Intelligent identification and early warning method of mine water inrush sources based on fusion of multiple data sources

Abstract

The invention discloses an intelligent identification and early warning method for a mine water inrush source based on the fusion of multiple data sources, and belongs to the technical field of mine water inrush source identification. The method specifically comprises: real-time collection of mine water inrush source data, including water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound wave and vibration signal, and preprocessing the collected mine water inrush source data, data fusion of the preprocessed mine water inrush source data, obtaining a comprehensive index after the fusion of the mine water inrush source data, establishing a mine water inrush source identification model, training the mine water inrush source identification model based on historical data of the mine water inrush source, using the trained model, according to the comprehensive index after the fusion of the mine water inrush source data, identifying the type of the mine water inrush source, and according to the identification result of the mine water inrush source identification model, combining the historical data of the mine water inrush source and the fused comprehensive index, performing trend analysis and early warning level determination.

Description

Technical Field

[0001] The present invention belongs to the technical field of mine water inrush source identification, and in particular to an intelligent mine water inrush source identification and early warning method based on multi-data source fusion. Background Art

[0002] With the complexity of hydrogeological conditions and the increase in mining depth over the years, mine water inrush has become the second largest production problem after coal mine gas. The premise of solving the problem of mine water inrush is to identify the type of water source of water inrush and effectively prevent and control water in advance, so as to prevent and reduce the occurrence of water disasters. Therefore, it is particularly important to study the characteristics of previous coal mine flood accidents, find methods to detect cave water and old pond water in real time, and fundamentally prevent similar coal mine flood accidents in the future.

[0003] When using traditional hydrochemical methods to identify the type of mine water source, the identification process takes a long time, the analysis and identification efficiency is low, and it cannot meet the real-time requirements for identifying the type of mine water inrush source. Other conventional detection methods may damage the sample and cannot achieve non-destructive identification. They also require professional personnel to operate, which is time-consuming and labor-intensive.

[0004] For example, the patent with publication number CN114239648A discloses a method for identifying the source of water inrush in a mine based on the EEMD-Isomap algorithm, which includes the following steps: obtaining Raman spectrum information of mine water; performing noise reduction processing on the Raman spectrum signal; dividing the Raman spectrum signal into a training set and a prediction set; using the dimension reduction processing of the training set and the prediction set; using the training set as the input of the kernel K nearest neighbor algorithm to establish a water inrush source identification model; inputting the prediction set into the water inrush source identification model to obtain the prediction result; comparing the prediction result with the actual result in the model to obtain the recognition accuracy; using the recognition accuracy and recognition time as evaluation indicators of the water inrush source type to identify the type of water inrush in the mine. The method proposed in this invention can intelligently identify the type of water inrush source, which has great reference value for effectively preventing and controlling water in advance, thereby preventing and reducing the occurrence of water disasters.

[0005] For example, the patent with publication number CN115752385A discloses a method and system for identifying, sensing and warning the source of water inrush in a mine. By acquiring mine information in real time, and transmitting, storing and analyzing the acquired mine information in real time, it is possible to identify and predict the source of water inrush in the mine, and to warn of water hazards in the mine, and to issue warning information. The system integrates the collection of key hydrogeological and hydrochemical characteristic information of the mine, and innovatively combines Python deep learning technology with the Internet of Things cloud platform to realize real-time monitoring, detection and forecasting and warning technology for rapid identification of water hazards in mines, taking into account data security, rapid and real-time storage, and real-time feedback of mine information and analysis results, to achieve real-time monitoring, dynamic and continuous perception and warning. It solves the problem that the existing technology cannot monitor and warn the source of water inrush in the mine around the clock, resulting in untimely information updates and failure to meet the theoretical requirements of current mining.

