Method and device for predicting underground water level of coal mine area and electronic equipment

Abstract

The invention provides a coal mine area underground water level prediction method and device and electronic equipment. The method comprises the steps that first monitoring data corresponding to a target pumping well and second monitoring data corresponding to a target monitoring well are obtained; performing wavelet packet decomposition on the first monitoring data to obtain a first high-frequency signal and a first low-frequency signal corresponding to the first monitoring data; performing wavelet packet decomposition on the second monitoring data to obtain a second high-frequency signal and a second low-frequency signal corresponding to the second monitoring data; and according to the first high-frequency signal, the first low-frequency signal, the second high-frequency signal and the second low-frequency signal, determining predicted water level data corresponding to the target monitoring well by using a water level prediction model obtained by pre-training. The underground water level of the coal mine area is accurately predicted.

Description

Technical Field

[0001] This disclosure relates to the field of data processing, and in particular to a method, apparatus and electronic equipment for predicting groundwater levels in coal mining areas. Background Technology

[0002] Groundwater level fluctuations in coal mining areas are a crucial indicator for ensuring safe production and protecting the regional ecological environment. Particularly with the significant westward shift of coal mining, concentrated mining activities can lead to abnormal fluctuations in groundwater levels, resulting in problems such as mine water inrush, surface subsidence, and water shortages, posing serious threats to production safety and the environment. Therefore, accurate prediction of groundwater levels in coal mining areas is of paramount importance for safe and green mining and the scientific management of water resources.

[0003] Currently, when making groundwater level predictions, the groundwater level data signals contain both long-term trend changes and short-term fluctuations. Affected by multi-scale disturbances, the accuracy of groundwater level predictions is limited. Summary of the Invention

[0004] In view of this, the purpose of this disclosure is to propose a method, device and electronic equipment for predicting groundwater levels in coal mining areas, so as to solve or partially solve the above problems.

[0005] To achieve the above objectives, the first aspect of this disclosure provides a method for predicting groundwater levels in coal mining areas, comprising: Obtain the first monitoring data corresponding to the target pumping well and the second monitoring data corresponding to the target monitoring well; The first monitoring data is decomposed by wavelet packet to obtain the first high-frequency signal and the first low-frequency signal corresponding to the first monitoring data. Wavelet packet decomposition is performed on the second monitoring data to obtain the second high-frequency signal and the second low-frequency signal corresponding to the second monitoring data; Based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal, the predicted water level data corresponding to the target monitoring well is determined using a pre-trained water level prediction model.

[0006] Based on the same inventive concept, a second aspect of this disclosure proposes a groundwater level prediction device for coal mining areas, comprising: The data acquisition module is configured to acquire the first monitoring data corresponding to the target pumping well and the second monitoring data corresponding to the target monitoring well. The first decomposition module is configured to perform wavelet packet decomposition on the first monitoring data to obtain the first high-frequency signal and the first low-frequency signal corresponding to the first monitoring data. The second decomposition module is configured to perform wavelet packet decomposition on the second monitoring data to obtain the second high-frequency signal and the second low-frequency signal corresponding to the second monitoring data. The water level prediction module is configured to determine the predicted water level data corresponding to the target monitoring well based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal, using a pre-trained water level prediction model.

[0007] Based on the same inventive concept, a third aspect of this disclosure proposes an electronic device including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method described above when executing the computer program.

[0008] Based on the same inventive concept, a fourth aspect of this disclosure provides a non-transitory computer-readable storage medium that stores computer instructions for causing a computer to perform the methods described above.

[0009] As can be seen from the above, this disclosure proposes a method, device, and electronic equipment for predicting groundwater levels in coal mining areas. It acquires first monitoring data corresponding to a target pumping well and second monitoring data corresponding to the target monitoring well. Wavelet packet decomposition is performed on the first monitoring data to obtain a first high-frequency signal and a first low-frequency signal corresponding to the first monitoring data. Wavelet packet decomposition is performed on the second monitoring data to obtain a second high-frequency signal and a second low-frequency signal corresponding to the second monitoring data. Based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal, a pre-trained water level prediction model is used to determine the predicted water level data corresponding to the target monitoring well. By performing wavelet packet decomposition on the first and second monitoring data to obtain corresponding high-frequency and low-frequency signals, water level prediction is performed based on the two high-frequency signals and two low-frequency signals. Wavelet packet decomposition effectively distinguishes signals of different frequencies. Furthermore, during water level prediction, the high-frequency signal after wavelet packet decomposition is not discarded; prediction is performed simultaneously based on both the low-frequency and high-frequency signals, resulting in more accurate predictions of the water level data for the target monitoring well. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in this disclosure or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1This is a flowchart of a method for predicting groundwater levels in coal mining areas, as described in this embodiment of the present disclosure. Figure 2 This is a schematic diagram of the structure of the water level prediction model according to an embodiment of the present disclosure; Figure 3 This is a schematic diagram illustrating how the optimal parameter combination of the first water level prediction model in this embodiment is determined. Figure 4 This is a flowchart of a groundwater level prediction method for coal mining areas according to another embodiment of the present disclosure; Figure 5 This is a schematic diagram of the structure of the groundwater level prediction device in a coal mining area according to an embodiment of the present disclosure; Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0013] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this disclosure should have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms "first," "second," and similar terms used in the embodiments of this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0014] The following are definitions of terms used in this disclosure: LSTM (Long Short-Term Memory) neural networks can be used to process time series data problems, filtering and integrating information to better capture long-range dependency features in the data. This overcomes the limitations of convolutional neural networks in handling long-term time series prediction problems. Building upon recurrent neural networks (RNNs), it adds the ability to preserve historical information of the sequence data, while also solving the problems of gradient explosion and gradient vanishing. Groundwater level fluctuations in coal mining areas are a crucial indicator for ensuring safe production and protecting the regional ecological environment. Particularly with the significant westward shift of coal mining, concentrated mining activities can lead to abnormal fluctuations in groundwater levels, resulting in problems such as mine water inrush, surface subsidence, and water shortages, posing serious threats to production safety and the environment. Therefore, accurate prediction of groundwater levels in coal mining areas is of paramount importance for safe and green mining and the scientific management of water resources.

