Goaf water disaster prevention and early warning method and system based on multi-source data fusion

Abstract

The invention provides a goaf water disaster prevention and early warning method and system based on multi-source data fusion, and the method comprises the steps: 1, data collection: collecting the multi-source data of a goaf, the multi-source data including the goaf water level, the aquifer water level, the waterproof sealing wall deformation, and the ground surface settlement; step 2, data standardization and fusion processing; the data fused in the second step are input into a water disaster analysis and early warning model, the water disaster analysis and early warning model adopts a support vector machine (SVM) model, and the water disaster analysis and early warning model outputs a water disaster risk probability value. And 4, performing real-time monitoring and early warning. According to the invention, a fusion mode based on correlation analysis is adopted, weighted average fusion is carried out on data with strong correlation, and feature level fusion is carried out on data with weak correlation, so that the geological environment and hydrological dynamics of the goaf are comprehensively and accurately reflected, and the accuracy of water disaster early warning is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of coal mines, relates to goafs, and specifically relates to a goaf water hazard prevention and control warning method and system based on multi-source data fusion. Background Technique

[0002] A goaf refers to an underground cavity area formed due to mineral resource exploitation. After long-term mining activities, these areas may experience unstable geological structures, which can lead to a series of geological disaster problems. Among them, water hazards are the most common and serious ones. Water hazards not only pose a threat to the safe production of goafs but also may have a serious impact on the surrounding environment and residents' lives. Therefore, how to effectively prevent and warn of water hazards in goafs has become an important topic in mine safety management.

[0003] Currently, traditional goaf water hazard warning methods mainly rely on the monitoring of a single data source, such as water hazard perception data or geological exploration data. These methods can reflect the water hazard risk in goafs to a certain extent, but due to the single data source and insufficient information, the accuracy and timeliness of warning are poor. In addition, traditional methods often use simple threshold judgment or empirical formulas for water hazard risk assessment, lacking in-depth analysis of complex data relationships and being difficult to cope with the changing geological environment of goafs. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a goaf water hazard prevention and control warning method and system based on multi-source data fusion to solve the technical problem that the accuracy of goaf water hazard prevention and control warning in the existing technology needs to be further improved.

[0005] To solve the above technical problems, the present invention is implemented by adopting the following technical solutions:

[0006] A goaf water hazard prevention and control warning method based on multi-source data fusion, the method includes the following steps:

[0007] Step 1, data collection:

[0008] Collect multi-source data of the goaf, and the multi-source data includes goaf water level, aquifer water level, deformation of water isolation and sealing wall, and ground settlement.

[0009] Step 2, data standardization and fusion processing:

[0010] Perform standardization processing on the multi-source data collected in Step 1, and use the Z-score standardization method to convert the multi-source data into standard data with a mean of 0 and a standard deviation of 1.

[0011] Adopt a fusion method based on correlation analysis, and set the correlation coefficient threshold as ρ threshold, when the correlation coefficient of two sets of standard data is greater than ρ threshold , weighted average fusion is adopted, otherwise, the fusion method after feature extraction is adopted, and finally the fused data is obtained.

[0012] Step 3, water disaster analysis and early warning:

[0013] Input the data fused in Step 2 into the water disaster analysis and early warning model. The water disaster analysis and early warning model adopts the support vector machine SVM model, and the water disaster analysis and early warning model outputs the water disaster risk probability value.

[0014] Step 4, real-time monitoring and early warning:

[0015] Real-time monitor the multi-source data of the goaf. When new multi-source data enters, immediately update Step 2 and Step 3; when the water disaster risk probability value output by the water disaster analysis and early warning model exceeds the set early warning threshold, send out a water disaster early warning message.

[0016] The present invention also has the following technical features:

[0017] In Step 1, the calculation method of the multi-source data includes the following steps:

[0018] Step 101, the water levels of the goaf and the aquifer are calculated by the time series analysis method, and the specific calculation formula is:

[0019]

[0020] In the formula:

[0021] W(t) represents the water level at time t;

[0022] W 0 represents the initial water level;

[0023] n represents the number of influencing factors;

[0024] a i represents the weight of the i-th influencing factor;

[0025] represents the influencing factor function of the i-th influencing factor at time t.

