accuracy of crack assessment.

[0059] (2) The present invention uses real-time crack monitoring technology to timely capture the dynamic changes of cracks, improve the response speed to potential instability risks, and enhance the ability to deal with emergencies.

[0060] (3) The present invention automatically generates response measures based on real-time monitoring and risk assessment results, which provide practical decision-making basis for relevant personnel. This decision support system significantly improves the efficiency of responding to emergencies and reduces the blindness and uncertainty in the manual decision-making process.

[0061] (4) By identifying the potential crack expansion and instability risks in advance, the present invention can take preventive measures to reduce the possible economic losses and casualties, enhance the safety of the structure, and ensure the safety of personnel and the surrounding environment.

[0062] Furthermore, if Figure 4As shown, based on the above-mentioned early warning method based on the fault instability assessment model, the present invention also provides an early warning system based on the fault instability assessment model, wherein the early warning system based on the fault instability assessment model includes:

[0063] A fracture multimodal data acquisition module 51 is used to acquire fracture data collected by multiple sensors, and to perform fracture extension analysis on the rock according to the fracture data to obtain comprehensive fracture multimodal data;

[0064] A crack parameter prediction module 52, used for inputting the crack multimodal data into a fault instability assessment model, wherein the fault instability assessment model couples the crack multimodal data to generate predicted crack parameters;

[0065] The crack parameter warning module 53 is used to match the corresponding warning level according to the predicted crack parameter when the predicted crack parameter reaches the target warning threshold, obtain the target warning level, and take corresponding response measures according to the target warning level.

[0066] Furthermore, if Figure 5 As shown, based on the above-mentioned early warning method and system based on the fault instability assessment model, the present invention also provides a terminal accordingly, and the terminal includes a processor 10, a memory 20 and a display 30. Figure 5 Only some components of the terminal are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0067] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory of the terminal. In other embodiments, the memory 20 may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal. Further, the memory 20 may also include both an internal storage unit of the terminal and an external storage device. The memory 20 is used to store application software and various types of data installed on the terminal, such as the program code of the installation terminal. The memory 20 may also be used to temporarily store data that has been output or is to be output. In one embodiment, an early warning program 40 based on a fault instability assessment model is stored on the memory 20, and the early warning program 40 based on the fault instability assessment model can be executed by the processor 10, thereby realizing the early warning method based on the fault instability assessment model in the present application.

[0068] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run the program code or process data stored in the memory 20, such as executing the early warning method based on the fault instability assessment model.

[0069] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 30 is used to display information on the terminal and to display a visual user interface. The terminals communicate with each other via a system bus.

[0070] In one embodiment, when the processor 10 executes the early warning program 40 based on the fault instability assessment model in the memory 20, the following steps are implemented:

[0071] Acquire fracture data collected by multiple sensors, perform fracture propagation analysis on rocks based on the fracture data, and obtain comprehensive fracture multimodal data;

[0072] Inputting the fracture multimodal data into a fault instability assessment model, wherein the fault instability assessment model couples the fracture multimodal data to generate predicted fracture parameters;

[0073] When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level, and corresponding response measures are taken according to the target warning level.

[0074] The method of acquiring the fracture data collected by multiple sensors and performing fracture extension analysis on the rock according to the fracture data to obtain comprehensive fracture multimodal data specifically includes:

[0075] Acquire strain sensors, temperature sensors, acoustic emission sensors and displacement sensors to collect crack data on the rock surface, and perform crack extension analysis on the rock based on the crack data to obtain stress, strain, temperature information and crack geometry parameters of the crack structure;

[0076] Comprehensive fracture multimodal data are generated according to the stress, the strain, the temperature information and the fracture geometric parameters.

[0077] The step of inputting the multimodal data of the cracks into a fault instability assessment model, wherein the fault instability assessment model couples the multimodal data of the cracks to generate predicted crack parameters, specifically includes:

[0078] Acquire structured values ​​of the multimodal data, and input the structured values ​​and the crack geometric parameters into a fault instability assessment model;

[0079] The fault instability assessment model fuses the structural values ​​and the crack geometric parameters through a coding layer to generate fusion feature information;

[0080] The fault instability assessment model maps the data distribution of the fused feature information through a transformation layer to generate predicted crack parameters.