[0015] Existing groundwater level prediction methods can be mainly divided into two categories. One category utilizes numerical simulation software such as COMSOL, GMS, and MODFLOW, which, based on certain physical backgrounds and geological parameters, offer high accuracy in predicting regional trends. However, these methods are computationally complex and sensitive to model parameters, making them unsuitable for scenarios with high real-time requirements. The other category utilizes statistical methods, such as linear correlation, BP neural networks, and support vector machines, which are effective for data-driven nonlinear modeling, offering fast computation speeds and good real-time processing capabilities.

[0016] However, existing technologies still have shortcomings in the following aspects regarding the prediction needs of groundwater levels in coal mining areas: They cannot fully exploit the correlations and differences between multi-source data; their capacity to remember long-term series data is limited, making it difficult to accurately capture key historical information; the influence of data noise leads to poor model accuracy; and because groundwater level data signals simultaneously contain long-term trend changes and short-term fluctuations, they are affected by multi-scale disturbances, resulting in limited accuracy in groundwater level prediction.

[0017] Furthermore, while wavelet packet decomposition is used to predict groundwater levels, reconstruction is typically performed after wavelet packet decomposition. This method discards high-frequency signals from the obtained data. Although the resulting low-frequency signals are smoother, the high-frequency signals are also real signals in the obtained data. If they are discarded directly, the predicted water level data will also contain errors.

[0018] Based on the above description, this embodiment proposes a method for predicting groundwater levels in coal mining areas, such as... Figure 1 As shown, the method specifically includes: Step 101: Obtain the first monitoring data corresponding to the target pumping well and the second monitoring data corresponding to the target monitoring well.

[0019] Step 102: Perform wavelet packet decomposition on the first monitoring data to obtain the first high-frequency signal and the first low-frequency signal corresponding to the first monitoring data.

[0020] Step 103: Perform wavelet packet decomposition on the second monitoring data to obtain the second high-frequency signal and the second low-frequency signal corresponding to the second monitoring data.

[0021] Step 104: Based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal, determine the predicted water level data corresponding to the target monitoring well using a pre-trained water level prediction model.

[0022] In specific implementation, the first monitoring data corresponding to the target pumping well is acquired. This first monitoring data represents the groundwater monitoring data corresponding to the target pumping well. The first monitoring data includes pumping volume data and first water level data. The pumping volume data of the target pumping well can be obtained through daily pumping volume records. The first water level data of the target pumping well can be obtained by collecting daily water level observations of the target pumping well in the coal mining area. This first water level data can directly reflect the water level change characteristics of the pumping well during operation, providing a basis for analyzing the impact of the pumping process on groundwater dynamics.

[0023] The second monitoring data corresponding to the target monitoring well is obtained, whereby the second monitoring data represents the groundwater monitoring data corresponding to the target monitoring well. In this embodiment, the target monitoring well is a monitoring well distributed within a preset distance range of the target pumping well, and the second monitoring data is the second water level data of the target monitoring well, which can be obtained by collecting the daily-scale water level monitoring data of the target monitoring well. The second water level data can reflect the degree of disturbance to the regional groundwater level field caused by pumping and the hydraulic connection between wells, which is of great significance for a comprehensive analysis of the groundwater system response.

[0024] After acquiring the first monitoring data and the second monitoring data, wavelet packet decomposition is performed on the first monitoring data to obtain the first high-frequency signal and the first low-frequency signal corresponding to the first monitoring data. Wavelet packet decomposition is then performed on the second monitoring data to obtain the second high-frequency signal and the second low-frequency signal corresponding to the second monitoring data.

[0025] In this embodiment, wavelet packet decomposition is a signal processing method that extracts signal features by decomposing the signal into sub-signals of different frequency ranges. During the decomposition process, wavelet functions and scaling functions are used to capture high-frequency and low-frequency information of the signal, respectively. High-frequency signals typically contain rapidly changing parts of the signal, such as noise and abrupt changes, while low-frequency signals contain the smooth trends and main components of the signal. Wavelet packet decomposition utilizes the localization properties of wavelet functions to finely divide the signal into different frequency bands, thereby simultaneously obtaining both high-frequency and low-frequency signals.

[0026] After determining the first high-frequency signal and the first low-frequency signal corresponding to the first monitoring data, and the second high-frequency signal and the second low-frequency signal corresponding to the second monitoring data, a pre-trained water level prediction model is obtained. Based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal, the predicted water level data corresponding to the target monitoring well is determined using the pre-trained water level prediction model.

[0027] In this embodiment, the water level prediction model is a neural network model, specifically an LSTM neural network structure. The structure of the water level prediction model is as follows: Figure 2 As shown, the water level prediction model includes an input layer, an LSTM layer, a Dropout layer, a fully connected layer, and an output layer.

[0028] The above scheme obtains first monitoring data and second monitoring data corresponding to the target pumping well. Wavelet packet decomposition is performed on the first monitoring data to obtain a first high-frequency signal and a first low-frequency signal. Wavelet packet decomposition is performed on the second monitoring data to obtain a second high-frequency signal and a second low-frequency signal. Based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal, a pre-trained water level prediction model is used to determine the predicted water level data corresponding to the target monitoring well. By performing wavelet packet decomposition on the first and second monitoring data to obtain corresponding high-frequency and low-frequency signals, water level prediction is performed based on the two high-frequency signals and two low-frequency signals. Wavelet packet decomposition effectively distinguishes signals of different frequencies. Furthermore, during water level prediction, the high-frequency signal after wavelet packet decomposition is not discarded; prediction is performed simultaneously based on both the low-frequency and high-frequency signals, resulting in more accurate predictions of the target monitoring well's water level data.

[0029] In some embodiments, the first monitoring data includes pumping volume data and first water level data. Step 102 involves performing wavelet packet decomposition on the first monitoring data to obtain a first high-frequency signal and a first low-frequency signal corresponding to the first monitoring data. Specifically, this includes: Step 1021: Perform wavelet packet decomposition on the pumping volume data to obtain the first high-frequency sub-signal and the first low-frequency sub-signal corresponding to the pumping volume data; Step 1022: Perform wavelet packet decomposition on the first water level data to obtain the second high-frequency sub-signal and the second low-frequency sub-signal corresponding to the first water level data; Step 1023: Use the first high-frequency sub-signal and the second high-frequency sub-signal as the first high-frequency signal, and use the first low-frequency sub-signal and the second low-frequency sub-signal as the first low-frequency signal.