[0026] Step 102, the deformation of the water-resisting sealed wall is calculated by the finite element analysis method to judge the stability of the water-resisting sealed wall, and the specific calculation formula is:

[0027]

[0028] In the formula:

[0029] ∈ represents the strain of the water-resisting sealed wall;

[0030] ΔL represents the change in the length of the water-blocking and airtight wall;

[0031] L 0 represents the initial length of the water-blocking and airtight wall.

[0032] Step 103: The surface settlement is calculated using the settlement rate to judge the settlement trend. The specific calculation formula is:

[0033] S(t) = S 0 + υ·t

[0034] In the formula:

[0035] S(t) represents the surface settlement at time t;

[0036] S 0 represents the initial surface settlement;

[0037] v represents the surface settlement rate.

[0038] In step three, the support vector machine SVM model is:

[0039]

[0040] In the formula:

[0041] f(x) represents the prediction function;

[0042] α i represents the weight of the support vector;

[0043] K(x i , x) represents the Gaussian kernel function;

[0044] b represents the bias term;

[0045] i represents the serial number of the support vector, i = 1, 2,..., n;

[0046] n represents the number of support vectors;

[0047] x represents the input sample vector;

[0048] x i represents the i-th support vector.

[0049] This method further includes the following steps:

[0050] Step five: Early warning information push:

[0051] Receive the water disaster early warning information sent in step four, push the early warning information to the staff, and record the push time and reception status at the same time to ensure the accurate transmission of the early warning information.

[0052] Step six: Data storage:

[0053] Store the multi-source data collected in Step 1, the fused data obtained in Step 2, the water hazard risk probability values output in Step 3, and the water hazard warning information sent in Step 4.

[0054] The present invention also protects a goaf water hazard prevention and warning system based on multi-source data fusion, and this system includes the following modules:

[0055] A data acquisition module, used for Step 1 as described above.

[0056] A data standardization and fusion processing module, used for Step 2 as described above.

[0057] A water hazard analysis and warning module, used for Step 3 as described above.

[0058] A real-time monitoring and warning module, used for Step 4 as described above.

[0059] A warning information push module, used for Step 5 as described above.

[0060] Data storage, used for Step 6 as described above.

[0061] Compared with the prior art, the present invention has the following technical effects:

[0062] (Ⅰ) The present invention adopts a fusion method based on correlation analysis, performs weighted average fusion on data with strong correlation, and performs feature-level fusion on data with weak correlation, so as to comprehensively and accurately reflect the geological environment and hydrological dynamics of the goaf, and improve the accuracy of water hazard warning.

[0063] (Ⅱ) The system of the present invention effectively improves the accuracy, timeliness and reliability of goaf water hazard warning through the coordinated work of multi-source data fusion, intelligent analysis and warning, multi-channel information push, data storage and security guarantee, convenient system management and maintenance, visualization display and historical data analysis functions, and provides strong technical support for the safe production and environmental protection of the goaf.

[0064] (Ⅲ) The present invention can achieve multi-source data fusion and improve the warning accuracy: This system collects multi-source data through the data acquisition module and performs standardization and fusion processing through the data fusion processing module. Adopting a fusion method based on correlation analysis, performing weighted average fusion on data with strong correlation, and performing feature-level fusion on data with weak correlation, so as to comprehensively and accurately reflect the geological environment and hydrological dynamics of the goaf and improve the accuracy of water hazard warning.

[0065] (Ⅳ) The present invention can achieve multi-intelligent water disaster analysis and early warning: The water disaster analysis and early warning module of this system constructs a water disaster prediction model based on support vector machines, and uses the Gaussian kernel function to map the data into a high-dimensional space for classification. Through the training of historical data and model optimization, it realizes the intelligent assessment of water disaster risks, issues water disaster early warning information in a timely and accurate manner, and avoids the problems of inaccurate and untimely early warnings caused by single data sources and simple threshold judgments in traditional methods.

[0066] (Ⅴ) The present invention can achieve multi-channel early warning information push: The early warning information push module supports multiple information push methods, including email, mobile phone text messages, WeChat push, and dedicated early warning system interface prompts, ensuring that early warning information is conveyed to relevant personnel in a timely and accurate manner. At the same time, the push content is detailed, including the water disaster risk level, the affected range, and recommended countermeasures, facilitating relevant personnel to quickly take effective countermeasures.