[0081] The method further comprises: inputting the multimodal data of the cracks into a fault instability assessment model, wherein the fault instability assessment model couples the multimodal data of the cracks to generate predicted crack parameters, and then:

[0082] Performing difference calculation based on the predicted crack parameters and the parameters of the fault instability assessment model to generate a loss value;

[0083] The loss value is transmitted back to the fault instability assessment model to optimize parameters and obtain optimal parameters.

[0084] The reverse transmission of the loss value to the fault instability assessment model to optimize parameters to obtain optimal parameters specifically includes:

[0085] Reversely transmitting the loss value to the fault instability assessment model;

[0086] The predicted crack parameters are compared with the actual crack parameters to obtain an evaluation result, and the parameters of the fault instability evaluation model are optimized according to the evaluation result and the loss value to obtain optimal parameters.

[0087] The step of inputting the multimodal data of the cracks into a fault instability assessment model, wherein the fault instability assessment model couples the multimodal data of the cracks to generate predicted crack parameters, further comprising:

[0088] Capturing time based on a recurrent neural network and a long short-term memory network to generate time series data, and processing the time series data according to preset rules to obtain time series prediction rules;

[0089] Inputting the time series prediction rule into the fault instability assessment model for training to obtain the trained fault instability assessment model;

[0090] The trained fault instability assessment model is used to predict the crack multimodal data.

[0091] The target warning levels include mild warning, moderate warning and severe warning;

[0092] When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level, and corresponding response measures are taken according to the target warning level, specifically including:

[0093] Dynamically adjust the warning threshold based on the adaptive algorithm to obtain the target warning threshold;

[0094] When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level;

[0095] When the target warning level is a mild warning, relevant personnel will be notified to strengthen daily inspections and data monitoring in the monitoring area;

[0096] When the target warning level is a moderate warning, relevant personnel are notified to conduct surveys and assessments of the monitoring area;

[0097] When the target warning level is a severe warning, emergency reinforcement measures are taken or the use of the rock area is suspended to ensure the safety of personnel and structures.

[0098] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores an early warning program based on a fault instability assessment model, and when the early warning program based on the fault instability assessment model is executed by a processor, the steps of the early warning method based on the fault instability assessment model as described above are implemented.

[0099] In summary, the present invention provides an early warning method, system and terminal based on a fault instability assessment model, the method comprising: obtaining crack data collected by multiple sensors, performing crack extension analysis on rocks according to the crack data, and obtaining comprehensive crack multimodal data; inputting the crack multimodal data into the fault instability assessment model, the fault instability assessment model couples the crack multimodal data, and generates predicted crack parameters; when the predicted crack parameters reach the target early warning threshold, matching the corresponding early warning level according to the predicted crack parameters, obtaining the target early warning level, and taking corresponding response measures according to the target early warning level. The present invention analyzes cracks according to sensor data, generates multimodal data, assesses the risk of instability, matches the early warning level and takes corresponding response measures, so as to quickly adjust the parameters related to fracturing and provide valuable data resources, so as to avoid potential dangers such as excessive fracturing leading to resource consumption or aggravation of geological disasters.

[0100] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or terminal system including the element.

[0101] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the program can be stored in a computer-readable storage medium that can be read by a computer, and the program can include the processes of the above-mentioned method embodiments when executed. The computer-readable storage medium can be a memory, a disk, an optical disk, etc.

[0102] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. An early warning method based on a fault instability assessment model, characterized in that: The early warning method based on the fault instability assessment model includes: Acquire fracture data collected by multiple sensors, perform fracture propagation analysis on rocks based on the fracture data, and obtain comprehensive fracture multimodal data; Inputting the fracture multimodal data into a fault instability assessment model, wherein the fault instability assessment model couples the fracture multimodal data to generate predicted fracture parameters; When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level, and corresponding response measures are taken according to the target warning level.

2. The early warning method based on the fault instability assessment model according to claim 1 is characterized in that: The method of acquiring the fracture data collected by multiple sensors and performing fracture extension analysis on the rock according to the fracture data to obtain comprehensive fracture multimodal data specifically includes: Acquire strain sensors, temperature sensors, acoustic emission sensors and displacement sensors to collect crack data on the rock surface, and perform crack extension analysis on the rock based on the crack data to obtain stress, strain, temperature information and crack geometry parameters of the crack structure; Comprehensive fracture multimodal data are generated according to the stress, the strain, the temperature information and the fracture geometric parameters.