[0006] The defects of the above patent are: it only identifies the chemical substances of the water source of mine water inrush, cannot accurately identify them, and cannot issue early warnings. Summary of the invention

[0007] In view of the shortcomings of the existing technology, the present invention proposes an intelligent identification and early warning method for mine water inrush sources based on the fusion of multiple data sources. By fusing multiple data sources, accurate identification and timely early warning of mine water inrush sources can be achieved, the false alarm rate can be reduced, and mine safety can be improved.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] The intelligent identification and early warning method of mine water inrush sources based on the fusion of multiple data sources includes:

[0010] Collect mine water inrush source data in real time, including water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound wave and vibration signal, and pre-process the collected mine water inrush source data;

[0011] The pre-processed mine water inrush source data is fused to obtain the comprehensive index after the fusion of the mine water inrush source data;

[0012] Establish a mine water inrush source identification model, train the mine water inrush source identification model based on the historical data of mine water inrush sources, and use the trained model to identify the type of mine water inrush source based on the comprehensive indicators after the mine water inrush source data is fused;

[0013] According to the identification results of the mine water inrush source identification model, combined with the historical data of mine water inrush sources and the integrated comprehensive indicators, trend analysis and warning level determination are carried out.

[0014] Specifically, the data fusion of the pre-processed mine water inrush source data includes:

[0015] Suppose the preprocessed mine water inrush source data set at time t is S t , S t ={s1 t ,s2 t ,...,s n t}, where s n t represents the pre-processed mine water inrush source data of the nth type at time t;

[0016] Calculate the weight of the i-th type of mine water inrush source data, the calculation formula is:

[0017] ,in, represents the weight of the i-th type of mine water inrush source data, n represents the number of mine water inrush source data types, H i represents the entropy value of the i-th type of mine water inrush source data, , where m represents the time window length, p ij Represents the probability distribution of the i-th type of data at the j-th moment;

[0018] The specific formula for predicting the water source data of the i-th type of mine water inrush at time t+1 is:

[0019] , where s i t+1’ represents the predicted data of the water inrush source of the i-th type of mine at time t+1, s i t represents the pre-processed mine water inrush source data of the i-th type at time t, A represents the state transfer matrix, B represents the control matrix, u t represents the control input at time t;

[0020] According to the actual data of the water inrush source of the i-th type of mine at time t+1, the predicted data of the water inrush source of the i-th type of mine at time t+1 is updated and iterated. The specific formula is:

[0021] ,in, represents the updated iterative data of the i-th type of mine water inrush source at time t+1, K t represents the Kalman gain, represents the actual data of the i-th type of mine water inrush source at time t+1, and H represents the observation matrix.

[0022] Specifically, the data fusion of the pre-processed mine water inrush source data also includes:

[0023] The mine water inrush source data at time t+1 are fused to obtain the comprehensive index after the fusion of the mine water inrush source data at time t+1. The specific formula is:

[0024] , where F(t+1) represents the comprehensive index after the fusion of the mine water inrush source data at time t+1.

[0025] Specifically, the establishment of a mine water inrush source identification model and training the mine water inrush source identification model based on historical data of mine water inrush sources include:

[0026] A mine water inrush source identification model is established, and the specific formula is:

[0027] , where Ts represents the mine water inrush source identification model, that is, the predicted water inrush source type, f(·) represents the activation function, which is used to introduce nonlinear relationships so that the model can learn complex relationships, W represents the weight of the mine water inrush source identification model, b represents the bias parameter of the mine water inrush source identification model, and x represents the input of the mine water inrush source identification model;

[0028] Assume that the historical data set of mine water inrush sources is Ls, Ls={(Zs1,lab1),(Zs2,lab2),...,(Zs m ,lab m )}, where Zs m represents the comprehensive index set after the historical data of the water inrush source of the mth mine is integrated, lab m The label representing the historical data of the mth mine water inrush source, that is, the type of mine water inrush source;

[0029] The historical data set Ls of mine water inrush sources is divided into a training set and a test set. The training set is input into the mine water inrush source identification model, and the loss function is selected to train the mine water inrush source identification model. The test set is input into the mine water inrush source identification model to identify the type of mine water inrush sources. The training is stopped when the evaluation value of the identification result converges and remains unchanged.

[0030] Specifically, identifying the type of water source of the mine water inrush includes:

[0031] The comprehensive index after the fusion of mine water inrush source data is input into the trained mine water inrush source identification model, and the trained mine water inrush source identification model is used to identify the comprehensive index after the fusion of mine water inrush source data to determine the type of mine water inrush source.