[0067] (Ⅵ) The present invention can achieve data storage and security guarantee: The data storage module uses cloud storage technology to store the multi-source data collected, the fused data, and the water disaster analysis and early warning results on the cloud server, with data backup and recovery functions to ensure the security and reliability of the data. At the same time, it supports the storage of structured and unstructured data and provides data encryption functions to protect the privacy and security of the data.

[0068] (Ⅶ) The present invention can achieve convenient system management and maintenance: The system management module provides user identity authentication and authorization management functions. Different users have different operation permissions to ensure the safe operation of the system. Administrators can perform system configuration and data management, and ordinary users can view relevant early warning information and emergency response plans. The system also includes device calibration and maintenance functions, regularly calibrating the data collection devices and sensors to ensure the normal operation of the devices and the accuracy of the data.

[0069] (Ⅷ) The present invention can achieve visual display and historical data analysis: The visualization module displays the real-time data of the goaf and the water disaster analysis and early warning results in an intuitive chart form on the display terminal. Users can view the status of the goaf and the water disaster risk situation in real time. The historical data analysis module deeply analyzes the stored historical data, uses data mining algorithms to find the potential correlations between different data features and water disaster events, analyzes the characteristics and laws of various water disaster events, and provides more targeted strategies for water disaster prevention and control.

[0070] The following further elaborates on the specific content of the present invention in conjunction with embodiments. Detailed implementation manners

[0071] It should be noted that all algorithms, models, functions, modules, units, technologies, devices and methods in the present invention, unless otherwise specified, all adopt the algorithms, models, functions, modules, units, technologies, devices and methods known in the prior art.

[0072] With the development of sensor technology, data acquisition technology and big data analysis technology, multi-source data fusion technology has gradually been applied to the field of geological disaster warning. Multi-source data fusion refers to the comprehensive processing and analysis of data from different sensors or data sources to obtain more comprehensive and accurate information. In the prevention and control of water hazards in goaf areas, multi-source data fusion technology can integrate various information, comprehensively reflect the geological environment and hydrogeological dynamics of the goaf area, and improve the accuracy and reliability of water hazard warning.

[0073] Therefore, the present invention proposes a goaf water hazard prevention and control warning system and method based on multi-source data fusion. Through the coordinated work of data acquisition, data fusion processing, water hazard analysis and warning, warning information push, data storage, system management, visualization display, equipment calibration, real-time monitoring and update, and historical data analysis modules, it realizes the comprehensive monitoring and timely warning of goaf water hazards, and improves the scientificity and effectiveness of goaf water hazard prevention and control.

[0074] The following gives specific embodiments of the present invention. It should be noted that the present invention is not limited to the following specific embodiments, and all equivalent transformations made on the basis of the technical solutions of this application fall within the protection scope of the present invention.

[0075] Embodiment 1:

[0076] This embodiment provides a goaf water hazard prevention and control warning method based on multi-source data fusion. The method includes the following steps:

[0077] Step 1, data acquisition:

[0078] Collect multi-source data of the goaf area, and the multi-source data includes the water level of the goaf area, the water level of the aquifer, the deformation of the water isolation and sealing wall, and the surface settlement.

[0079] In step 1, the calculation method of the multi-source data includes the following steps:

[0080] Step 101, the water level of the goaf area and the water level of the aquifer are calculated by the time series analysis method, and the specific calculation formula is:

[0081]

[0082] In the formula:

[0083] W(t) represents the water level at time t;

[0084] W 0Indicates the initial water level;

[0085] n represents the number of influencing factors;

[0086] a i represents the weight of the i-th influencing factor;

[0087] Represents the influencing factor function of the i-th influencing factor at time t.

[0088] Step 102, the deformation of the watertight sealed wall is calculated using a finite element analysis method to determine the stability of the watertight sealed wall. The specific calculation formula is:

[0089]

[0090] Where:

[0091] ∈ represents the strain of the watertight wall;

[0092] ΔL represents the change in the length of the watertight wall;

[0093] L 0 Indicates the initial length of the watertight confined wall.

[0094] Step 103, the surface settlement is calculated by using the settlement rate to determine the settlement trend. The specific calculation formula is:

[0095] S(t)=S 0 +υ·t

[0096] Where:

[0097] S(t) represents the surface subsidence at time t;

[0098] S 0 represents the initial surface settlement;

[0099] v represents the surface subsidence rate.

[0100] Furthermore, in this embodiment, a preliminary quality check is performed on the collected data. The preliminary quality check includes checking the integrity, rationality and consistency of the data, and using an outlier detection algorithm to eliminate obviously abnormal data.