[0030] In practice, the first monitoring data for the target pumping well includes pumping volume data and first water level data. Wavelet packet decomposition is performed on the pumping volume data and the first water level data respectively, that is, the high-frequency and low-frequency signals in the pumping volume data and the high-frequency and low-frequency signals in the first water level data are determined respectively.

[0031] Specifically, wavelet packet decomposition is performed on the pumping volume data to obtain a first high-frequency sub-signal and a first low-frequency sub-signal corresponding to the pumping volume data. Wavelet packet decomposition is performed on the first water level data to obtain a second high-frequency sub-signal and a second low-frequency sub-signal corresponding to the first water level data.

[0032] In this embodiment, the process of performing wavelet packet decomposition on the pumping volume data specifically includes: Step 10211: Decompose the pumping volume data based on a preset wavelet function to obtain a first low-frequency component and a first high-frequency component. Step 10212: Decompose the first low-frequency component based on a preset wavelet function to obtain the second low-frequency component and the second high-frequency component. Step 10213: Decompose the second low-frequency component based on a preset wavelet function to obtain the third low-frequency component and the third high-frequency component. Step 10214: Use the first high-frequency component, the second high-frequency component, and the third high-frequency component as the first high-frequency sub-signal corresponding to the pumping volume data, and use the third low-frequency component as the first low-frequency sub-signal corresponding to the pumping volume data.

[0033] In practice, when performing wavelet decomposition, a higher number of decomposition levels yields better detail signals but also increases errors. Therefore, the number of decomposition levels should not be too many or too few. In this embodiment, the number of decomposition levels is three, meaning the pumping volume data is decomposed into three levels.

[0034] A preset wavelet function is obtained, and the pumping volume data is decomposed based on the preset wavelet function to obtain a first low-frequency component and a second high-frequency component, wherein the first low-frequency component and the first high-frequency component are expressed by the following formula:

[0035] in, For pumping volume data, The first low-frequency component, This is the first high-frequency component.

[0036] The first low-frequency component is decomposed based on a preset wavelet function to obtain a second low-frequency component and a second high-frequency component, wherein the second low-frequency component and the second high-frequency component are expressed by the following formula:

[0037] in, The second low-frequency component, It is the second high-frequency component.

[0038] The second low-frequency component is decomposed based on a preset wavelet function to obtain a third low-frequency component and a third high-frequency component, wherein the third low-frequency component and the third high-frequency component are expressed by the following formula:

[0039] in, The third low-frequency component, It is the third high-frequency component.

[0040] Specifically, the preset wavelet function can be either the sym wavelet function or the db wavelet function. Since the db wavelet is an approximately orthogonal wavelet function of the symxi wavelet and has better regularity, the preset wavelet function in this embodiment is preferably the db4 wavelet function.

[0041] The first high-frequency component, the second high-frequency component, and the third high-frequency component are used as the first high-frequency sub-signal corresponding to the pumping volume data, and the third low-frequency component is used as the first low-frequency sub-signal corresponding to the pumping volume data.

[0042] In this embodiment, the process of performing wavelet packet decomposition on the first water level data specifically includes: Step 10221: Decompose the first water level data based on a preset wavelet function to obtain a fourth low-frequency component and a fourth high-frequency component. Step 10222: Decompose the fourth low-frequency component based on a preset wavelet function to obtain the fifth low-frequency component and the fifth high-frequency component. Step 10223: Decompose the fifth low-frequency component based on a preset wavelet function to obtain the sixth low-frequency component and the sixth high-frequency component. Step 10224: The fourth high-frequency component, the fifth high-frequency component, and the sixth high-frequency component are used as the second high-frequency sub-signal corresponding to the first water level data, and the sixth low-frequency component is used as the second low-frequency sub-signal corresponding to the first water level data.

[0043] In specific implementation, a preset wavelet function is obtained to perform a three-level decomposition on the first water level data. Specifically, the first water level data is decomposed based on the preset wavelet function to obtain a fourth low-frequency component and a fourth high-frequency component, wherein the fourth low-frequency component and the fourth high-frequency component are expressed by the following formula:

[0044] in, This is the first water level data. It is the fourth low-frequency component. It is the fourth high-frequency component.

[0045] The fourth low-frequency component is decomposed based on a preset wavelet function to obtain a fifth low-frequency component and a fifth high-frequency component, wherein the fifth low-frequency component and the fifth high-frequency component are expressed by the following formula:

[0046] in, The fifth low-frequency component, It is the fifth high-frequency component.

[0047] The fifth low-frequency component is decomposed based on a preset wavelet function to obtain a sixth low-frequency component and a sixth high-frequency component, wherein the sixth low-frequency component and the sixth high-frequency component are expressed by the following formula:

[0048] in, The sixth low-frequency component, It is the sixth high-frequency component.

[0049] Specifically, the preset wavelet function can be either the sym wavelet function or the db wavelet function. Since the db wavelet is an approximately orthogonal wavelet function of the symxi wavelet and has better regularity, the preset wavelet function in this embodiment is preferably the db4 wavelet function.

[0050] The fourth high-frequency component, the fifth high-frequency component, and the sixth high-frequency component are used as the second high-frequency sub-signal corresponding to the first water level data, and the sixth low-frequency component is used as the second low-frequency sub-signal corresponding to the first water level data.

[0051] The first high-frequency sub-signal and the second high-frequency sub-signal are used as the first high-frequency signal, and the first low-frequency sub-signal and the second low-frequency sub-signal are used as the first low-frequency signal. That is, the first high-frequency component, the second high-frequency component, the third high-frequency component, the fourth high-frequency component, the fifth high-frequency component, and the sixth high-frequency component are used as the first high-frequency signal. The third low-frequency component and the sixth low-frequency component are used as the first low-frequency signal.