[0032] Specifically, the warning level determination includes:

[0033] According to the current data of the mine water inrush source and the type of the mine water inrush source, the stage of the mine water inrush in the mine water inrush source type is judged, and different warning levels are determined according to the stage of the mine water inrush.

[0034] The system for intelligent identification and early warning of water inrush sources in mines based on fusion of multiple data sources is used to implement the method for intelligent identification and early warning of water inrush sources in mines based on fusion of multiple data sources, including: a data acquisition module, a data fusion module, a water source type identification module and an early warning module;

[0035] The data acquisition module is used to collect mine water inrush source data in real time, including water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound wave and vibration signal, and pre-process the collected mine water inrush source data;

[0036] The data fusion module is used to fuse the pre-processed mine water inrush source data to obtain a comprehensive index after the fusion of the mine water inrush source data;

[0037] The water source type identification module is used to establish a mine water inrush source identification model, train the mine water inrush source identification model based on the mine water inrush source historical data, and use the trained model to identify the mine water inrush source type according to the comprehensive indicators after the mine water inrush source data is fused;

[0038] The early warning module is used to perform trend analysis and early warning level determination based on the identification results of the mine water inrush source identification model, combined with the historical data of the mine water inrush source and the integrated comprehensive indicators.

[0039] Specifically, the water source type identification module includes: a water source type identification model building unit and a water source type identification unit;

[0040] The water source type identification model establishment unit is used to establish a water source type identification model and train the mine water inrush source identification model using historical data of mine water inrush sources;

[0041] The water source type identification unit is used to use the trained model to identify the type of mine water inrush source according to the comprehensive indicators after the mine water inrush source data is fused.

[0042] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for intelligent identification and early warning of water sources in mines based on fusion of multiple data sources are implemented.

[0043] A storage medium stores computer instructions, which, when executed, execute the steps of a method for intelligent identification and early warning of water sources in mines based on fusion of multiple data sources.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. The present invention proposes an intelligent identification and early warning method for mine water inrush sources based on the fusion of multiple data sources. By integrating multiple data sources, such as water level, flow, temperature, chemical composition, etc., and using data mining and machine learning algorithms, it can more accurately identify the source of water inrush, thereby reducing false alarms and missed alarms.

[0046] 2. The present invention proposes an intelligent identification and early warning method for mine water inrush sources based on the fusion of multiple data sources. The method can realize real-time monitoring of the mine environment. When early signs of water inrush events are detected, early warning signals are issued in time to gain valuable response time for miners and managers, thereby reducing casualties and property losses.

[0047] 3. The present invention proposes an intelligent identification and early warning method for mine water inrush sources based on the fusion of multiple data sources. By utilizing the fusion technology of multiple data sources, different types of data can be comprehensively analyzed to obtain more comprehensive and in-depth information, thereby improving the understanding and prediction capabilities of water inrush events. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flow chart of the method for intelligent identification and early warning of water inrush sources in mines based on fusion of multiple data sources provided by the present invention;

[0049] Figure 2 This is an architecture diagram of the mine water inrush source intelligent identification and early warning system based on multi-data source fusion provided by the present invention;

[0050] Figure 3 A schematic diagram of an electronic device provided by the present invention;

[0051] Figure 4 This is a schematic diagram of the storage medium structure provided by the present invention. DETAILED DESCRIPTION

[0052] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, in the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside" and the like indicate the orientation or position relationship based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "No. 1", "No. 2" and "No. 3" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. The present invention is further explained below in conjunction with specific embodiments.

[0053] Example 1

[0054] See also Figure 1 An embodiment of the present invention provides a method for intelligently identifying and warning the source of water inrush in a mine based on fusion of multiple data sources, comprising the following specific steps:

[0055] Step S1: deploy multiple sensor nodes to collect various types of mine water inrush source data in real time, including but not limited to water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound waves and vibration signals, and use data cleaning algorithms to clean the raw data to remove noise and outliers;

[0056] The sensors include but are not limited to pressure sensors, flow sensors, temperature sensors, pH sensors, acoustic wave sensors and microseismic sensors. The microseismic sensors utilize the characteristics of water to collect tiny vibrations generated by the flow of water.