[0101] Furthermore, in this embodiment, multi-source data are presented in the three-dimensional space model of the goaf. The specific presentation method is: using WebGL to create and render the three-dimensional space model of the goaf, thereby establishing the spatial information of the goaf; connecting the multi-source data to the three-dimensional space model of the goaf, so that the multi-source data is displayed in the three-dimensional space model of the goaf.

[0102] Step 2: Data standardization and fusion processing:

[0103] Perform standardization processing on the multi-source data collected in Step 1, and use the Z-score standardization method to convert the multi-source data into standard data with a mean of 0 and a standard deviation of 1.

[0104] Specifically in this embodiment, the data G = {g 1 , g 2 ,..., g n} is standardized to where is the mean of G, and S G is the standard deviation of G.

[0105] Adopt a fusion method based on correlation analysis. Let the correlation coefficient threshold be ρ threshold . When the correlation coefficient of two groups of standard data is greater than ρ threshold , weighted average fusion is adopted; otherwise, the fusion method after feature extraction is adopted to finally obtain the fused data.

[0106] Step 3, water damage analysis and early warning:

[0107] Input the data fused in Step 2 into the water damage analysis and early warning model. The water damage analysis and early warning model adopts the support vector machine SVM model, and the water damage analysis and early warning model outputs the water damage risk probability value.

[0108] In Step 3, the support vector machine SVM model is:

[0109]

[0110] In the formula:

[0111] f(x) represents the prediction function;

[0112] α i represents the weight of the support vector;

[0113] K(x i , x) represents the Gaussian kernel function;

[0114] b represents the bias term;

[0115] i represents the serial number of the support vector, i = 1, 2,..., n, used to identify different support vectors;

[0116] n represents the number of support vectors, that is, the total number of support vectors participating in the calculation of the prediction function;

[0117] x represents the input sample vector, that is, the sample data to be predicted;

[0118] x i represents the i-th support vector, which is the sample vector in the training set that has an important influence on the classification hyperplane.

[0119] Specifically, in this embodiment, the water hazard analysis and early warning model maps the data fused in Step 2 to a high-dimensional space through a kernel function for classification. The kernel function used is the Gaussian kernel function K(x, y) = exp(-γ||x - y|| 2 ).

[0120] In the formula:

[0121] x represents a sample vector in the data fused in Step 2;

[0122] y represents another sample vector in the data fused in Step 2;

[0123] γ represents the kernel parameter.

[0124] Specifically, in this embodiment, based on the support vector machine SVM model, the support vector machine SVM model is trained with the collected historical water hazard data, and the random gradient descent algorithm is used to optimize the parameters of the support vector machine SVM model until the model converges to obtain the water hazard analysis and early warning model.

[0125] Specifically, in the training process, the cross-validation method is used to verify the support vector machine SVM model to prevent overfitting of the support vector machine SVM model and improve the generalization ability of the model.

[0126] Specifically, in this embodiment, the structure of the water hazard analysis and early warning model includes an input layer, multiple hidden layers, and an output layer; the input layer is used to receive the data fused in Step 2; the hidden layer uses the ReLU activation function for nonlinear transformation; the output layer outputs the water hazard risk probability value.

[0127] Step 4, real-time monitoring and early warning:

[0128] Real-time monitor the multi-source data of the goaf. When new multi-source data enters, immediately update Step 2 and Step 3; when the water hazard risk probability value output by the water hazard analysis and early warning model exceeds the set early warning threshold, issue a water hazard early warning message.

[0129] In this embodiment, the water hazard early warning information determines the early warning level according to the following formula:

[0130]

[0131] In the formula:

[0132] L represents the early warning level;

[0133] P represents the water hazard risk probability value;

[0134] T 1 >T 2 >T 3 >…>TN represent a series of gradually decreasing water hazard risk thresholds;

[0135] C 1 ,C 2 ,C 3 …,C N represent the coupling relationship conditions between corresponding different data.

[0136] More specifically in this embodiment, the classification method of the warning level is as follows: when the water hazard risk assessment value is less than the first threshold, it is determined as a first-level warning, indicating that the water hazard risk is extremely low; when the water hazard risk assessment value is between the first threshold and the second threshold, it is determined as a second-level warning, indicating that the water hazard risk is relatively low; when the water hazard risk assessment value is between the second threshold and the third threshold, it is determined as a third-level warning, indicating that the water hazard risk is relatively high; when the water hazard risk assessment value is greater than the third threshold, it is determined as a fourth-level warning, indicating that the water hazard risk is extremely high, where the first threshold is less than the second threshold is less than the third threshold, and the above thresholds are set and adjusted according to historical water hazard data.