[0052] In some embodiments, the second monitoring data includes second water level data. Step 103 involves performing wavelet packet decomposition on the second monitoring data to obtain a second high-frequency signal and a second low-frequency signal corresponding to the second monitoring data. Specifically, this includes: Step 1031: Decompose the second water level data based on a preset wavelet function to obtain the seventh low-frequency component and the seventh high-frequency component; Step 1032: Decompose the seventh low-frequency component based on a preset wavelet function to obtain the eighth low-frequency component and the eighth high-frequency component. Step 1033: Decompose the eighth low-frequency component based on a preset wavelet function to obtain the ninth low-frequency component and the ninth high-frequency component. Step 1034: The seventh high-frequency component, the eighth high-frequency component, and the ninth high-frequency component are used as the second high-frequency signal corresponding to the second water level data, and the ninth low-frequency component is used as the second low-frequency signal corresponding to the second water level data.

[0053] In specific implementation, the second monitoring data of the target monitoring well includes the second water level data. A preset wavelet function is obtained, and the second water level data is decomposed into three levels. Specifically, the second water level data is decomposed based on the preset wavelet function to obtain a seventh low-frequency component and a seventh high-frequency component, wherein the seventh low-frequency component and the seventh high-frequency component are expressed by the following formula:

[0054] in, This is the second water level data. It is the seventh low-frequency component. It is the seventh high-frequency component.

[0055] The seventh low-frequency component is decomposed based on a preset wavelet function to obtain an eighth low-frequency component and an eighth high-frequency component. The eighth low-frequency component and the eighth high-frequency component are expressed by the following formula:

[0056] in, This is the eighth low-frequency component. It is the eighth high-frequency component.

[0057] The eighth low-frequency component is decomposed based on a preset wavelet function to obtain a ninth low-frequency component and a ninth high-frequency component, wherein the ninth low-frequency component and the ninth high-frequency component are expressed by the following formula:

[0058] in, It is the ninth low-frequency component. It is the ninth high-frequency component.

[0059] Specifically, the preset wavelet function can be either the sym wavelet function or the db wavelet function. Since the db wavelet is an approximately orthogonal wavelet function of the symxi wavelet and has better regularity, the preset wavelet function in this embodiment is preferably the db4 wavelet function.

[0060] The seventh high-frequency component, the eighth high-frequency component, and the ninth high-frequency component are used as the second high-frequency signal corresponding to the second water level data, and the ninth low-frequency component is used as the second low-frequency signal corresponding to the second water level data.

[0061] In some embodiments, step 104, which involves determining the predicted water level data corresponding to the target monitoring well using a pre-trained water level prediction model based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal, specifically includes: Step 1041: Based on the first high-frequency signal and the second high-frequency signal, determine the predicted high-frequency signal based on the pre-trained first water level prediction model; Step 1042: Input the first low-frequency signal and the second low-frequency signal into the pre-trained second water level prediction model, process them through the second water level prediction model, and output the predicted low-frequency signal. Step 1043: The predicted high-frequency signal and the predicted low-frequency signal are merged to obtain the predicted water level data corresponding to the target monitoring well.

[0062] In specific implementation, a pre-trained first water level prediction model is obtained, and a predicted high-frequency signal is determined based on the first high-frequency signal and the second high-frequency signal according to the first water level prediction model.

[0063] A pre-trained second water level prediction model is obtained. The first low-frequency signal and the second low-frequency signal are input into the second water level prediction model. The second water level prediction model processes the signal and outputs the predicted low-frequency signal.

[0064] The predicted high-frequency signal and the predicted low-frequency signal are combined to obtain the predicted water level data corresponding to the target monitoring well. In other words, the predicted high-frequency signal and the predicted low-frequency signal are summed to obtain the signal representation of the predicted water level data corresponding to the target monitoring well.

[0065] In this embodiment, the first high-frequency signal includes a first high-frequency sub-signal and a second high-frequency sub-signal. The first high-frequency sub-signal includes a first high-frequency component, a second high-frequency component, and a third high-frequency component. The second high-frequency sub-signal includes a fourth high-frequency component, a fifth high-frequency component, and a sixth high-frequency component. The second high-frequency signal also includes a seventh high-frequency component, an eighth high-frequency component, and a ninth high-frequency component. Step 1041, based on the first high-frequency signal and the second high-frequency signal, determines the predicted high-frequency signal based on the pre-trained first water level prediction model. Specifically, this includes: Step 10411: Input the first high-frequency component, the fourth high-frequency component and the seventh high-frequency component into the pre-trained first water level prediction sub-model, process them through the first water level prediction sub-model, and output the first prediction high-frequency sub-signal. Step 10412: Input the second high-frequency component, the fifth high-frequency component and the eighth high-frequency component into the pre-trained second water level prediction sub-model, process them through the second water level prediction sub-model, and output the second prediction high-frequency sub-signal. Step 10413: Input the third high-frequency component, the sixth high-frequency component and the ninth high-frequency component into the pre-trained third water level prediction sub-model, process them through the third water level prediction sub-model, and output the third prediction high-frequency sub-signal. Step 10414: The first predicted high-frequency sub-signal, the second predicted high-frequency sub-signal, and the third predicted high-frequency sub-signal are used as the predicted high-frequency signal.

[0066] In specific implementation, based on the aforementioned description, wavelet packet decomposition is performed on the pumping volume data of the target pumping well to obtain the first high-frequency component, the second high-frequency component, the third high-frequency component, and the third low-frequency component; wavelet packet decomposition is performed on the first water level data of the target pumping well to obtain the fourth high-frequency component, the fifth high-frequency component, the sixth high-frequency component, and the sixth low-frequency component; and wavelet packet decomposition is performed on the second water level data of the target monitoring well to obtain the seventh high-frequency component, the eighth high-frequency component, the ninth high-frequency component, and the ninth low-frequency component.

[0067] The first water level prediction model specifically includes a first water level prediction sub-model, a second water level prediction sub-model, and a third water level prediction sub-model. The first high-frequency component, the fourth high-frequency component, and the seventh high-frequency component are input into the pre-trained first water level prediction sub-model, and processed by the first water level prediction sub-model to output the first predicted high-frequency sub-signal.

[0068] The second high-frequency component, the fifth high-frequency component, and the eighth high-frequency component are input into the pre-trained second water level prediction sub-model. After processing by the second water level prediction sub-model, the second prediction high-frequency sub-signal is output.

