A large-inclination coal seam overburden rock breakage migration prediction method and system based on machine learning

Abstract

The application discloses a large-inclination coal seam overburden rock breakage migration prediction method and system based on machine learning, and the method comprises the following steps: S1, collecting geological data of a large-inclination coal seam and performing missing value filling and abnormal value detection on the geological data to obtain pretreatment data; S2, extracting static features from the pretreatment data, collecting mining dynamic features in a mining process, and performing breakage migration simulation based on the static features and the dynamic features; and S3, constructing an overburden rock breakage migration prediction model based on a simulation result, inputting to-be-detected breakage migration data into the overburden rock breakage migration prediction model, outputting a prediction result, and completing the large-inclination coal seam overburden rock breakage migration prediction based on machine learning.

Description

Technical Field

[0001] This invention relates to the field of overburden fracturing and migration prediction technology, specifically to a method and system for predicting overburden fracturing and migration in steeply dipped coal seams based on machine learning. Background Technology

[0002] The fracturing and migration of overlying strata in steeply inclined coal seams refers to the deformation, fracturing, and displacement of the overlying strata due to mining activities as the working face advances during the mining process.

[0003] Existing prediction models may fit data well in specific mining areas or under specific conditions, but their generalization ability in other mining areas or under different mining conditions needs improvement. While using complex models such as deep learning and ensemble learning can improve prediction accuracy, they are computationally expensive. In practical applications, a better balance needs to be found between model complexity and computational efficiency.

[0004] In actual mining operations, timely and accurate prediction of overburden fracturing and migration is crucial for guiding mining operations. However, current prediction models may experience delays in the entire process from data acquisition to result output. Furthermore, while prediction results can provide some reference for mining, actual mining operations require consideration of numerous practical factors, such as limitations of mining equipment and personnel operating habits. Therefore, how to better integrate prediction results with actual mining practices is a pressing issue that needs to be addressed. Summary of the Invention

[0005] To address the aforementioned problems, this invention proposes a machine learning-based method and system for predicting the fracturing and migration of overlying coal seams at steep angles. The method specifically includes:

[0006] Step S1: Collect geological data of steeply dipped coal seams and perform missing value filling and outlier detection on the geological data to obtain preprocessed data;

[0007] Step S2: Extract static features from the preprocessed data, collect dynamic features of the mining process, and perform fracture and migration simulation based on the static and dynamic features;

[0008] Step S3: Construct a strata fracturing and migration prediction model based on simulation results. Input the fracturing and migration data to be detected into the strata fracturing and migration prediction model, and output the prediction results to complete the prediction of strata fracturing and migration of steeply inclined coal seams based on machine learning.

[0009] Optionally, in step S1, the geological data includes coal seam dip angle, coal seam thickness, roof lithology, physical and mechanical parameters of the roof strata, geostress data, and mining depth.

[0010] Optionally, the missing value detection process includes:

[0011] Set the dataset segmentation length and the missing value rate, segment the geological data based on the segmentation length, traverse all segmented geological data, and generate a sample dataset by using the geological data with a missing value rate less than the preset missing value rate as sample data.

[0012] An infill model is constructed based on a deep learning model. The infill model is trained using the sample dataset, MSE, and SIV loss to obtain an optimized infill model.

[0013] The optimized filling model is used to fill in the missing values ​​in the geological data.

[0014] Optionally, the content of the optimized filling model includes:

[0015]

[0016] Where α is the combined weight between the loss functions MSE and SIV. The loss value of MSE, This represents the loss value for SIV.

[0017] Optionally, in step S2, the method for extracting static features specifically includes:

[0018] PCA is used to reduce the dimensionality of the preprocessed data to obtain dimensionality-reduced data.

[0019] The MBK-means clustering algorithm was used to compress a portion of the dimensionality-reduced data to obtain compressed data;

[0020] The other part of the dimensionality-reduced data is labeled using the uncertain sampling method in AL to obtain labeled data;

[0021] The CNN is trained using the compressed data and the labeled data;

[0022] Use a trained CNN model to extract static features.

[0023] This invention also discloses a machine learning-based prediction system for the fracturing and migration of overburden in steeply inclined coal seams, the system comprising:

[0024] The data acquisition module is used to collect geological data of steeply inclined coal seams and to fill in missing values ​​and detect outliers in the geological data to obtain preprocessed data.