[0137] Step Five, warning information push:

[0138] Receive the water hazard warning information sent in Step Four, push the warning information to the staff, and record the push time and reception status at the same time to ensure the accurate transmission of the warning information.

[0139] In this embodiment, the push methods include but are not limited to email, mobile phone text messages, and dedicated warning system interface prompts, and the push content includes the water hazard risk level and the affected area information.

[0140] Step Six, data storage:

[0141] Store the multi-source data collected in Step One, the fused data obtained in Step Two, the water hazard risk probability value output in Step Three, and the water hazard warning information sent in Step Four.

[0142] In this embodiment, cloud storage technology is adopted for storage, and the data is stored in a cloud server, which has data backup and recovery functions. The data storage format includes structured data and unstructured data, and data encryption function is provided at the same time.

[0143] Embodiment 2:

[0144] This embodiment provides a goaf water hazard prevention and control warning system based on multi-source data fusion, and the system includes the following modules:

[0145] A data collection module for implementing Step One in Embodiment 1.

[0146] A data standardization and fusion processing module for implementing Step Two in Embodiment 1.

[0147] The water damage analysis and early warning module is used to implement Step 3 in Embodiment 1.

[0148] Furthermore, the water damage analysis and early warning module further includes a self-learning unit. This self-learning unit automatically optimizes the parameters of the support vector machine model through continuous accumulation and analysis of historical data, improving the accuracy of water damage prediction.

[0149] The real-time monitoring and early warning module is used to implement Step 4 in Embodiment 1.

[0150] The early warning information push module is used to implement Step 5 in Embodiment 1.

[0151] Furthermore, the early warning information push module further includes an early warning information confirmation unit. This unit is used to receive the confirmation feedback of relevant personnel on the early warning information and store the feedback information in the data storage module for subsequent analysis and improvement of the early warning mechanism.

[0152] Data storage is used to implement Step 6 in Embodiment 1.

[0153] As a further solution of this embodiment, the system further includes the following modules:

[0154] The system management module is used to manage and maintain the goaf water damage prevention and control early warning system for multi-source data fusion, including user identity authentication and authorization management. Different users have different operation permissions. Among them, the administrator has the highest permissions for system configuration and data management, and ordinary users can only view relevant early warning information. It also includes real-time monitoring of the system operation status, and can automatically issue an alarm and record fault information when the system is abnormal.

[0155] The visualization module is used to display the relevant data of the goaf and the results of water damage analysis and early warning in an intuitive chart form, including drawing the trend chart of the goaf water level change and the distribution map of water damage risk areas. The visualization interface supports interactive operations, and users can view specific data information through click and zoom operations.

[0156] Furthermore, the visualization module further includes a 3D visualization unit. This unit displays the geological structure, water level change, and risk area information of the goaf in the form of a 3D model, and realizes the display of sensor layout, monitoring equipment, data collection, and risk points, providing an intuitive analysis view.

[0157] The equipment calibration module is used to regularly calibrate the data collection equipment and sensors to ensure the accuracy of the collected data. It adopts a combination of automatic calibration and manual calibration. Automatic calibration is carried out according to the preset calibration algorithm and time interval. The calibration parameters and results are recorded during the calibration process for subsequent analysis and traceability.

[0158] Further, the device calibration module further includes a remote calibration unit, which remotely triggers and executes the calibration operations of the data acquisition device and the sensor through the network, reducing the workload and cost of on-site maintenance.

[0159] The historical data analysis module is used to analyze the stored historical data, mine the laws and trends of goaf water disasters, adopt time series analysis methods such as the ARIMA model to model and analyze the water level data, predict the future water level change trend, and generate the key factors causing water disasters under different conditions through the analysis of the data related to historical water disaster events.