[0069] The third, sixth, and ninth high-frequency components are input into a pre-trained third water level prediction sub-model. The sub-model processes the data and outputs a third predicted high-frequency sub-signal. The first, second, and third predicted high-frequency sub-signals are used as the predicted high-frequency signal.

[0070] In this embodiment, the first water level prediction sub-model, the second water level prediction sub-model, and the third water level prediction sub-model can be trained separately or simultaneously. It is understood that training the first water level prediction model simultaneously is equivalent to training the third water level prediction sub-model.

[0071] Specifically, taking the training of the first water level prediction model as an example, the training process includes: Obtain the first dataset, which is divided into a training set, a validation set, and a test set. The training set is used to train and learn the first water level prediction model and adjust parameters such as the weight matrix. Then, the parameter combination is adjusted through the validation set to improve the prediction accuracy of the first water level prediction model. The output results obtained from the test set can be used to evaluate the generalization ability of the first water level prediction sub-model.

[0072] To investigate the impact of the input time window length on the simulation performance of the first water level prediction model, different input time windows of varying lengths were calculated and compared. Specifically, sequence data decomposed by wavelet analysis was used as the input to the first water level prediction model to predict the high-frequency components of the water level changes in the target monitoring well in the coal mine area over the next day. To ensure high accuracy during testing, the number of neurons was set to 100, and the initial learning rate was set to 0.001.

[0073] The first water level prediction model also uses an LSTM neural network structure. Key hyperparameters of LSTM include the number of LSTM layers, learning rate, number of hidden layer neurons, and sample length L. Optimal hyperparameters can improve the performance and predictive ability of the first water level prediction model. In this embodiment, the parameters or hyperparameters of the first water level prediction model are determined through cross-validation, specifically as follows: The optimal parameter combination for the first water level prediction model is selected by iterating through parameter combinations in the training set and cross-validating with the validation set using an increasing window. For example... Figure 3 As shown, Figure 3 To determine the optimal parameter combination for the first water level prediction model in this embodiment, the validation set Valid is divided into 6 folds, each containing n sample periods. RMSE1 is calculated using Train1 as the first training set and Valid1 as the first test set. The training set is then expanded by n1 periods, with Train2 as the second training set and Valid2 as the second test set, and RMSE2 is calculated. This process continues, with n periods added to the training set each time, until Valid6 becomes the test set, at which point RMSE6 is calculated. Finally, the average root mean square error (RMSE) is calculated using the model training results from the 6 validation sets, and the model parameters with better performance are selected. Cross-validation improves the robustness of the first water level prediction model, preventing significantly abnormal predictions and making the results more stable and reliable.

[0074] In some embodiments, the training process of the water level prediction model in step 104 specifically includes: Step A: Obtain the initial water level prediction model and historical dataset, wherein the historical dataset includes historical pumping volume data corresponding to the target pumping well, first historical water level data corresponding to the target pumping well, second historical water level data corresponding to the target monitoring well, and actual predicted water level corresponding to the target monitoring well. Step B: Train the initial water level prediction model using the historical dataset until the preset training termination condition is met, and obtain the water level prediction model.

[0075] In specific implementation, an initial water level prediction model and a historical dataset are obtained. The historical dataset includes historical pumping volume data corresponding to the target pumping well, first historical water level data corresponding to the target pumping well, second historical water level data corresponding to the target monitoring well, and the actual predicted water level corresponding to the target monitoring well.

[0076] The historical dataset is input into the initial water level prediction model, and the initial water level prediction model is trained using the historical dataset until the preset training termination condition is met, thus obtaining the water level prediction model.

[0077] In this embodiment, the preset training termination condition includes at least one of the following: determining that all data in the training dataset has been input into the initial water level prediction model for training, determining that the loss function of the initial water level prediction model has converged to a preset convergence threshold, or determining that the initial water level prediction model has been iterated for training to a preset number of iterations.

[0078] For example, the preset training termination condition is that all data in the training dataset has been input into the initial water level prediction model for training: The training dataset contains fifty sets of data. The preset training termination condition is that all data in the training dataset has been input into the initial water level prediction model for training. That is, when all fifty sets of data have been input into the initial water level prediction model, there is no training data in the training dataset that has not yet been input into the initial water level prediction model. At this point, the initial water level prediction model training is considered complete, and the water level prediction model is obtained.

[0079] Another example is that the preset training termination condition is to determine that the loss function of the initial water level prediction model converges to a preset convergence threshold: Training data from the training dataset is input into the initial water level prediction model for training, and the training results are output. A loss function is determined based on the training results and the actual predicted water level. The loss function may include at least one of the following: mean squared error loss function, cross-entropy loss function, logarithmic loss function, exponential loss function, squared loss function, or absolute value loss function, etc. When the loss function converges to a preset convergence threshold, the preset training termination condition is met, and the water level prediction model is obtained.

[0080] Another example is that the preset training termination condition is to determine the initial water level prediction model and iterate iteratively train it to a preset number of iterations.

[0081] The training data in the training dataset is input into the initial water level prediction model for iterative training. The number of iterations is recorded. When the number of iterations equals the preset number of iterations, the preset training termination condition is met, and the water level prediction model is obtained.

[0082] Based on the same inventive concept, another embodiment of this disclosure provides a method for predicting groundwater levels in coal mining areas, such as... Figure 4 As shown, the method specifically includes: Step 201: Obtain the pumping volume data and first water level data corresponding to the target pumping well, and obtain the second water level data corresponding to the target monitoring well.

[0083] Step 202: Perform wavelet packet decomposition on the pumping volume data, the first water level data, and the second water level data, respectively.

[0084] Step 203 yields the first low-frequency signal, the second low-frequency signal, the first high-frequency component, the second high-frequency component, the third high-frequency component, the fourth high-frequency component, the fifth high-frequency component, the sixth high-frequency component, the seventh high-frequency component, the eighth high-frequency component, and the ninth high-frequency component.

[0085] In specific implementation, the pumping volume data and first water level data corresponding to the target pumping well are acquired, and the second water level data corresponding to the target monitoring well is acquired. The pumping volume data is decomposed based on a preset wavelet function to obtain a first low-frequency component and a first high-frequency component. The first low-frequency component is then decomposed based on a preset wavelet function to obtain a second low-frequency component and a second high-frequency component. The second low-frequency component is then decomposed based on a preset wavelet function to obtain a third low-frequency component and a third high-frequency component. The first high-frequency component, the second high-frequency component, and the third high-frequency component are used as the first high-frequency sub-signal corresponding to the pumping volume data, and the third low-frequency component is used as the first low-frequency sub-signal corresponding to the pumping volume data.