3. The early warning method based on the fault instability assessment model according to claim 2 is characterized in that: The step of inputting the multi-modal data of the cracks into a fault instability assessment model, wherein the fault instability assessment model couples the multi-modal data of the cracks to generate predicted crack parameters, specifically includes: Acquire structured values ​​of the multimodal data, and input the structured values ​​and the crack geometric parameters into a fault instability assessment model; The fault instability assessment model fuses the structural values ​​and the crack geometric parameters through a coding layer to generate fusion feature information; The fault instability assessment model maps the data distribution of the fused feature information through a transformation layer to generate predicted crack parameters.

4. The early warning method based on the fault instability assessment model according to claim 1 is characterized in that: The method further comprises: inputting the multi-modal data of the cracks into a fault instability assessment model, wherein the fault instability assessment model couples the multi-modal data of the cracks to generate predicted crack parameters, and then: Performing difference calculation based on the predicted crack parameters and the parameters of the fault instability assessment model to generate a loss value; The loss value is transmitted back to the fault instability assessment model to optimize parameters and obtain optimal parameters.

5. The early warning method based on the fault instability assessment model according to claim 4 is characterized in that: The reverse transmission of the loss value to the fault instability assessment model to optimize parameters to obtain optimal parameters specifically includes: Reversely transmitting the loss value to the fault instability assessment model; The predicted crack parameters are compared with the actual crack parameters to obtain an evaluation result, and the parameters of the fault instability evaluation model are optimized according to the evaluation result and the loss value to obtain optimal parameters.

6. The early warning method based on the fault instability assessment model according to claim 1 is characterized in that: The step of inputting the multi-modal data of the cracks into a fault instability assessment model, wherein the fault instability assessment model couples the multi-modal data of the cracks to generate predicted crack parameters, further comprising: Capturing time based on a recurrent neural network and a long short-term memory network to generate time series data, and processing the time series data according to preset rules to obtain time series prediction rules; Inputting the time series prediction rule into the fault instability assessment model for training to obtain the trained fault instability assessment model; The trained fault instability assessment model is used to predict the crack multimodal data.

7. The early warning method based on the fault instability assessment model according to claim 1 is characterized in that: The target warning levels include mild warning, moderate warning and severe warning; When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level, and corresponding response measures are taken according to the target warning level, specifically including: Dynamically adjust the warning threshold based on the adaptive algorithm to obtain the target warning threshold; When the predicted crack parameter reaches the target warning threshold, the corresponding warning level is matched according to the predicted crack parameter to obtain the target warning level; When the target warning level is a mild warning, relevant personnel will be notified to strengthen daily inspections and data monitoring in the monitoring area; When the target warning level is a moderate warning, relevant personnel are notified to conduct surveys and assessments of the monitoring area; When the target warning level is a severe warning, emergency reinforcement measures are taken or the use of the rock area is suspended to ensure the safety of personnel and structures.

8. An early warning system based on a fault instability assessment model, characterized in that: The early warning system based on the fault instability assessment model includes: A fracture multimodal data acquisition module is used to acquire fracture data collected by multiple sensors, and to perform fracture extension analysis on rocks based on the fracture data to obtain comprehensive fracture multimodal data; A crack parameter prediction module, used for inputting the crack multimodal data into a fault instability assessment model, wherein the fault instability assessment model couples the crack multimodal data to generate predicted crack parameters; The crack parameter warning module is used to match the corresponding warning level according to the predicted crack parameter when the predicted crack parameter reaches the target warning threshold, obtain the target warning level, and take corresponding response measures according to the target warning level.

9. A terminal, characterized in that: The terminal includes: a memory, a processor, and an early warning program based on a fault instability assessment model stored in the memory and executable on the processor. When the early warning program based on the fault instability assessment model is executed by the processor, the steps of the early warning method based on the fault instability assessment model as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an early warning program based on a fault instability assessment model, and when the early warning program based on a fault instability assessment model is executed by a processor, the steps of the early warning method based on a fault instability assessment model as described in any one of claims 1 to 7 are implemented.

Images