Claims

1. A water hazard prevention and early warning method for goaf areas based on multi-source data fusion, characterized in that: The method comprises the following steps: Step 1: Data collection: Collecting multi-source data of the goaf, wherein the multi-source data includes water level in the goaf, water level in the aquifer, deformation of the water-proof and airtight wall, and surface settlement; Step 2: Data standardization and fusion processing: The multi-source data collected in step 1 are standardized, and the Z-score standardization method is used to convert the multi-source data into standard data with a mean of 0 and a standard deviation of 1; Adopt the fusion method based on correlation analysis, and set the correlation coefficient threshold as ρ threshold , when the correlation coefficient of two sets of standard data is greater than ρ threshold When the weighted average fusion is used, otherwise the feature extraction and fusion method is used to finally obtain the fused data; Step 3: Water hazard analysis and early warning: The data fused in step 2 is input into a water hazard analysis and early warning model, wherein the water hazard analysis and early warning model adopts a support vector machine (SVM) model, and the water hazard analysis and early warning model outputs a water hazard risk probability value; Step 4: Real-time monitoring and early warning: Real-time monitoring of multi-source data in the goaf, when new multi-source data enters, immediately update steps 2 and 3; when the water hazard risk probability value output by the water hazard analysis and early warning model exceeds the set early warning threshold, a water hazard early warning message is issued.

2. The method for early warning of water hazard prevention in goaf areas based on multi-source data fusion according to claim 1, characterized in that: In step 1, the method for calculating multi-source data includes the following steps: Step 101, the water level of the goaf and the water level of the aquifer are calculated by time series analysis method, and the specific calculation formula is: Where: W(t) represents the water level at time t; W0 represents the initial water level; n represents the number of influencing factors; a i represents the weight of the i-th influencing factor; f i (t) represents the influencing factor function of the i-th influencing factor at time t.

3. The method for early warning of water hazard prevention in goaf areas based on multi-source data fusion according to claim 1, characterized in that: In step 1, the method for calculating multi-source data includes the following steps: Step 102, the deformation of the watertight sealed wall is calculated using a finite element analysis method to determine the stability of the watertight sealed wall. The specific calculation formula is: Where: ∈ represents the strain of the watertight wall; ΔL represents the change in the length of the watertight wall; L0 represents the initial length of the watertight wall.

4. The method for early warning of water hazard prevention in goaf areas based on multi-source data fusion according to claim 1, characterized in that: In step 1, the method for calculating multi-source data includes the following steps: Step 103, the surface settlement is calculated by using the settlement rate to determine the settlement trend. The specific calculation formula is: S(t)=S0+v·tWhere: S(t) represents the surface subsidence at time t; S0 represents the initial surface settlement; v represents the surface subsidence rate.

5. The method for early warning of water hazard prevention in goaf area based on multi-source data fusion according to claim 1, characterized in that: In step 3, the support vector machine SVM model is: Where: f(x) represents the prediction function; α i represents the weight of the support vector; K(x i , x) represents the Gaussian kernel function; b represents the bias term; i represents the serial number of the support vector, i = 1, 2, ..., n; n represents the number of support vectors; x represents the input sample vector; x i represents the i-th support vector.

6. The method for early warning of water hazard prevention in goaf area based on multi-source data fusion according to claim 1, characterized in that: The method further comprises the following steps: Step 5: Push warning information: Receive the flood warning information issued in step 4, push the warning information to the staff, and record the push time and receiving status to ensure the accurate transmission of the warning information; Step 6: Data storage: The multi-source data collected in step one, the fused data obtained in step two, the flood risk probability value output in step three, and the flood warning information issued in step four are stored.

7. A water hazard prevention and early warning system for goaf areas based on multi-source data fusion, characterized in that: The system includes the following modules: A data acquisition module, used to implement step 1 of the method for early warning of water hazard prevention in goaf areas based on multi-source data fusion as claimed in claim 6; A data standardization and fusion processing module, used to implement step 2 of the goaf area water hazard prevention and early warning method based on multi-source data fusion as claimed in claim 6; A water hazard analysis and early warning module, used to implement step 3 of the method for early warning of water hazard prevention in goaf areas based on multi-source data fusion as claimed in claim 6; A real-time monitoring and early warning module, used to implement step 4 of the method for early warning of water hazard prevention in goaf areas based on multi-source data fusion as claimed in claim 6; An early warning information push module, used to implement step 5 of the method for early warning of water hazard prevention and control in goaf areas based on multi-source data fusion as claimed in claim 6; Data storage is used to implement step 6 of the goaf area water hazard prevention and early warning method based on multi-source data fusion as described in claim 6.

Images