[0086] The first water level data is decomposed using a preset wavelet function to obtain a fourth low-frequency component and a fourth high-frequency component. The fourth low-frequency component is then decomposed using the same preset wavelet function to obtain a fifth low-frequency component and a fifth high-frequency component. The fifth low-frequency component is then decomposed using the same preset wavelet function to obtain a sixth low-frequency component and a sixth high-frequency component. The fourth, fifth, and sixth high-frequency components are used as the second high-frequency sub-signal corresponding to the first water level data, and the sixth low-frequency component is used as the second low-frequency sub-signal corresponding to the first water level data.

[0087] The second water level data is decomposed using a preset wavelet function to obtain a seventh low-frequency component and a seventh high-frequency component. The seventh low-frequency component is then decomposed using the same preset wavelet function to obtain an eighth low-frequency component and an eighth high-frequency component. The eighth low-frequency component is then decomposed using the same preset wavelet function to obtain a ninth low-frequency component and a ninth high-frequency component. The seventh, eighth, and ninth high-frequency components are used as the second high-frequency signal corresponding to the second water level data, and the ninth low-frequency component is used as the second low-frequency signal corresponding to the second water level data.

[0088] Step 204: Input the first low-frequency signal and the second low-frequency signal into the pre-trained second water level prediction model, process them through the second water level prediction model, and output the predicted low-frequency signal.

[0089] Step 205: Input the first high-frequency component, the fourth high-frequency component, and the seventh high-frequency component into the pre-trained first water level prediction sub-model, process them through the first water level prediction sub-model, and output the first predicted high-frequency sub-signal.

[0090] Step 206: Input the second high-frequency component, the fifth high-frequency component and the eighth high-frequency component into the pre-trained second water level prediction sub-model, process them through the second water level prediction sub-model, and output the second prediction high-frequency sub-signal.

[0091] Step 207: Input the third high-frequency component, the sixth high-frequency component, and the ninth high-frequency component into the pre-trained third water level prediction sub-model, process them through the third water level prediction sub-model, and output the third prediction high-frequency sub-signal.

[0092] Step 208: The first predicted high-frequency sub-signal, the second predicted high-frequency sub-signal, the third predicted high-frequency sub-signal, and the predicted low-frequency signal are merged to obtain the predicted water level data corresponding to the target monitoring well.

[0093] In this embodiment, after obtaining the predicted water level data, an error assessment is performed on the predicted water level data, and commonly used evaluation indicators such as mean square error (MSE), mean absolute error (MAE), and coefficient of determination (R²) are output to quantitatively analyze the predictive performance of the water level prediction model. Simultaneously, the prediction errors under different operating conditions are grouped and compared to reveal the applicability and stability of the water level prediction model in different scenarios.

[0094] In this embodiment, the predicted water level data output by the water level prediction model is compared with the measured values, and plotted using methods such as time series curves and scatter plots to demonstrate the prediction accuracy and the degree of agreement with the trend. For studies with strong spatial distribution, water level contour maps or three-dimensional water level change maps can also be drawn in conjunction with the location of monitoring wells to intuitively reflect the dynamic characteristics of water levels in different areas.

[0095] In this embodiment, the predicted water level data is organized into tables or reports, including predicted water level values, measured water level values, and their deviations for each time period, facilitating subsequent review and decision-making. For long-term prediction results, water level change trend statistics can also be output, such as the average annual decline and key indicators like the delay time between pumping and water level response.

[0096] In this embodiment, the predicted water level data, evaluation indicators, and visualization maps are archived to form a complete technical report or database, which can be reused in mine management, academic research, and subsequent model optimization. Simultaneously, the results can be exported to various formats (such as Excel, image files, or GIS data formats) as needed to support use on different platforms and application scenarios.

[0097] The groundwater level prediction method for coal mining areas proposed in this embodiment can be specifically applied to mine water hazard prevention and control, providing water level early warnings for mine scheduling systems, and enabling advance drainage and protective measures deployment. It can also be applied to groundwater resource management, assisting in the formulation of water extraction plans and control strategies to ensure the sustainable use of regional water resources. For environmental monitoring and protection, it allows for continuous monitoring of groundwater dynamics in and around coal mining areas, reducing the risk of geological disasters. Finally, it can also be applied to scientific research and simulation analysis, providing high-precision predictive data support for hydrogeological research.

[0098] It should be noted that the method of this disclosure embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this disclosure embodiment, and the multiple devices will interact with each other to complete the method described.

[0099] It should be noted that the above description describes some embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] Based on the same inventive concept, another embodiment of this disclosure proposes a groundwater level prediction device for coal mining areas, such as... Figure 5 As shown, it specifically includes: The data acquisition module 501 is configured to acquire the first monitoring data corresponding to the target pumping well and the second monitoring data corresponding to the target monitoring well; The first decomposition module 502 is configured to perform wavelet packet decomposition on the first monitoring data to obtain the first high-frequency signal and the first low-frequency signal corresponding to the first monitoring data. The second decomposition module 503 is configured to perform wavelet packet decomposition on the second monitoring data to obtain the second high-frequency signal and the second low-frequency signal corresponding to the second monitoring data. The water level prediction module 504 is configured to determine the predicted water level data corresponding to the target monitoring well based on the first high-frequency signal, the first low-frequency signal, the second high-frequency signal, and the second low-frequency signal using a pre-trained water level prediction model.

[0101] In some embodiments, the first monitoring data includes pumping volume data and first water level data, and the first decomposition module 502 is specifically configured to: Wavelet packet decomposition is performed on the pumping volume data to obtain the first high-frequency sub-signal and the first low-frequency sub-signal corresponding to the pumping volume data; Wavelet packet decomposition is performed on the first water level data to obtain the second high-frequency sub-signal and the second low-frequency sub-signal corresponding to the first water level data. The first high-frequency sub-signal and the second high-frequency sub-signal are used as the first high-frequency signal, and the first low-frequency sub-signal and the second low-frequency sub-signal are used as the first low-frequency signal.