[0025] The mining simulation module is used to extract static features from the preprocessed data, collect dynamic features of the mining process, and perform fracture and migration simulation based on the static and dynamic features.

[0026] The migration prediction module is used to construct a strata fracturing and migration prediction model based on simulation results. The fracturing and migration data to be detected are input into the strata fracturing and migration prediction model, and the prediction results are output to complete the strata fracturing and migration prediction of steeply inclined coal seams based on machine learning.

[0027] Optionally, the missing value detection process includes:

[0028] Set the dataset segmentation length and the missing value rate, segment the geological data based on the segmentation length, traverse all segmented geological data, and generate a sample dataset by using the geological data with a missing value rate less than the preset missing value rate as sample data.

[0029] An infill model is constructed based on a deep learning model. The infill model is trained using the sample dataset, MSE, and SIV loss to obtain an optimized infill model.

[0030] The optimized filling model is used to fill in the missing values ​​in the geological data.

[0031] Optionally, the content of the optimized filling model includes:

[0032]

[0033] Where α is the combined weight between the loss functions MSE and SIV. The loss value of MSE, This represents the loss value for SIV.

[0034] Optional methods for extracting static features include:

[0035] PCA is used to reduce the dimensionality of the preprocessed data to obtain dimensionality-reduced data.

[0036] The MBK-means clustering algorithm was used to compress a portion of the dimensionality-reduced data to obtain compressed data;

[0037] The other part of the dimensionality-reduced data is labeled using the uncertain sampling method in AL to obtain labeled data;

[0038] The CNN is trained using the compressed data and the labeled data;

[0039] Use a trained CNN model to extract static features.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] This invention fills in missing values ​​using a deep learning model and uses CNTS for outlier detection, solving the problem of numerous missing and outlier values ​​in geological data, improving data integrity and accuracy, and providing a high-quality data foundation for subsequent modeling.

[0042] This invention employs a method combining PCA dimensionality reduction, MBK-means clustering, AL uncertain sampling, and CNN training to extract static features. This method can extract features that are of great significance for predicting overburden fracture and migration from high-dimensional and complex geological data. At the same time, it combines dynamic features during the mining process to simulate fracture and migration, thus achieving an effective fusion of static and dynamic features and solving the problems of inaccurate and incomplete feature extraction.

[0043] This invention combines the physical mechanism of overlying rock fracture and migration to verify the consistency of prediction results, and generates a dynamic trend chart based on the final prediction results. This solves the problems of lack of verification methods and low credibility of prediction results, making the prediction results more scientific and practical. Attached Figure Description

[0044] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart of the method for predicting the fracturing and migration of overlying coal seams at steep angles based on machine learning, as described in an embodiment of the present invention.

[0046] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0047] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] Example 1

[0049] A machine learning-based method for predicting the fracturing and migration of overburden in steeply dip coal seams, such as... Figure 1 As shown, the method includes:

[0050] Step S1: Collect geological data of steeply dipped coal seams and perform missing value filling and outlier detection on the geological data to obtain preprocessed data;

[0051] In step S1, the geological data includes coal seam dip angle, coal seam thickness, roof lithology, physical and mechanical parameters of the roof strata, geostress data, and mining depth.

[0052] The missing value detection process includes:

[0053] Set the dataset segmentation length and the missing value rate, segment the geological data based on the segmentation length, traverse all segmented geological data, and generate a sample dataset by using the geological data with a missing value rate less than the preset missing value rate as sample data.

[0054] An infill model is constructed based on a deep learning model. The infill model is trained using the sample dataset, MSE, and SIV loss to obtain an optimized infill model.

[0055] The optimized filling model is used to fill in the missing values ​​in the geological data.

[0056] Optionally, the content of the optimized filling model includes:

[0057]

[0058] Where α is the combined weight between the loss functions MSE and SIV. The loss value of MSE, This represents the loss value for SIV.

[0059] Based on this embodiment, the specific calculation methods for the two loss functions are as follows:

[0060]

[0061] Where, x L For sample set The data in the sample; remove num from each sample. obs *γ values, where num obs *γ represents the sum of the number of values ​​already filled and the number of observations in the sample. And the data to be removed is defined as x. drop The remaining data is x obs It is worth noting that x drop The portion of the value to be filled is defined as The portion belonging to the observation value is defined as Get x obs Then, it can be input into the model to obtain x. drop Filling results The same It also needs to be done according to x drop The classification methods are divided into and Used for subsequent loss calculations.