[0057] Step S2: performing data fusion on the pre-processed mine water inrush source data to obtain a comprehensive index after the fusion of the mine water inrush source data;

[0058] The benefits of this step: Fusion of multiple sensor data can provide more information and perspectives, and enable a more comprehensive understanding of the actual situation of the mine. For example, the data of water level sensors, flow rate sensors, chemical sensors, and temperature sensors can be combined to more accurately determine the occurrence and source of water inrush events; the data of a single sensor may be affected by noise and failures, resulting in inaccurate identification and early warning. By fusing data from multiple sensors, the impact of a single sensor failure or noise on the overall situation can be reduced, and robustness and noise resistance can be improved; the weights of each sensor can be dynamically adjusted through methods such as information entropy. The present invention can adjust the importance distribution of data according to actual conditions, and is more adaptable when facing different environments and conditions;

[0059] The specific steps of step S2 are:

[0060] Step S201: Set the preprocessed mine water inrush source data set at time t as S t , S t ={s1 t ,s2 t ,...,s n t}, where s n t represents the pre-processed mine water inrush source data of the nth type at time t, that is, the pre-processed mine water inrush source sample data collected by the nth sensor at time t;

[0061] Step S202: Calculate the weight of the i-th type of mine water inrush source data, and the calculation formula is:

[0062] ,in, represents the weight of the i-th type of mine water inrush source data, n represents the number of mine water inrush source data types, H i represents the entropy value of the i-th type of mine water inrush source data, , where m represents the time window length, p ij Represents the probability distribution of the i-th type of data at the j-th moment;

[0063] Step S203: predicting the water source data of the i-th type of mine water inrush at time t+1, the specific formula is:

[0064] , where s i t+1’ represents the predicted data of the water inrush source of the i-th type of mine at time t+1, s i t represents the pre-processed mine water inrush source data of the i-th type at time t, A represents the state transfer matrix, B represents the control matrix, u t represents the control input at time t;

[0065] Step S204: According to the actual data of the water inrush source of the i-th type of mine at time t+1, the predicted data of the water inrush source of the i-th type of mine at time t+1 is updated and iterated. The specific formula is:

[0066] ,in, represents the updated iterative data of the i-th type of mine water inrush source at time t+1, K t represents the Kalman gain, represents the actual data of the i-th type of mine water inrush source at time t+1, and H represents the observation matrix;

[0067] K t The specific formula is: , P t represents the error covariance prediction value at time k, H represents the observation matrix, R represents the observation noise covariance matrix, and the error covariance prediction value P at time k t The specific formula is: t =(IK t ×H)P t-1 , I represents the identity matrix, P t-1 Represents the error covariance prediction value at time k-1;

[0068] Step S205: Fusing the mine water inrush source data at time t+1 to obtain a comprehensive index after the fusion of the mine water inrush source data at time t+1. The specific formula is:

[0069] , where F(t+1) represents the comprehensive index after the fusion of the mine water inrush source data at time t+1.

[0070] If no water inrush occurs, the mine water inrush source data will all be 0. If water inrush occurs but has not yet reached the well, some of the mine water inrush source data will be 0, such as water pressure, flow, temperature, conductivity, pH value, chemical composition, etc., because they cannot be detected and need to rely on sound waves and vibration signals.

[0071] Step S3: Establish a mine water inrush source identification model, train the mine water inrush source identification model based on the mine water inrush source historical data, and use the trained model to identify the mine water inrush source type according to the comprehensive indicators after the mine water inrush source data is fused;

[0072] The specific steps of step S3 are:

[0073] Step S301: Establish a mine water inrush source identification model, the specific formula is:

[0074] , where Ts represents the mine water inrush source identification model, which is used to predict the mine water inrush source type, f(·) represents the activation function, which is used to introduce nonlinear relationships so that the model can learn complex relationships, W represents the weight of the mine water inrush source identification model, b represents the bias parameter of the mine water inrush source identification model, and x represents the input of the mine water inrush source identification model;