[0102] In some embodiments, the first decomposition module 502 is specifically configured to: The pumping volume data is decomposed based on a preset wavelet function to obtain a first low-frequency component and a first high-frequency component. The first low-frequency component is decomposed based on a preset wavelet function to obtain the second low-frequency component and the second high-frequency component. The second low-frequency component is decomposed based on a preset wavelet function to obtain the third low-frequency component and the third high-frequency component. The first high-frequency component, the second high-frequency component, and the third high-frequency component are used as the first high-frequency sub-signal corresponding to the pumping volume data, and the third low-frequency component is used as the first low-frequency sub-signal corresponding to the pumping volume data.

[0103] In some embodiments, the first decomposition module 502 is specifically configured to: The first water level data is decomposed based on a preset wavelet function to obtain a fourth low-frequency component and a fourth high-frequency component. The fourth low-frequency component is decomposed based on a preset wavelet function to obtain the fifth low-frequency component and the fifth high-frequency component. The fifth low-frequency component is decomposed based on a preset wavelet function to obtain the sixth low-frequency component and the sixth high-frequency component. The fourth high-frequency component, the fifth high-frequency component, and the sixth high-frequency component are used as the second high-frequency sub-signal corresponding to the first water level data, and the sixth low-frequency component is used as the second low-frequency sub-signal corresponding to the first water level data.

[0104] In some embodiments, the second monitoring data includes second water level data, and the second decomposition module 503 is specifically configured to: The second water level data is decomposed based on a preset wavelet function to obtain the seventh low-frequency component and the seventh high-frequency component. The seventh low-frequency component is decomposed based on a preset wavelet function to obtain the eighth low-frequency component and the eighth high-frequency component. The eighth low-frequency component is decomposed based on a preset wavelet function to obtain the ninth low-frequency component and the ninth high-frequency component. The seventh high-frequency component, the eighth high-frequency component, and the ninth high-frequency component are used as the second high-frequency signal corresponding to the second water level data, and the ninth low-frequency component is used as the second low-frequency signal corresponding to the second water level data.

[0105] In some embodiments, the water level prediction module 504 is specifically configured as follows: Based on the first high-frequency signal and the second high-frequency signal, and using a pre-trained first water level prediction model, a predicted high-frequency signal is determined. The first low-frequency signal and the second low-frequency signal are input into the pre-trained second water level prediction model, and the predicted low-frequency signal is output after processing by the second water level prediction model. The predicted high-frequency signal and the predicted low-frequency signal are combined to obtain the predicted water level data corresponding to the target monitoring well.

[0106] In some embodiments, the first high-frequency signal includes a first high-frequency sub-signal and a second high-frequency sub-signal. The first high-frequency sub-signal includes a first high-frequency component, a second high-frequency component, and a third high-frequency component. The second high-frequency sub-signal includes a fourth high-frequency component, a fifth high-frequency component, and a sixth high-frequency component. The second high-frequency signal also includes a seventh high-frequency component, an eighth high-frequency component, and a ninth high-frequency component. The water level prediction module 504 is specifically configured as follows: The first high-frequency component, the fourth high-frequency component, and the seventh high-frequency component are input into the pre-trained first water level prediction sub-model, and the first prediction high-frequency sub-signal is output after processing by the first water level prediction sub-model. The second high-frequency component, the fifth high-frequency component, and the eighth high-frequency component are input into the pre-trained second water level prediction sub-model, and after processing by the second water level prediction sub-model, the second prediction high-frequency sub-signal is output. The third high-frequency component, the sixth high-frequency component, and the ninth high-frequency component are input into the pre-trained third water level prediction sub-model, and processed by the third water level prediction sub-model to output the third prediction high-frequency sub-signal. The first predicted high-frequency sub-signal, the second predicted high-frequency sub-signal, and the third predicted high-frequency sub-signal are used as the predicted high-frequency signal.

[0107] In some embodiments, the apparatus further includes a training module, which is specifically configured to: Obtain an initial water level prediction model and a historical dataset, wherein the historical dataset includes historical pumping volume data corresponding to the target pumping well, first historical water level data corresponding to the target pumping well, second historical water level data corresponding to the target monitoring well, and actual predicted water level corresponding to the target monitoring well; The initial water level prediction model is trained using the historical dataset until the preset training termination condition is met, thus obtaining the water level prediction model.

[0108] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing this disclosure, the functions of each module can be implemented in one or more software and / or hardware.

[0109] The apparatus described above is used to implement the corresponding groundwater level prediction method in any of the following embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0110] Based on the same inventive concept, corresponding to 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, wherein the processor executes the program to implement the groundwater level prediction method for coal mining areas as described in any of the above embodiments.

[0111] Figure 6 This embodiment illustrates a more specific hardware structure of an electronic device. The device 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.

[0112] 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.

[0113] 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.

[0114] 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.

[0115] The communication interface 1040 is used to connect a 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, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0116] 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.

[0117] 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.

[0118] The electronic devices described above are used to implement the corresponding groundwater level prediction method for coal mining areas in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0119] 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 groundwater level prediction method for coal mining areas as described in any of the above embodiments.

[0120] 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 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.

[0121] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the groundwater level prediction method in the coal mining area as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0122] It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0123] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0124] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0125] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0126] 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.

[0127] 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., circuits) 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 can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0128] 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.

[0129] This disclosure is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method of predicting the water table in a coalfield, characterised in that, The method comprises the following steps: obtaining first monitoring data corresponding to a target pumping well and second monitoring data corresponding to a target monitoring well; wavelet packet decomposing the first monitoring data to obtain first high-frequency signals and first low-frequency signals corresponding to the first monitoring data; wavelet packet decomposing the second monitoring data to obtain second high-frequency signals and second low-frequency signals corresponding to the second monitoring data; using a water level prediction model trained in advance to determine predicted water level data corresponding to the target monitoring well according to the first high-frequency signals, the first low-frequency signals, the second high-frequency signals and the second low-frequency signals.