[0062] In this embodiment, outlier detection is calculated and implemented using CNTS.

[0063] The goal of reconstruction-based anomaly detection models is to learn an optimized model that can accurately reconstruct normal data samples while failing to effectively reconstruct anomalies (Ruffetal., 2021). Anomaly detection using reconstruction models is based on the assumption that reconstructed data can distinguish between normal and anomalous data.

[0064]

[0065] The effectiveness of a reconstruction model in detecting anomalies depends on its ability to produce small reconstruction errors for normal values ​​and large reconstruction errors for outliers. This will result in a relatively low objective function result. In practical modeling, the labels of anomalies are always unknown, and the objective function is as follows:

[0066]

[0067] The CNTS reconstructor (R) and detector (D) work together to achieve similar goals, ultimately accomplishing their respective objectives. Specifically, the goal of R is to reconstruct normal samples and distinguish them from abnormal samples, while the goal of D is to detect anomalies. These two goals overlap to some extent and can reinforce each other.

[0068] Therefore, the mutual utilization of information between the two networks D and R has the following advantages. The reconstruction error of R can serve as anomaly labels for D during its training process, guiding its training. On the other hand, D can provide R with information about anomalies, reducing the impact of potential outliers on R's training. This separation of the anomaly detection task from the data reconstruction task allows CNTS to be modeled as a multi-objective programming problem. The objective function formula is:

[0069]

[0070] Where D(x) i θ D ) is a parameter of θ D D,R(x) i θ R ) is a parameter of θ R R, and Let D and R represent the losses respectively. It is a category label, if D(x) i θ D The value of ) accounts for all D(x) i θ D The first k% of ) It equals 1, otherwise it equals 0.

[0071] Step S2: Extract static features from the preprocessed data, collect dynamic features of the mining process, and perform fracture and migration simulation based on the static and dynamic features;

[0072] In step S2, the method for extracting static features specifically includes:

[0073] PCA is used to reduce the dimensionality of the preprocessed data to obtain dimensionality-reduced data.

[0074] The MBK-means clustering algorithm was used to compress a portion of the dimensionality-reduced data to obtain compressed data;

[0075] The other part of the dimensionality-reduced data is labeled using the uncertain sampling method in AL to obtain labeled data;

[0076] The CNN is trained using the compressed data and the labeled data;

[0077] Use a trained CNN model to extract static features.

[0078] Specifically: Step 1: Randomly select 2% of the training samples from each class as the initial training set and feed them into C for training; Step 2: Based on the class probabilities provided by the trained CNN, select a certain number of the most informative samples from the remaining unlabeled samples, i.e., the samples with the smallest difference between the best and second-best class probabilities predicted by the CNN, and perform expert annotation; Step 3: Add the difficult-to-classify samples selected in Step 2 to the initial training set as a new training set and feed it into the CNN for training again; Step 4: Iterate through Step 2 and Step 3 until the required number of iterations is reached.

[0079] Principal component analysis (PCA) is used to reduce the dimensionality of multidimensional feature data to a lower-dimensional space by extracting the main feature directions. For example, in the prediction of overburden fracturing and migration, there may be more than ten feature dimensions, such as coal seam dip angle, roof lithology, and mining depth. PCA can reduce these to three or four main feature dimensions, which can explain more than 80% of the variability of the original data.

[0080] In this embodiment, it is important to note that: steeply dipped coal seams refer to coal seams with dip angles between 35 and 55 degrees, and the thickness ranges from thin (1.3m) to extra-thick (8m). This angle range significantly impacts the selection of mining equipment and roadway support schemes. Coal seam thickness is generally between 0.8 and 3 meters, directly determining the mining process and equipment selection. Roof lithology includes different types such as sandstone and shale, and its compressive strength and stability are crucial for roof management planning. Mining depths vary from hundreds to thousands of meters; the greater the depth, the greater the ground pressure and the higher the safety risks. During mining, dynamic characteristics such as gas emission and roof activity intensity are combined with static characteristics to form a temporal characteristic matrix. For example, in a certain mining area, with a coal seam dip angle of 40 degrees, a coal seam thickness of 2.5 meters, a sandstone roof, and a mining depth of 800 meters, the corresponding gas emission exhibits a periodic variation pattern, with a peak gas emission occurring approximately every 100 meters of advance per working face. Clustering algorithms classify time-series feature matrices, and density-based clustering methods can be used. Analysis reveals that areas with similar mining depths often exhibit similar gas emission patterns, allowing mining areas to be categorized according to mining depth. When data distribution characteristics do not meet preset conditions, regression algorithms are needed for correction. For example, if a region has abnormally high gas emission rates, multiple linear regression analysis reveals that the high degree of roof fragmentation in this area leads to an increase in gas channels.