[0075] Step S302: Set the historical data set of mine water inrush sources as Ls, Ls={(Zs1,lab1),(Zs2,lab2),...,(Zs m ,lab m )}, where Zs m represents the comprehensive index set after the historical data of the water inrush source of the mth mine is integrated, lab m The label representing the historical data of the mth mine water inrush source, that is, the type of mine water inrush source;

[0076] Step S303: Divide the historical data set Ls of the mine water inrush source into a training set and a test set, input the training set into the mine water inrush source identification model, select a loss function, train the mine water inrush source identification model, input the test set into the mine water inrush source identification model, identify the mine water inrush source, and stop training when the evaluation value of the identification result converges and remains unchanged;

[0077] The specific formula of the loss function is:

[0078] , where L represents the loss function and y j represents the true label of the historical data of the mth mine water inrush source, represents the predicted label of the historical data of the water source of the mth mine water inrush, and log represents the logarithmic function;

[0079] When training the model, the model parameters are adjusted according to the evaluation value of the recognition result until the recognition effect is optimal, and the model training is completed;

[0080] Step S304: using the trained mine water inrush source identification model to identify the comprehensive indicators after the fusion of the mine water inrush source data, and determine the type of water inrush source.

[0081] Step S4: according to the identification result of the mine water inrush source identification model, combined with the historical data of the mine water inrush source and the integrated comprehensive indicators, trend analysis and warning level determination are performed;

[0082] The warning level determination in step S4 specifically includes: judging the stage of the mine water inrush in the mine water inrush source type according to the current data of the mine water inrush source and the type of the mine water inrush source, and determining different warning levels according to different stages.

[0083] Step S5: Send the warning information to the mine monitoring center and relevant personnel through the wireless transmission network to prompt the possible water inrush risk in real time. The warning information includes the type of water inrush source, warning level, warning time, etc.

[0084] The present invention can extract and predict mine water inrush accidents. When collecting data, it can judge whether a mine water inrush accident will occur by fusing mine sound wave and vibration signal data. If a mine water inrush accident is about to occur, the type of the mine water inrush source can be judged at the same time, and an early warning can be issued. When a mine water inrush accident occurs, the water pressure, flow rate, temperature, conductivity, pH value, chemical composition and other data of the mine water inrush source are collected, and after fusion analysis, the type of the mine water inrush source can be judged, and a warning message can be issued.

[0085] The differences from traditional methods are: 1) Traditional methods: usually rely on a single or a few data sources, such as single sensor data, manual observations, etc.; Multi-data source fusion method: integrates multiple data sources, including water level, flow, temperature, chemical composition, geological data, historical data, etc.; 2) Traditional methods: mainly rely on experience judgment and simple data processing techniques, such as threshold method and statistical analysis; Multi-data source fusion method: uses advanced data mining, machine learning and artificial intelligence algorithms to perform complex multi-source data fusion and intelligent analysis; 3) Traditional methods: low warning accuracy, prone to false alarms and missed alarms; Multi-data source fusion method: by integrating multiple data and intelligent algorithms, the accuracy and precision of warnings are improved.

[0086] Example 2

[0087] See also Figure 2 , another embodiment provided by the present invention: a mine water inrush source intelligent identification and early warning system based on multi-data source fusion, comprising: a data acquisition module, a data fusion module, a water source type identification module and an early warning module;

[0088] The data acquisition module is used to collect mine water inrush source data in real time, including water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound wave and vibration signal, and pre-process the collected mine water inrush source data;

[0089] The data fusion module is used to fuse the pre-processed mine water inrush source data to obtain a comprehensive index after the fusion of the mine water inrush source data;

[0090] The water source type identification module is used to establish a mine water inrush source identification model, train the mine water inrush source identification model based on the mine water inrush source historical data, and use the trained model to identify the mine water inrush source type according to the comprehensive indicators after the mine water inrush source data is fused;

[0091] The early warning module is used to perform trend analysis and early warning level determination based on the identification results of the mine water inrush source identification model, combined with the historical data of the mine water inrush source and the integrated comprehensive indicators.