2. The method of claim 1, wherein, The first monitoring data comprises pumping volume data and first water level data, The wavelet packet decomposing the first monitoring data to obtain first high-frequency signals and first low-frequency signals corresponding to the first monitoring data comprises the following steps: wavelet packet decomposing the pumping volume data to obtain first high-frequency sub-signals and first low-frequency sub-signals corresponding to the pumping volume data; wavelet packet decomposing the first water level data to obtain second high-frequency sub-signals and second low-frequency sub-signals corresponding to the first water level data; taking the first high-frequency sub-signals and the second high-frequency sub-signals as the first high-frequency signals and taking the first low-frequency sub-signals and the second low-frequency sub-signals as the first low-frequency signals.

3. The method of claim 2, wherein, The wavelet packet decomposing the pumping volume data to obtain first high-frequency sub-signals and first low-frequency sub-signals corresponding to the pumping volume data comprises the following steps: decomposing the pumping volume data based on a preset wavelet function to obtain a first low-frequency component and a first high-frequency component; decomposing the first low-frequency component based on the preset wavelet function to obtain a second low-frequency component and a second high-frequency component; decomposing the second low-frequency component based on the preset wavelet function to obtain a third low-frequency component and a third high-frequency component; taking the first high-frequency component, the second high-frequency component and the third high-frequency component as the first high-frequency sub-signals corresponding to the pumping volume data and taking the third low-frequency component as the first low-frequency sub-signals corresponding to the pumping volume data.

4. The method of claim 3, wherein, The wavelet packet decomposing the first water level data to obtain second high-frequency sub-signals and second low-frequency sub-signals corresponding to the first water level data comprises the following steps: decomposing the first water level data based on a preset wavelet function to obtain a fourth low-frequency component and a fourth high-frequency component; decomposing the fourth low-frequency component based on the preset wavelet function to obtain a fifth low-frequency component and a fifth high-frequency component; decomposing the fifth low-frequency component based on the preset wavelet function to obtain a sixth low-frequency component and a sixth high-frequency component; taking the fourth high-frequency component, the fifth high-frequency component and the sixth high-frequency component as the second high-frequency sub-signals corresponding to the first water level data and taking the sixth low-frequency component as the second low-frequency sub-signals corresponding to the first water level data.

5. The method of claim 1, wherein, The second monitoring data comprises second water level data, The wavelet packet decomposing the second monitoring data to obtain second high-frequency signals and second low-frequency signals corresponding to the second monitoring data comprises the following steps: The second water level data are decomposed based on a preset wavelet function to obtain a seventh low-frequency component and a seventh high-frequency component; The seventh low-frequency component is decomposed based on a preset wavelet function to obtain an eighth low-frequency component and an eighth high-frequency component; The eighth low-frequency component is decomposed based on a preset wavelet function to obtain a ninth low-frequency component and a ninth high-frequency component; The seventh high-frequency component, the eighth high-frequency component and the ninth high-frequency component are taken as a second high-frequency signal corresponding to the second water level data, and the ninth low-frequency component is taken as a second low-frequency signal corresponding to the second water level data.

6. The method of claim 1, wherein, The water level prediction model is trained by using the first high-frequency signal, the first low-frequency signal, the second high-frequency signal and the second low-frequency signal, and the target monitoring well corresponding predicted water level data are determined, including: The first high-frequency signal and the second high-frequency signal are used to determine a predicted high-frequency signal based on a first water level prediction model trained in advance; The first low-frequency signal and the second low-frequency signal are input into a second water level prediction model trained in advance, and a predicted low-frequency signal is output through processing of the second water level prediction model; The predicted high-frequency signal and the predicted low-frequency signal are combined to obtain the target monitoring well corresponding predicted water level data.

7. The method of claim 6, wherein, The first high-frequency signal includes a first high-frequency sub-signal and a second high-frequency sub-signal, the first high-frequency sub-signal includes a first high-frequency component, a second high-frequency component and a third high-frequency component, the second high-frequency sub-signal includes a fourth high-frequency component, a fifth high-frequency component and a sixth high-frequency component, the second high-frequency signal includes a seventh high-frequency component, an eighth high-frequency component and a ninth high-frequency component, The first high-frequency signal and the second high-frequency signal are used to determine a predicted high-frequency signal based on a first water level prediction model trained in advance, including: The first high-frequency component, the fourth high-frequency component and the seventh high-frequency component are input into a first water level prediction sub-model trained in advance, and a first predicted high-frequency sub-signal is output through processing of the first water level prediction sub-model; The second high-frequency component, the fifth high-frequency component and the eighth high-frequency component are input into a second water level prediction sub-model trained in advance, and a second predicted high-frequency sub-signal is output through processing of the second water level prediction sub-model; The third high-frequency component, the sixth high-frequency component and the ninth high-frequency component are input into a third water level prediction sub-model trained in advance, and a third predicted high-frequency sub-signal is output through processing of the third water level prediction sub-model; The first predicted high-frequency sub-signal, the second predicted high-frequency sub-signal and the third predicted high-frequency sub-signal are taken as the predicted high-frequency signal.

8. The method of claim 1, wherein, The training process of the water level prediction model includes: An initial water level prediction model and a historical data set are obtained, wherein the historical data set includes historical pumping amount data corresponding to a target pumping well, first historical water level data corresponding to the target pumping well, second historical water level data corresponding to a target monitoring well and actual predicted water level data corresponding to the target monitoring well; The initial water level prediction model is trained by using the historical data set until a preset training end condition is met, so as to obtain a water level prediction model.

9. A groundwater level prediction device for coal mining areas, characterized in that, The method comprises the following steps: The data acquisition module is configured to acquire first monitoring data corresponding to a target pumping well and second monitoring data corresponding to a target monitoring well; The first decomposition module is configured to perform wavelet packet decomposition on the first monitoring data to obtain first high-frequency signals and first low-frequency signals corresponding to the first monitoring data; The second decomposition module is configured to perform wavelet packet decomposition on the second monitoring data to obtain second high-frequency signals and second low-frequency signals corresponding to the second monitoring data; The water level prediction module is configured to determine predicted water level data corresponding to the target monitoring well by using a water level prediction model that is obtained by pre-training, according to the first high-frequency signals, the first low-frequency signals, the second high-frequency signals, and the second low-frequency signals.

10. An electronic device, comprising: The computer program is stored in the memory and can be run on the processor, and the processor implements the method according to any one of claims 1 to 8 when executing the program.

Images