[0081] Static simulation is performed using the 3D finite element method. Static features are input into AnSYS software for simulation, and simulation results are obtained.

[0082] The dynamic simulation was performed using the three-dimensional discrete element method software 3DEC to obtain simulation results of the dynamic features.

[0083] Step S3: Construct a strata fracturing and migration prediction model based on simulation results. Input the fracturing and migration data to be detected into the strata fracturing and migration prediction model, and output the prediction results to complete the prediction of strata fracturing and migration of steeply inclined coal seams based on machine learning.

[0084] In this embodiment, the prediction model uses a vertical prediction method, and the calculation formula is as follows:

[0085] In(D a )=f(M,V,In(D s )

[0086] Among them, D a The data to be detected is the fracture transport data, where M is the anomaly level, V is the fracture transport cycle velocity, and D is the fracture transport period velocity. s This is the fault distance.

[0087] Real-time monitoring data is acquired to extract key features of overburden fracturing and migration. An ensemble method is used to fuse random forest and neural network models to calculate optimized values. Prediction results are regenerated based on these optimized values, and the error is checked against a preset threshold. If the error exceeds the threshold, the neural network parameters are adjusted, and the prediction results are recalculated. The consistency between the prediction results and the physical mechanism of overburden fracturing and migration is verified. If the consistency meets preset conditions, the final prediction result is output. Based on the final prediction result, a dynamic trend map of overburden fracturing and migration is generated.

[0088] Example 2

[0089] A machine learning-based prediction system for the fracturing and migration of overburden in steeply dip coal seams, comprising:

[0090] The data acquisition module is used to collect geological data of steeply inclined coal seams and to fill in missing values ​​and detect outliers in the geological data to obtain preprocessed data.

[0091] Geological data includes coal seam images, coal seam dip angle, coal seam thickness, roof lithology, and mining depth.

[0092] The missing value detection process includes:

[0093] Set the dataset segmentation length and the missing value rate, segment the geological data based on the segmentation length, traverse all segmented geological data, and generate a sample dataset by using the geological data with a missing value rate less than the preset missing value rate as sample data.

[0094] An infill model is constructed based on a deep learning model. The infill model is trained using the sample dataset, MSE, and SIV loss to obtain an optimized infill model.

[0095] The optimized filling model is used to fill in the missing values ​​in the geological data.

[0096] Optionally, the content of the optimized filling model includes:

[0097]

[0098] Where α is the combined weight between the loss functions MSE and SIV. The loss value of MSE, This represents the loss value for SIV.

[0099] Based on this embodiment, the specific calculation methods for the two loss functions are as follows:

[0100]

[0101] Where, x L For sample set The data in the sample; remove num from each sample. obs *γ values, where num obs *γ represents the sum of the number of values ​​already filled and the number of observations in the sample. And the data to be removed is defined as x. drop The remaining data is x obs It is worth noting that x drop The portion of the value to be filled is defined as The portion belonging to the observation value is defined as Get x obs Then, it can be input into the model to obtain x. drop Filling results The same It also needs to be done according to x drop The classification methods are divided into and Used for subsequent loss calculations.

[0102] In this embodiment, outlier detection is calculated and implemented using CNTS.

[0103] The goal of reconstruction-based anomaly detection models is to learn an optimized model that can accurately reconstruct normal data samples while failing to effectively reconstruct anomalies (Ruffetal., 2021). Anomaly detection using reconstruction models is based on the assumption that reconstructed data can distinguish between normal and anomalous data.

[0104]

[0105] The effectiveness of a reconstruction model in detecting anomalies depends on its ability to produce small reconstruction errors for normal values ​​and large reconstruction errors for outliers. This will result in a relatively low objective function result. In practical modeling, the labels of anomalies are always unknown, and the objective function is as follows:

[0106]

[0107] The CNTS reconstructor (R) and detector (D) work together to achieve similar goals, ultimately accomplishing their respective objectives. Specifically, the goal of R is to reconstruct normal samples and distinguish them from abnormal samples, while the goal of D is to detect anomalies. These two goals overlap to some extent and can reinforce each other.