[0092] The water source type identification module includes: a water source type identification model building unit and a water source type identification unit;

[0093] The water source type identification model establishment unit is used to establish a water source type identification model and train the mine water inrush source identification model using historical data of mine water inrush sources;

[0094] The water source type identification unit is used to use the trained model to identify the type of mine water inrush source according to the comprehensive indicators after the mine water inrush source data is fused.

[0095] Example 3

[0096] Figure 3 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 3 As shown, the present application also provides an electronic device 500. The electronic device 500 may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by one or more processors, a method for intelligent identification and early warning of water inrush sources in mines based on fusion of multiple data sources may be executed.

[0097] The method or system according to the embodiment of the present application can also be used by Figure 3 The electronic device architecture shown in FIG. Figure 3 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a ROM 503, a RAM 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store the intelligent identification and early warning method for water inrush in a mine based on fusion of multiple data sources provided in the present application. The intelligent identification and early warning method of mine water inrush sources based on the fusion of multiple data sources may, for example, include: real-time collection of mine water inrush source data, including water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound waves and vibration signals, and preprocessing the collected mine water inrush source data, fusing the preprocessed mine water inrush source data to obtain comprehensive indicators after the fusion of mine water inrush source data, establishing a mine water inrush source identification model, training the mine water inrush source identification model based on the historical data of mine water inrush sources, using the trained model, according to the comprehensive indicators after the fusion of mine water inrush source data, identifying the type of mine water inrush source, and performing trend analysis and early warning level determination according to the identification results of the mine water inrush source identification model, combined with the historical data of mine water inrush sources and the fused comprehensive indicators. Further, the electronic device 500 may also include a user interface 508. Of course, Figure 3 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 3 One or more components of an electronic device are shown.

[0098] Example 4

[0099] Figure 4 Schematic diagram of a storage medium structure provided by an embodiment of the present application. Figure 4 As shown, the present application discloses a storage medium 600. Computer-readable instructions are stored on the storage medium 600. When the computer-readable instructions are executed by the processor, the intelligent identification and early warning method of the source of water inrush in a mine based on the fusion of multiple data sources according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0100] It should be understood that the methods, apparatuses, and devices of the present application can be implemented in many ways. For example, the methods, apparatuses, and devices of the present application can be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is only for illustration, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application can also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers a recording medium storing a program for executing the method according to the present application.

[0101] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0102] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for intelligent identification and early warning of mine water inrush sources based on fusion of multiple data sources, characterized in that: include: Collect mine water inrush source data in real time, including water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound wave and vibration signal, and pre-process the collected mine water inrush source data; The pre-processed mine water inrush source data is fused to obtain the comprehensive index after the fusion of the mine water inrush source data; Establish a mine water inrush source identification model, train the mine water inrush source identification model based on the historical data of mine water inrush sources, and use the trained model to identify the type of mine water inrush source based on the comprehensive indicators after the mine water inrush source data is fused; According to the identification results of the mine water inrush source identification model, combined with the historical data of mine water inrush sources and the integrated comprehensive indicators, trend analysis and warning level determination are carried out; The data fusion of the pre-processed mine water inrush source data includes: Suppose the preprocessed mine water inrush source data set at time t is S t , S t ={s1 t ,s2 t ,...,s n t }, where s n t represents the pre-processed mine water inrush source data of the nth type at time t; Calculate the weight of the i-th type of mine water inrush source data, the calculation formula is: ,in, represents the weight of the i-th type of mine water inrush source data, n represents the number of mine water inrush source data types, H i represents the entropy value of the i-th type of mine water inrush source data, , where m represents the time window length, p ij Represents the probability distribution of the i-th type of data at the j-th moment; The specific formula for predicting the water source data of the i-th type of mine water inrush at time t+1 is: , where s i t+1’ represents the predicted data of the water inrush source of the i-th type of mine at time t+1, s i t represents the pre-processed mine water inrush source data of the i-th type at time t, A represents the state transfer matrix, B represents the control matrix, u t represents the control input at time t; According to the actual data of the water inrush source of the i-th type of mine at time t+1, the predicted data of the water inrush source of the i-th type of mine at time t+1 is updated and iterated. The specific formula is: ,in, represents the updated iterative data of the i-th type of mine water inrush source at time t+1, K t represents the Kalman gain, represents the actual data of the i-th type of mine water inrush source at time t+1, and H represents the observation matrix; The mine water inrush source data at time t+1 are fused to obtain the comprehensive index after the fusion of the mine water inrush source data at time t+1. The specific formula is: , where F(t+1) represents the comprehensive index after the fusion of the mine water inrush source data at time t+1.