[0108] Therefore, the mutual utilization of information between the two networks D and R has the following advantages. The reconstruction error of R can serve as anomaly labels for D during its training process, guiding its training. On the other hand, D can provide R with information about anomalies, reducing the impact of potential outliers on R's training. This separation of the anomaly detection task from the data reconstruction task allows CNTS to be modeled as a multi-objective programming problem. The objective function formula is:

[0109]

[0110] Where D(x) i θ D ) is a parameter of θ D D,R(x) i θ R ) is a parameter of θ R R, and Let D and R represent the losses respectively. It is a category label, if D(x) i ,θ D The value of ) accounts for all D(x) i ,θ D The first k% of ) It equals 1, otherwise it equals 0.

[0111] The mining simulation module is used to extract static features from the preprocessed data, collect dynamic features of the mining process, and perform fracture and migration simulation based on the static and dynamic features.

[0112] Methods for extracting static features specifically include:

[0113] PCA is used to reduce the dimensionality of the preprocessed data to obtain dimensionality-reduced data.

[0114] The MBK-means clustering algorithm was used to compress a portion of the dimensionality-reduced data to obtain compressed data;

[0115] The other part of the dimensionality-reduced data is labeled using the uncertain sampling method in AL to obtain labeled data;

[0116] The CNN is trained using the compressed data and the labeled data;

[0117] Use a trained CNN model to extract static features.

[0118] Specifically: Step 1: Randomly select 2% of the training samples from each class as the initial training set and feed them into C for training; Step 2: Based on the class probabilities provided by the trained CNN, select a certain number of the most informative samples from the remaining unlabeled samples, i.e., the samples with the smallest difference between the best and second-best class probabilities predicted by the CNN, and perform expert annotation; Step 3: Add the difficult-to-classify samples selected in Step 2 to the initial training set as a new training set and feed it into the CNN for training again; Step 4: Iterate through Step 2 and Step 3 until the required number of iterations is reached.

[0119] Principal component analysis (PCA) is used to reduce the dimensionality of multidimensional feature data to a lower-dimensional space by extracting the main feature directions. For example, in the prediction of overburden fracturing and migration, there may be more than ten feature dimensions, such as coal seam dip angle, roof lithology, and mining depth. PCA can reduce these to three or four main feature dimensions, which can explain more than 80% of the variability of the original data.

[0120] In this embodiment, it is important to note that: steeply dipped coal seams refer to coal seams with dip angles between 35 and 55 degrees, and the thickness ranges from thin (1.3m) to extra-thick (8m). This angle range significantly impacts the selection of mining equipment and roadway support schemes. Coal seam thickness is generally between 0.8 and 3 meters, directly determining the mining process and equipment selection. Roof lithology includes different types such as sandstone and shale, and its compressive strength and stability are crucial for roof management planning. Mining depths vary from hundreds to thousands of meters; the greater the depth, the greater the ground pressure and the higher the safety risks. During mining, dynamic characteristics such as gas emission and roof activity intensity are combined with static characteristics to form a temporal characteristic matrix. For example, in a certain mining area, with a coal seam dip angle of 40 degrees, a coal seam thickness of 2.5 meters, a sandstone roof, and a mining depth of 800 meters, the corresponding gas emission exhibits a periodic variation pattern, with a peak gas emission occurring approximately every 100 meters of advance per working face. Clustering algorithms classify time-series feature matrices, and density-based clustering methods can be used. Analysis reveals that areas with similar mining depths often exhibit similar gas emission patterns, allowing mining areas to be categorized according to mining depth. When data distribution characteristics do not meet preset conditions, regression algorithms are needed for correction. For example, if a region has abnormally high gas emission rates, multiple linear regression analysis reveals that the high degree of roof fragmentation in this area leads to an increase in gas channels.

[0121] The migration prediction module is used to construct a strata fracturing and migration prediction model based on simulation results. The fracturing and migration data to be detected are input into the strata fracturing and migration prediction model, and the prediction results are output to complete the strata fracturing and migration prediction of steeply inclined coal seams based on machine learning.

[0122] Real-time monitoring data is acquired to extract key features of overburden fracturing and migration. An ensemble method is used to fuse random forest and neural network models to calculate optimized values. Prediction results are regenerated based on these optimized values, and the error is checked against a preset threshold. If the error exceeds the threshold, the neural network parameters are adjusted, and the prediction results are recalculated. The consistency between the prediction results and the physical mechanism of overburden fracturing and migration is verified. If the consistency meets preset conditions, the final prediction result is output. Based on the final prediction result, a dynamic trend map of overburden fracturing and migration is generated.