2. The method for intelligent identification and early warning of mine water inrush sources based on multi-data source fusion as claimed in claim 1, characterized in that: The method of establishing a mine water inrush source identification model and training the mine water inrush source identification model based on historical data of mine water inrush sources includes: A mine water inrush source identification model is established, and the specific formula is: , where Ts represents the mine water inrush source identification model, which is used to predict the mine water inrush source type, f(·) represents the activation function, which is used to introduce nonlinear relationship into the mine water inrush source identification model, W represents the weight of the mine water inrush source identification model, b represents the bias parameter of the mine water inrush source identification model, and x represents the input of the mine water inrush source identification model; Assume that the historical data set of mine water inrush sources is Ls, Ls={(Zs1,lab1),(Zs2,lab2),...,(Zs m ,lab m )}, where Zs m represents the comprehensive index set after the historical data of the water inrush source of the mth mine is integrated, lab m The label representing the historical data of the mth mine water inrush source, that is, the type of mine water inrush source; The historical data set Ls of mine water inrush sources is divided into a training set and a test set. The training set is input into the mine water inrush source identification model, and the loss function is selected to train the mine water inrush source identification model. The test set is input into the mine water inrush source identification model to identify the type of mine water inrush sources. The training is stopped when the evaluation value of the identification result converges and remains unchanged.

3. The method for intelligent identification and early warning of mine water inrush sources based on multi-data source fusion as claimed in claim 2, characterized in that: The method of identifying the type of water source of the mine water inrush includes: The comprehensive index after the fusion of mine water inrush source data is input into the trained mine water inrush source identification model, and the trained mine water inrush source identification model is used to identify the comprehensive index after the fusion of mine water inrush source data to determine the type of mine water inrush source.

4. The method for intelligent identification and early warning of mine water inrush sources based on fusion of multiple data sources as claimed in claim 3, characterized in that: The warning level determination includes: According to the current data of the mine water inrush source and the type of the mine water inrush source, the stage of the mine water inrush in the mine water inrush source type is judged, and different warning levels are determined according to the stage of the mine water inrush.

5. A system for intelligent identification and early warning of water inrush sources in mines based on fusion of multiple data sources, used to implement the method for intelligent identification and early warning of water inrush sources in mines based on fusion of multiple data sources as described in any one of claims 1 to 4, characterized in that: include: Data acquisition module, data fusion module, water source type identification module and early warning module; The data acquisition module is used to collect mine water inrush source data in real time, including water pressure, flow, temperature, conductivity, pH value, chemical composition, mine sound wave and vibration signal, and pre-process the collected mine water inrush source data; The data fusion module is used to fuse the pre-processed mine water inrush source data to obtain a comprehensive index after the fusion of the mine water inrush source data; The water source type identification module is used to establish a mine water inrush source identification model, train the mine water inrush source identification model based on the mine water inrush source historical data, and use the trained model to identify the mine water inrush source type according to the comprehensive indicators after the mine water inrush source data is fused; The early warning module is used to perform trend analysis and early warning level determination based on the identification results of the mine water inrush source identification model, combined with the historical data of the mine water inrush source and the integrated comprehensive indicators; The water source type identification module includes: a water source type identification model building unit and a water source type identification unit; The water source type identification model establishment unit is used to establish a water source type identification model and train the mine water inrush source identification model using historical data on mine water inrush sources; The water source type identification unit is used to use the trained model to identify the type of mine water inrush source according to the comprehensive indicators after the mine water inrush source data is fused.

6. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for intelligent identification and early warning of mine water inrush sources based on fusion of multiple data sources described in any one of claims 1-4 are implemented.

7. A storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed, the steps of the method for intelligent identification and early warning of mine water inrush sources based on fusion of multiple data sources described in any one of claims 1-4 are executed.

Images