[0123] Example 3

[0124] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the machine learning-based method for predicting the fracturing and migration of overlying coal seams at steep inclinations as described in any of the above embodiments.

[0125] Figure 2 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0126] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0127] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0128] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0129] The communication interface 1040 is used to connect the communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB (Universal Serial Bus), network cable, etc.) or wireless means (such as mobile network, WIFI (Wireless Fidelity), Bluetooth, etc.).

[0130] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0131] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0132] The system described in the above embodiments is used to implement the corresponding machine learning-based method for predicting the fracturing and migration of overlying coal seams at steep angles in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0133] Example 4

[0134] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the machine learning-based method for predicting the fracturing and migration of overlying coal seams at steep inclinations as described in any of the above embodiments.

[0135] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0136] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the machine learning-based method for predicting the fracturing and migration of overlying coal seams at steep angles as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0137] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this disclosure (including the claims) is limited to these examples; within the framework of this disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this disclosure as described above, which are not provided in detail for the sake of brevity.

[0138] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this disclosure, the provided drawings may or may not show well-known power / ground connections to integrated circuit (IC) chips and other components. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this disclosure, and this also takes into account the fact that the details of implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this disclosure will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuitry) have been set forth to describe exemplary embodiments of this disclosure, it will be apparent to those skilled in the art that the embodiments of this disclosure may be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0139] Although this disclosure has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0140] Therefore, the units of the various examples described in the embodiments of this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0141] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for predicting the fracturing and migration of overlying strata in steeply dip coal seams based on machine learning, characterized in that, The method includes: Step S1: Collect geological data of steeply dipped coal seams and perform missing value filling and outlier detection on the geological data to obtain preprocessed data; The missing value imputation process includes: Set the dataset segmentation length and the missing value rate, segment the geological data based on the segmentation length, traverse all segmented geological data, and generate a sample dataset by using the geological data with a missing value rate less than the preset missing value rate as sample data. An infill model is constructed based on a deep learning model. The infill model is trained using the sample dataset, MSE, and SIV loss to obtain an optimized infill model. The optimized filling model is used to fill in the missing value data in the geological data; The optimized filling model includes: Where α is the combined weight between the loss functions MSE and SIV. The loss value of MSE, This represents the loss value for SIV; Step S2: Extract static features from the preprocessed data, collect dynamic features of the mining process, and perform fracture and migration simulation based on the static and dynamic features; In step S2, the method for extracting static features specifically includes: PCA is used to reduce the dimensionality of the preprocessed data to obtain dimensionality-reduced data. The MBK-means clustering algorithm was used to compress a portion of the dimensionality-reduced data to obtain compressed data; The other part of the dimensionality-reduced data is labeled using the uncertain sampling method in AL to obtain labeled data; The CNN is trained using the compressed data and the labeled data; Extract static features using a trained CNN model; Step S3: Construct a strata fracturing and migration prediction model based on simulation results. Input the fracturing and migration data to be detected into the strata fracturing and migration prediction model, and output the prediction results to complete the prediction of strata fracturing and migration of steeply inclined coal seams based on machine learning.

2. The method for predicting the fracturing and migration of overlying strata in steeply dipped coal seams based on machine learning according to claim 1, characterized in that, In step S1, the geological data includes coal seam dip angle, coal seam thickness, roof lithology, physical and mechanical parameters of roof strata, geostress data, and mining depth.

3. A machine learning-based prediction system for the fracturing and migration of overlying coal seams at steep inclinations, the system being used to implement the prediction method described in any one of claims 1-2, characterized in that the system... include: The data acquisition module is used to collect geological data of steeply inclined coal seams and to fill in missing values ​​and detect outliers in the geological data to obtain preprocessed data. The mining simulation module is used to extract static features from the preprocessed data, collect dynamic features of the mining process, and perform fracture and migration simulation based on the static and dynamic features. The migration prediction module is used to construct a strata fracturing and migration prediction model based on simulation results. The fracturing and migration data to be detected are input into the strata fracturing and migration prediction model, and the prediction results are output to complete the strata fracturing and migration prediction of steeply inclined coal seams based on machine learning.

Images